draft-irtf-t2trg-iot-edge.xml

<?xml version='1.0' encoding='utf-8'?>
<!DOCTYPE rfc [
  <!ENTITY nbsp    "&#160;">
  <!ENTITY zwsp   "&#8203;">
  <!ENTITY nbhy   "&#8209;">
  <!ENTITY wj     "&#8288;">
]>
<?xml-stylesheet type="text/xsl" href="rfc2629.xslt" ?>
<!-- generated by https://github.com/cabo/kramdown-rfc version 1.6.39 (Ruby 3.0.2) -->
<?rfc iprnotified="yes"?>
<?rfc strict="yes"?>
<?rfc compact="yes"?>
<?rfc colonspace="yes"?>
<?rfc rfcedstyle="no"?>
<rfc xmlns:xi="http://www.w3.org/2001/XInclude" ipr="trust200902" docName="draft-irtf-t2trg-iot-edge-10" category="info" submissionType="IRTF" tocDepth="4" tocInclude="true" sortRefs="true" symRefs="true" version="3">
  <!-- xml2rfc v2v3 conversion 3.18.0 -->
  <front>
    <title abbrev="IoT Edge Computing">IoT Edge Challenges and Functions</title>
    <seriesInfo name="Internet-Draft" value="draft-irtf-t2trg-iot-edge-10"/>
    <author initials="J." surname="Hong" fullname="Jungha Hong">
      <organization>ETRI</organization>
      <address>
        <postal>
          <street>218 Gajeong-ro, Yuseung-Gu</street>
          <city>Daejeon</city>
          <code>34129</code>
          <country>Republic of Korea</country>
        </postal>
        <email>jhong@etri.re.kr</email>
      </address>
    </author>
    <author initials="Y.-G." surname="Hong" fullname="Yong-Geun Hong">
      <organization>Daejeon University</organization>
      <address>
        <postal>
          <street>62 Daehak-ro, Dong-gu</street>
          <city>Daejeon</city>
          <code>300716</code>
          <country>Republic of Korea</country>
        </postal>
        <email>yonggeun.hong@gmail.com</email>
      </address>
    </author>
    <author initials="X." surname="de Foy" fullname="Xavier de Foy">
      <organization>InterDigital Communications, LLC</organization>
      <address>
        <postal>
          <street>1000 Sherbrooke West</street>
          <city>Montreal</city>
          <code>H3A 3G4</code>
          <country>Canada</country>
        </postal>
        <email>xavier.defoy@interdigital.com</email>
      </address>
    </author>
    <author initials="M." surname="Kovatsch" fullname="Matthias Kovatsch">
      <organization>Huawei Technologies Duesseldorf GmbH</organization>
      <address>
        <postal>
          <street>Riesstr. 25 C // 3.OG</street>
          <city>Munich</city>
          <code>80992</code>
          <country>Germany</country>
        </postal>
        <email>ietf@kovatsch.net</email>
      </address>
    </author>
    <author initials="E." surname="Schooler" fullname="Eve Schooler">
      <organization>Intel</organization>
      <address>
        <postal>
          <street>2200 Mission College Blvd.</street>
          <city>Santa Clara, CA</city>
          <code>95054-1537</code>
          <country>USA</country>
        </postal>
        <email>eve.schooler@gmail.com</email>
      </address>
    </author>
    <author initials="D." surname="Kutscher" fullname="Dirk Kutscher">
      <organization>Hong Kong University of Science and Technology (Guangzhou)</organization>
      <address>
        <postal>
          <street>No.1 Du Xue Rd</street>
          <city>Guangzhou</city>
          <country>China</country>
        </postal>
        <email>ietf@dkutscher.net</email>
      </address>
    </author>
    <date year="2023" month="September" day="15"/>
    <area>T2TRG</area>
    <abstract>
      <?line 879?>

<t>Many Internet of Things (IoT) applications have requirements that cannot be satisfied by traditional cloud-based systems (i.e., cloud computing). These include time sensitivity, data volume, connectivity cost, operation in the face of intermittent services, privacy, and security. As a result, IoT is driving the Internet toward edge computing. This document outlines the requirements of the emerging IoT Edge and its challenges. It presents a general model and major components of the IoT Edge to provide a common basis for future discussions in the T2TRG and other IRTF and IETF groups. This document is a product of the IRTF Thing-to-Thing Research Group (T2TRG).</t>
    </abstract>
  </front>
  <middle>
    <?line 883?>

<section anchor="introduction">
      <name>Introduction</name>
      <t>Currently, many IoT services leverage cloud computing platforms, because they provide virtually unlimited storage and processing power. The reliance of IoT on back-end cloud computing provides additional advantages such as scalability and efficiency.  Today's IoT systems are fairly static with respect to integrating and supporting computation.  It is not that there is no computation, but that systems are often limited to static configurations (edge gateways and cloud services).</t>
      <t>However, IoT devices generate large amounts of data at the edges of the network. To meet IoT use case requirements, data is increasingly being stored, processed, analyzed, and acted upon close to the data sources. These requirements include time sensitivity, data volume, connectivity cost, and resiliency in the presence of intermittent connectivity, privacy, and security, which cannot be addressed by centralized cloud computing. A more flexible approach is necessary to address these needs effectively. This involves distributing computing (and storage) and seamlessly integrating it into the edge-cloud continuum. We refer to this integration of edge computing and IoT as "IoT edge computing". This draft describes the related background, use cases, challenges, system models, and functional components.</t>
      <t>Owing to the dynamic nature of the IoT edge computing landscape, this document does not list existing projects in this field. <xref target="sec-overview"/> presents a high-level overview of the field, based on a limited review of standards, research, open-source and proprietary products in <xref target="I-D.defoy-t2trg-iot-edge-computing-background"/>.</t>
      <t>This document represents the consensus of the Thing-to-Thing Research Group (T2TRG). It has been reviewed extensively by the Research Group (RG) members who are actively involved in the research and development of the technology covered by this document. It is not an IETF product and is not a standard.</t>
    </section>
    <section anchor="background">
      <name>Background</name>
      <section anchor="internet-of-things-iot">
        <name>Internet of Things (IoT)</name>
        <t>Since the term "Internet of Things" (IoT) was coined by Kevin Ashton in 1999 working on Radio-Frequency Identification (RFID) technology <xref target="Ashton"/>, the concept of IoT has evolved. It now reflects a vision of connecting the physical world to the virtual world of computers using (often wireless) networks over which things can send and receive information without human intervention.  Recently, the term has become more literal by connecting things to the Internet and converging on Internet and Web technologies.</t>
        <t>A Thing is a physical item made available in the IoT, thereby enabling digital interaction with the physical world for humans, services, and/or other Things (<xref target="I-D.irtf-t2trg-rest-iot"/>). In this document we will use the term "IoT device" to designate the embedded system attached to the Thing.</t>
        <t>Resource-constrained Things such as sensors, home appliances and wearable devices often have limited storage and processing power, which can provide challenges with respect to reliability, performance, energy consumption, security, and privacy <xref target="Lin"/>. Some, less resource-constrained Things, can generate a voluminous amount of data. This range of factors led IoT designs that integrate Things into larger distributed systems, for example edge or cloud computing systems.</t>
      </section>
      <section anchor="cloud-computing">
        <name>Cloud Computing</name>
        <t>Cloud computing has been defined in <xref target="NIST"/>: "cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction".  The low cost and massive availability of storage and processing power enabled the realization of another computing model, in which virtualized resources can be leased in an on-demand fashion and be provided as general utilities. Platform-as-a-Service and cloud computing platforms widely adopted this paradigm for delivering services over the Internet, gaining both economical and technical benefits <xref target="Botta"/>.</t>
        <t>Today, an unprecedented volume and variety of data is generated by Things, and applications deployed at the network edge consume this data.  In this context, cloud-based service models are not suitable for some classes of applications which require very short response times, access to local personal data, or generate vast amounts of data.  These applications may instead leverage edge computing.</t>
      </section>
      <section anchor="edge-computing">
        <name>Edge Computing</name>
        <t>Edge computing, also referred to as fog computing in some settings, is a new paradigm in which substantial computing and storage resources are placed at the edge of the Internet, close to mobile devices, sensors, actuators, or machines.  Edge computing happens near data sources <xref target="Mahadev"/>, as well as close to where decisions are made or where interactions with the physical world take place ("close" here can refer to a distance which is topological, physical, latency-based, etc.).  It processes both downstream data (originating from cloud services) and upstream data (originating from end devices or network elements).  The term "fog computing" usually represents the notion of multi-tiered edge computing, that is, several layers of compute infrastructure between end devices and cloud services.</t>
        <t>An edge device is any computing or networking resource residing between end-device data sources and cloud-based data centers.  In edge computing, end devices consume and produce data. At the network edge, devices not only request services and information from the Cloud but also handle computing tasks including processing, storage, caching, and load balancing on data sent to and from the Cloud <xref target="Shi"/>.  This does not preclude end devices from hosting computation themselves, when possible, independently or as part of a distributed edge computing platform.</t>
        <t>Several standards developing organization (SDO) and industry forums have provided definitions of edge and fog computing:</t>
        <ul spacing="normal">
          <li>ISO defines edge computing as a "form of distributed computing in which significant processing and data storage takes place on nodes which are at the edge of the network" <xref target="ISO_TR"/>.</li>
          <li>ETSI defines multi-access edge computing as a "system which provides an IT service environment and cloud-computing capabilities at the edge of an access network which contains one or more type of access technology, and in close proximity to its users" <xref target="ETSI_MEC_01"/>.</li>
          <li>The Industry IoT Consortium (IIC, now incorporating what was formerly OpenFog) defines fog computing as "a horizontal, system-level architecture that distributes computing, storage, control and networking functions closer to the users along a cloud-to-thing continuum" <xref target="OpenFog"/>.</li>
        </ul>
        <t>Based on these definitions, we can summarize a general philosophy of edge computing as distributing the required functions close to users and data, while the difference to classic local systems is the usage of management and orchestration features adopted from cloud computing.</t>
        <t>Actors from various industries approach edge computing using different terms and reference models although, in practice, these approaches are not incompatible and may integrate with each other:</t>
        <ul spacing="normal">
          <li>The telecommunication industry tends to use a model where edge computing services are deployed over Network Function Virtualization (NFV) infrastructure, at aggregation points or in proximity to the user equipment (e.g., gNodeBs) <xref target="ETSI_MEC_03"/>.</li>
          <li>Enterprise and campus solutions often interpret edge computing as an "edge cloud", that is, a smaller data center directly connected to the local network (often referred to as "on-premise").</li>
          <li>The automation industry defines the edge as the connection point between IT and OT (Operational Technology). Hence, edge computing sometimes refers to applying IT solutions to OT problems, such as analytics, more flexible user interfaces, or simply having more computing power than an automation controller.</li>
        </ul>
      </section>
      <section anchor="sec-uc">
        <name>Examples of IoT Edge Computing Use Cases</name>
        <t>IoT edge computing can be used in home, industry, grid, healthcare, city, transportation, agriculture, and/or educational scenarios. Here, we discuss only a few examples of such use cases, to identify differentiating requirements, providing references to other use cases.</t>
        <t><strong>Smart Factory</strong></t>
        <t>As part of the 4th industrial revolution, smart factories run real-time processes based on IT technologies, such as artificial intelligence and big data. Even a very small environmental change in a smart factory can lead to a situation in which production efficiency decreases or product quality problems occur. Therefore, simple but time-sensitive processing can be performed at the edge, for example, controlling the temperature and humidity in the factory, or operating machines based on the real-time collection of the operational status of each machine. However, data requiring highly precise analysis, such as machine lifecycle management or accident risk prediction, can be transferred to a central data center for processing.</t>
        <t>The use of edge computing in a smart factory can reduce the cost of network and storage resources by reducing the communication load to the central data center or server. It is also possible to improve process efficiency and facility asset productivity through real-time prediction of failures and to reduce the cost of failure through preliminary measures. In the existing manufacturing field, production facilities are manually run according to a program entered in advance; however, edge computing in a smart factory enables tailoring solutions by analyzing data at each production facility and machine level. Digital twins <xref target="Jones"/> of IoT devices have been jointly used with edge computing in industrial IoT scenarios <xref target="Chen"/>.</t>
        <t><strong>Smart Grid</strong></t>
        <t>In future smart city scenarios, the Smart Grid will be critical in ensuring highly available/efficient energy control in city-wide electricity management.  Edge computing is expected to play a significant role in these systems to improve the transmission efficiency of electricity, to react to, and restore power after a disturbance, to reduce operation costs, and to reuse energy effectively, since these operations involve local decision-making. In addition, edge computing can help monitor power generation and power demand, and make local electrical energy storage decisions in smart grid systems.</t>
        <t><strong>Smart Agriculture</strong></t>
        <t>Smart agriculture integrates information and communication technologies with farming technology. Intelligent farms use IoT technology to measure and analyze parameters, such as the temperature, humidity, sunlight, carbon dioxide, and soil quality, in crop cultivation facilities. Depending on the analysis results, control devices are used to set the environmental parameters to an appropriate state. Remote management is also possible through mobile devices such as smartphones.</t>
        <t>In existing farms, simple systems such as management according to temperature and humidity can be easily and inexpensively implemented using IoT technology. Field sensors gather data on field and crop condition. This data is then transmitted to cloud servers that process data and recommend actions. The use of edge computing can reduce the volume of back-and-forth data transmissions significantly, resulting in cost and bandwidth savings. Locally generated data can be processed at the edge, and local computing and analytics can drive local actions. With edge computing, it is easy for farmers to select large amounts of data for processing, and data can be analyzed even in remote areas with poor access conditions. Other applications include enabling dashboarding, for example, to visualize the farm status, as well as enhancing Extended Reality (XR) applications that require edge audio/video processing. As the number of people working on farming has been decreasing over time, increasing automation enabled by edge computing can be a driving force for future smart agriculture.</t>
        <t><strong>Smart Construction</strong></t>
        <t>Safety is critical at construction sites. Every year, many construction workers lose their lives because of falls, collisions, electric shocks, and other accidents.  Therefore, solutions have been developed to improve construction site safety, including the real-time identification of workers, monitoring of equipment location, and predictive accident prevention. To deploy these solutions, many cameras and IoT sensors have been installed on construction sites, to measure noise, vibration, gas concentration, etc. Typically, the data generated from these measurements is collected in on-site gateways and sent to remote cloud servers for storage and analysis. Thus, an inspector can check the information stored on the cloud server to investigate an incident. However, this approach can be expensive because of transmission costs, for example, of video streams over a mobile network connection, and because usage fees of private cloud services.</t>
        <t>Using edge computing, data generated at the construction site can be processed and analyzed on an edge server located within or near the site. Only the result of this processing needs to be transferred to a cloud server, thus reducing transmission costs. It is also possible to locally generate warnings to prevent accidents in real-time.</t>
        <t><strong>Self-Driving Car</strong></t>
        <t>Edge computing plays a crucial role in safety-focused self-driving car systems. With a multitude of sensors, such as high-resolution cameras, radar, LIDAR, sonar sensors, and GPS systems, autonomous vehicles generate vast amounts of real-time data. Local processing utilizing edge computing nodes allows for efficient collection and analysis of this data to monitor vehicle distances and road conditions and respond promptly to unexpected situations. Roadside computing nodes can also be leveraged to offload tasks when necessary, for example, when the local processing capacity of the car is insufficient because of hardware constraints or a large data volume.</t>
        <t>For instance, when the car ahead slows, a self-driving car adjusts its speed to maintain a safe distance, or when a roadside signal changes, it adapts its behavior accordingly. In another example, cars equipped with self-parking features utilize local processing to analyze sensor data, determine suitable parking spots, and execute precise parking maneuvers without relying on external processing or connectivity. It is also possible to use in-cabin cameras coupled with local processing to monitor the driver's attention level and detect signs of drowsiness or distraction. The system can issue warnings or implement preventive measures to ensure driver safety.</t>
        <t>Edge computing empowers self-driving cars by enabling real-time processing, reducing latency, enhancing data privacy, and optimizing bandwidth usage. By leveraging local processing capabilities, self-driving cars can make rapid decisions, adapt to changing environments, and ensure safer and more efficient autonomous driving experiences.</t>
        <t><strong>Digital Twin</strong></t>
        <t>A digital twin can simulate different scenarios and predict outcomes based on real-time data collected from the physical environment. This simulation capability empowers proactive maintenance, optimization of operations, and the prediction of potential issues or failures. Decision makers can use digital twins to test and validate different strategies, identify inefficiencies, and optimize performance.</t>
        <t>With edge computing, real-time data is collected, processed, and analyzed directly at the edge, allowing for the accurate monitoring and simulation of physical assets. Moreover, edge computing effectively minimizes latency, enabling rapid responses to dynamic conditions as computational resources are brought closer to the physical object. Running digital twin processing at the edge enables organizations to obtain timely insights and make informed decisions that maximize efficiency and performance.</t>
        <t><strong>Other Use Cases</strong></t>
        <t>AI/ML systems at the edge empower real-time analysis, faster decision-making, reduced latency, improved operational efficiency, and personalized experiences across various industries, by bringing artificial intelligence and machine learning capabilities closer to edge devices.</t>
        <t>In addition, oneM2M has studied several IoT edge computing use cases, which are documented in <xref target="oneM2M-TR0001"/>, <xref target="oneM2M-TR0018"/> and <xref target="oneM2M-TR0026"/>. The edge computing related requirements raised through the analysis of these use cases are captured in <xref target="oneM2M-TS0002"/>.</t>
      </section>
    </section>
    <section anchor="sec-challenges">
      <name>IoT Challenges Leading Towards Edge Computing</name>
      <t>This section describes the challenges faced by IoT that are motivating the adoption of edge computing. These are distinct from the research challenges applicable to IoT edge computing, some of which are mentioned in <xref target="sec-functions"/>.</t>
      <t>IoT technology is used with increasingly demanding applications, for example, in industrial, automotive and healthcare domains, leading to new challenges.  For example, industrial machines such as laser cutters produce over 1 terabyte of data per hour, and similar amounts can be generated in autonomous cars <xref target="NVIDIA"/>.  90% of IoT data is expected to be stored, processed, analyzed, and acted upon close to the source <xref target="Kelly"/>, as cloud computing models alone cannot address these new challenges <xref target="Chiang"/>.</t>
      <t>Below, we discuss IoT use case requirements that are moving cloud capabilities to be more proximate, distributed, and disaggregated.</t>
      <section anchor="time-sensitivity">
        <name>Time Sensitivity</name>
        <t>Many industrial control systems, such as manufacturing systems, smart grids, and oil and gas systems often require stringent end-to-end latency between the sensor and control nodes.  While some IoT applications may require latency below a few tens of milliseconds <xref target="Weiner"/>, industrial robots and motion control systems have use cases for cycle times in the order of microseconds <xref target="_60802"/>.  In some cases, speed-of-light limitations may simply prevent a cloud-based solutions; however, this is not the only challenge relative to time sensitivity.  Guarantees for bounded latency and jitter (<xref target="RFC8578"/> section 7) are also important for industrial IoT applications.  This means that control packets must arrive with as little variation as possible and within a strict deadline.  Given the best-effort characteristics of the Internet, this challenge is virtually impossible to address, without using end-to-end guarantees for individual message delivery and continuous data flows.</t>
      </section>
      <section anchor="connectivity-cost">
        <name>Connectivity Cost</name>
        <t>Some IoT deployments may not face bandwidth constraints when uploading data to the Cloud.  5G and Wi-Fi 6 networks both theoretically top out at 10 gigabits per second (i.e., 4.5 terabytes per hour), allowing to transfer large amounts of uplink data.  However, the cost of maintaining continuous high-bandwidth connectivity for such usage is unjustifiable and impractical for most IoT applications.  In some settings, for example, in aeronautical communication, higher communication costs reduce the amount of data that can be practically uploaded even further.  Minimizing reliance on high-bandwidth connectivity is therefore a requirement, for example, by processing data at the edge and deriving summarized or actionable insights that can be transmitted to the Cloud.</t>
      </section>
      <section anchor="resilience-to-intermittent-services">
        <name>Resilience to Intermittent Services</name>
        <t>Many IoT devices, such as sensors, actuators, and controllers, have very limited hardware resources and cannot rely solely on their own resources to meet their computing and/or storage needs.  They require reliable, uninterrupted, or resilient services to augment their capabilities to fulfill their application tasks.  This is difficult and partly impossible to achieve using cloud services for systems such as vehicles, drones, or oil rigs that have intermittent network connectivity.  Conversely, a cloud back-end might want to device data even if it is currently asleep.</t>
      </section>
      <section anchor="sec-priv">
        <name>Privacy and Security</name>
        <t>When IoT services are deployed at home, personal information can be learned from detected usage data.  For example, one can extract information about employment, family status, age, and income by analyzing smart-meter data <xref target="ENERGY"/>.  Policy makers have begun to provide frameworks that limit the usage of personal data and impose strict requirements on data controllers and processors.  Data stored indefinitely in the Cloud also increases the risk of data leakage, for instance, through attacks on rich targets.</t>
        <t>It is often argues that industrial systems do not provide privacy implications, as no personal data is gathered.  However, data from such systems is often highly sensitive, as one might be able to infer trade secrets such as the setup of production lines.  Hence, owners of these systems are generally reluctant to upload IoT data to the Cloud.</t>
        <t>Furthermore, passive observers can perform traffic analysis on device-to-cloud paths.  Therefore, hiding traffic patterns associated with sensor networks can be another requirement for edge computing.</t>
      </section>
    </section>
    <section anchor="sec-functions">
      <name>IoT Edge Computing Functions</name>
      <t>We first look at the current state of IoT edge computing (<xref target="sec-overview"/>), and then define a general system model (<xref target="sec-model"/>). This provides a context for IoT edge-computing functions, which are listed in <xref target="sec-components-oam"/>, <xref target="sec-components-functional"/> and <xref target="sec-components-app"/>.</t>
      <section anchor="sec-overview">
        <name>Overview of IoT Edge Computing Today</name>
        <t>This section provides an overview of today's IoT edge computing field based on a limited review of standards, research, open-source and proprietary products in <xref target="I-D.defoy-t2trg-iot-edge-computing-background"/>.</t>
        <t>IoT gateways, both open-source (such as EdgeX Foundry or Home Edge) and proprietary products, represent a common class of IoT edge-computing products, where the gateway provides a local service on customer premises and is remotely managed through a cloud service. IoT communication protocols are typically used between IoT devices and the gateway, including CoAP <xref target="RFC7252"/>, MQTT <xref target="mqtt5"/>, and many specialized IoT protocols (such as OPC UA and DDS in the Industrial IoT space), while the gateway communicates with the distant cloud typically using HTTPS. Virtualization platforms enable the deployment of virtual edge computing functions (using VMs and application containers), including IoT gateway software, on servers in the mobile network infrastructure (at base stations and concentration points), edge data centers (in central offices), and regional data centers located near central offices. End devices are envisioned to become computing devices in forward-looking projects, but are not commonly used today.</t>
        <t>In addition to open-source and proprietary solutions, a horizontal IoT service layer is standardized by the oneM2M standards body to reduce fragmentation, increase interoperability and promote reuse in the IoT ecosystem. Furthermore, ETSI MEC developed an IoT API <xref target="ETSI_MEC_33"/> that enables the deployment of heterogeneous IoT platforms and provides a means to configure the various components of an IoT system.</t>
        <t>Physical or virtual IoT gateways can host application programs that are typically built using an SDK to access local services through a programmatic API.  Edge cloud system operators host their customers' application VMs or containers on servers located in or near access networks that can implement local edge services. For example, mobile networks can provide edge services for radio-network information, location, and bandwidth management.</t>
        <t>Resilience in the IoT can entail the ability to operate autonomously in periods of disconnectedness to preserve the integrity and safety of the controlled system, possibly in a degraded mode. IoT devices and gateways are often expected to operate in always-on and unattended modes, using fault detection and unassisted recovery functions.</t>
        <t>The life cycle management of services and applications on physical IoT gateways is generally cloud-based.  Edge cloud management platforms and products (such as StarlingX, Akraino Edge Stack, or proprietary products from major Cloud providers) adapt cloud management technologies (e.g., Kubernetes) to the edge cloud, that is, to smaller, distributed computing devices running outside a controlled data center. The service and application life-cycle is typically using an NFV-like management and orchestration model.</t>
        <t>The platform typically enables advertising or consuming services hosted on the platform (e.g., the Mp1 interface in ETSI MEC supports service discovery and communication), and enables communication with local and remote endpoints (e.g., message routing function in IoT gateways).  The platform is typically extensible to edge applications because it can advertise a service that other edge applications can consume. The IoT communication services include protocol translation, analytics, and transcoding.  Communication between edge-computing devices is enabled in tiered or distributed deployments.</t>
        <t>An edge cloud platform may enable pass-through without storage or local storage (e.g., on IoT gateways). Some edge cloud platforms use distributed storage such as that provided by a distributed storage platform (e.g., EdgeFS, Ceph), or, in more experimental settings, by an ICN network, for example, systems such as Chipmunk <xref target="chipmunk"/> and Kua <xref target="kua"/> have been proposed as distributed information-centric objects stores.  External storage, for example, on databases in distant or local IT cloud, is typically used for filtered data deemed worthy of long-term storage, although in some cases it may be for all data, for example when required for regulatory reasons.</t>
        <t>Stateful computing is supported on platforms that host native programs, VMs, or containers. Stateless computing is supported on platforms providing a "serverless computing" service (also known as function-as-a-service, e.g., using stateless containers), or on systems based on named function networking.</t>
        <t>In many IoT use cases, a typical network usage pattern is a high volume uplink with some form of traffic reduction enabled by processing over edge-computing devices. Alternatives to traffic reduction include deferred transmission (to off-peak hours or using physical shipping). Downlink traffic includes application control and software updates. Downlink-heavy traffic patterns are not excluded but are more often associated with non-IoT usage (e.g., video CDNs).</t>
      </section>
      <section anchor="sec-model">
        <name>General Model</name>
        <t>Edge computing is expected to play an important role in deploying new IoT services integrated with Big Data and AI enabled by flexible in-network computing platforms. Although there are many approaches to edge computing, in this section, we attempt to lay out a general model and the list associated logical functions. In practice, this model can be mapped to different architectures, such as:</t>
        <ul spacing="normal">
          <li>A single IoT gateway, or a hierarchy of IoT gateways, typically connected to the cloud (e.g., to extend the traditional cloud-based management of IoT devices and data to the edge). The IoT gateway plays a common role in providing access to a heterogeneous set of IoT devices/sensors, handling IoT data, and delivering IoT data to its final destination in a cloud network. Whereas an IoT gateway requires interactions with the cloud, it can also operate independently in a disconnected mode.</li>
          <li>A set of distributed computing nodes, for example, embedded in switches, routers, edge cloud servers, or mobile devices. Some IoT devices have sufficient computing capabilities to participate in such distributed systems owing to advances in hardware technology. In this model, edge-computing nodes can collaborate to share resources.</li>
          <li>A hybrid system involving both IoT gateways and supporting functions in distributed computing nodes.</li>
        </ul>
        <t>In the general model described in <xref target="rl-fig1"/>, the edge computing domain is interconnected with IoT devices (southbound connectivity), possibly with a remote/cloud network (northbound connectivity), and with a service operator's system.  Edge-computing nodes provide multiple logical functions or components that may not be present in a given system. They may be implemented in a centralized or distributed fashion, at the network edge, or through interworking between the edge network and remote cloud networks.</t>
        <figure anchor="rl-fig1">
          <name>Model of IoT Edge Computing</name>
          <artset>
            <artwork type="svg" align="center"><svg xmlns="http://www.w3.org/2000/svg" version="1.1" height="656" width="392" viewBox="0 0 392 656" class="diagram" text-anchor="middle" font-family="monospace" font-size="13px" stroke-linecap="round">
                <path d="M 8,128 L 8,528" fill="none" stroke="black"/>
                <path d="M 24,576 L 24,624" fill="none" stroke="black"/>
                <path d="M 32,32 L 32,80" fill="none" stroke="black"/>
                <path d="M 64,528 L 64,576" fill="none" stroke="black"/>
                <path d="M 96,576 L 96,624" fill="none" stroke="black"/>
                <path d="M 128,80 L 128,128" fill="none" stroke="black"/>
                <path d="M 136,576 L 136,624" fill="none" stroke="black"/>
                <path d="M 184,528 L 184,576" fill="none" stroke="black"/>
                <path d="M 208,32 L 208,80" fill="none" stroke="black"/>
                <path d="M 208,576 L 208,624" fill="none" stroke="black"/>
                <path d="M 232,48 L 232,96" fill="none" stroke="black"/>
                <path d="M 272,560 L 272,624" fill="none" stroke="black"/>
                <path d="M 288,96 L 288,128" fill="none" stroke="black"/>
                <path d="M 320,528 L 320,560" fill="none" stroke="black"/>
                <path d="M 344,560 L 344,624" fill="none" stroke="black"/>
                <path d="M 360,48 L 360,96" fill="none" stroke="black"/>
                <path d="M 384,128 L 384,544" fill="none" stroke="black"/>
                <path d="M 32,32 L 208,32" fill="none" stroke="black"/>
                <path d="M 232,48 L 360,48" fill="none" stroke="black"/>
                <path d="M 32,80 L 208,80" fill="none" stroke="black"/>
                <path d="M 232,96 L 360,96" fill="none" stroke="black"/>
                <path d="M 8,128 L 384,128" fill="none" stroke="black"/>
                <path d="M 8,528 L 248,528" fill="none" stroke="black"/>
                <path d="M 312,528 L 328,528" fill="none" stroke="black"/>
                <path d="M 272,560 L 344,560" fill="none" stroke="black"/>
                <path d="M 24,576 L 96,576" fill="none" stroke="black"/>
                <path d="M 136,576 L 208,576" fill="none" stroke="black"/>
                <path d="M 24,624 L 96,624" fill="none" stroke="black"/>
                <path d="M 136,624 L 208,624" fill="none" stroke="black"/>
                <path d="M 272,624 L 344,624" fill="none" stroke="black"/>
                <g class="text">
                  <text x="84" y="52">Remote</text>
                  <text x="144" y="52">network</text>
                  <text x="60" y="68">(e.g.,</text>
                  <text x="112" y="68">cloud</text>
                  <text x="172" y="68">network)</text>
                  <text x="288" y="68">Service</text>
                  <text x="292" y="84">Operator</text>
                  <text x="124" y="148">Edge</text>
                  <text x="184" y="148">Computing</text>
                  <text x="252" y="148">Domain</text>
                  <text x="48" y="180">One</text>
                  <text x="76" y="180">or</text>
                  <text x="108" y="180">more</text>
                  <text x="168" y="180">Computing</text>
                  <text x="232" y="180">Nodes</text>
                  <text x="52" y="196">(IoT</text>
                  <text x="108" y="196">gateway,</text>
                  <text x="160" y="196">end</text>
                  <text x="212" y="196">devices,</text>
                  <text x="288" y="196">switches,</text>
                  <text x="68" y="212">routers,</text>
                  <text x="168" y="212">mini/micro-data</text>
                  <text x="268" y="212">centers,</text>
                  <text x="328" y="212">etc.)</text>
                  <text x="48" y="244">OAM</text>
                  <text x="108" y="244">Components</text>
                  <text x="40" y="260">-</text>
                  <text x="84" y="260">Resource</text>
                  <text x="160" y="260">Discovery</text>
                  <text x="216" y="260">and</text>
                  <text x="292" y="260">Authentication</text>
                  <text x="40" y="276">-</text>
                  <text x="68" y="276">Edge</text>
                  <text x="140" y="276">Organization</text>
                  <text x="208" y="276">and</text>
                  <text x="268" y="276">Federation</text>
                  <text x="40" y="292">-</text>
                  <text x="104" y="292">Multi-Tenancy</text>
                  <text x="176" y="292">and</text>
                  <text x="232" y="292">Isolation</text>
                  <text x="40" y="308">-</text>
                  <text x="64" y="308">...</text>
                  <text x="76" y="340">Functional</text>
                  <text x="164" y="340">Components</text>
                  <text x="40" y="356">-</text>
                  <text x="92" y="356">In-Network</text>
                  <text x="184" y="356">Computation</text>
                  <text x="40" y="372">-</text>
                  <text x="68" y="372">Edge</text>
                  <text x="120" y="372">Caching</text>
                  <text x="40" y="388">-</text>
                  <text x="104" y="388">Communication</text>
                  <text x="40" y="404">-</text>
                  <text x="72" y="404">Other</text>
                  <text x="132" y="404">Services</text>
                  <text x="40" y="420">-</text>
                  <text x="64" y="420">...</text>
                  <text x="80" y="452">Application</text>
                  <text x="172" y="452">Components</text>
                  <text x="40" y="468">-</text>
                  <text x="64" y="468">IoT</text>
                  <text x="112" y="468">Devices</text>
                  <text x="188" y="468">Management</text>
                  <text x="40" y="484">-</text>
                  <text x="68" y="484">Data</text>
                  <text x="132" y="484">Management</text>
                  <text x="192" y="484">and</text>
                  <text x="248" y="484">Analytics</text>
                  <text x="40" y="500">-</text>
                  <text x="64" y="500">...</text>
                  <text x="264" y="532">-</text>
                  <text x="280" y="532">-</text>
                  <text x="296" y="532">-</text>
                  <text x="344" y="532">-</text>
                  <text x="360" y="532">-</text>
                  <text x="376" y="532">-</text>
                  <text x="248" y="548">|</text>
                  <text x="248" y="580">|</text>
                  <text x="304" y="580">compute</text>
                  <text x="384" y="580">|</text>
                  <text x="56" y="596">End</text>
                  <text x="168" y="596">End</text>
                  <text x="232" y="596">...</text>
                  <text x="308" y="596">node/end</text>
                  <text x="52" y="612">Device</text>
                  <text x="88" y="612">1</text>
                  <text x="164" y="612">Device</text>
                  <text x="200" y="612">2</text>
                  <text x="236" y="612">...|</text>
                  <text x="300" y="612">device</text>
                  <text x="336" y="612">n</text>
                  <text x="384" y="612">|</text>
                  <text x="248" y="644">+</text>
                  <text x="264" y="644">-</text>
                  <text x="280" y="644">-</text>
                  <text x="296" y="644">-</text>
                  <text x="312" y="644">-</text>
                  <text x="328" y="644">-</text>
                  <text x="344" y="644">-</text>
                  <text x="360" y="644">-</text>
                  <text x="380" y="644">-+</text>
                </g>
              </svg>
            </artwork>
            <artwork type="ascii-art" align="center"><![CDATA[
   +---------------------+
   |   Remote network    |  +---------------+
   |(e.g., cloud network)|  |   Service     |
   +-----------+---------+  |   Operator    |
               |            +------+--------+
               |                   |
+--------------+-------------------+-----------+
|            Edge Computing Domain             |
|                                              |
|   One or more Computing Nodes                |
|   (IoT gateway, end devices, switches,       |
|   routers, mini/micro-data centers, etc.)    |
|                                              |
|   OAM Components                             |
|   - Resource Discovery and Authentication    |
|   - Edge Organization and Federation         |
|   - Multi-Tenancy and Isolation              |
|   - ...                                      |
|                                              |
|   Functional Components                      |
|   - In-Network Computation                   |
|   - Edge Caching                             |
|   - Communication                            |
|   - Other Services                           |
|   - ...                                      |
|                                              |
|   Application Components                     |
|   - IoT Devices Management                   |
|   - Data Management and Analytics            |
|   - ...                                      |
|                                              |
+------+--------------+-------- - - - -+- - - -+
       |              |       |        |       |
       |              |          +-----+--+
  +----+---+    +-----+--+    |  |compute |    |
  |  End   |    |  End   | ...   |node/end|
  |Device 1|    |Device 2| ...|  |device n|    |
  +--------+    +--------+       +--------+
                              + - - - - - - - -+
]]></artwork>
          </artset>
        </figure>
        <t>In the distributed model described in <xref target="rl-fig2"/>, the edge-computing domain is composed of IoT edge gateways and IoT devices which are also used as computing nodes.  Edge computing domains are connected to a remote/cloud network and their respective service operator's system. IoT devices/computing nodes provide logical functions, for example as part of distributed machine learning or distributed image processing applications. The processing capabilities in IoT devices are limited; they require the support of other nodes, and in a distributed machine learning application, the training process for AI services can be executed at IoT edge gateways or cloud networks and the prediction (inference) service is executed in the IoT devices. In a distributed image processing application, some image processing functions can be similarly executed at the edge or in the cloud, while preprocessing, which helps limiting the amount of uploaded data, is performed by the IoT device.</t>
        <figure anchor="rl-fig2">
          <name>Example: Machine Learning over a Distributed IoT Edge Computing System</name>
          <artset>
            <artwork type="svg" align="center"><svg xmlns="http://www.w3.org/2000/svg" version="1.1" height="576" width="392" viewBox="0 0 392 576" class="diagram" text-anchor="middle" font-family="monospace" font-size="13px" stroke-linecap="round">
                <path d="M 8,32 L 8,240" fill="none" stroke="black"/>
                <path d="M 8,352 L 8,560" fill="none" stroke="black"/>
                <path d="M 24,80 L 24,144" fill="none" stroke="black"/>
                <path d="M 24,448 L 24,512" fill="none" stroke="black"/>
                <path d="M 32,176 L 32,208" fill="none" stroke="black"/>
                <path d="M 32,272 L 32,320" fill="none" stroke="black"/>
                <path d="M 32,384 L 32,416" fill="none" stroke="black"/>
                <path d="M 64,144 L 64,176" fill="none" stroke="black"/>
                <path d="M 64,416 L 64,448" fill="none" stroke="black"/>
                <path d="M 96,80 L 96,144" fill="none" stroke="black"/>
                <path d="M 96,448 L 96,512" fill="none" stroke="black"/>
                <path d="M 128,208 L 128,272" fill="none" stroke="black"/>
                <path d="M 128,320 L 128,384" fill="none" stroke="black"/>
                <path d="M 136,80 L 136,144" fill="none" stroke="black"/>
                <path d="M 136,448 L 136,512" fill="none" stroke="black"/>
                <path d="M 176,144 L 176,176" fill="none" stroke="black"/>
                <path d="M 176,416 L 176,448" fill="none" stroke="black"/>
                <path d="M 208,80 L 208,144" fill="none" stroke="black"/>
                <path d="M 208,272 L 208,320" fill="none" stroke="black"/>
                <path d="M 208,448 L 208,512" fill="none" stroke="black"/>
                <path d="M 232,272 L 232,320" fill="none" stroke="black"/>
                <path d="M 280,80 L 280,144" fill="none" stroke="black"/>
                <path d="M 280,448 L 280,512" fill="none" stroke="black"/>
                <path d="M 288,208 L 288,272" fill="none" stroke="black"/>
                <path d="M 288,320 L 288,384" fill="none" stroke="black"/>
                <path d="M 320,144 L 320,176" fill="none" stroke="black"/>
                <path d="M 320,416 L 320,448" fill="none" stroke="black"/>
                <path d="M 344,176 L 344,208" fill="none" stroke="black"/>
                <path d="M 344,384 L 344,416" fill="none" stroke="black"/>
                <path d="M 352,80 L 352,144" fill="none" stroke="black"/>
                <path d="M 352,272 L 352,320" fill="none" stroke="black"/>
                <path d="M 352,448 L 352,512" fill="none" stroke="black"/>
                <path d="M 384,32 L 384,240" fill="none" stroke="black"/>
                <path d="M 384,352 L 384,560" fill="none" stroke="black"/>
                <path d="M 8,32 L 384,32" fill="none" stroke="black"/>
                <path d="M 24,80 L 96,80" fill="none" stroke="black"/>
                <path d="M 136,80 L 208,80" fill="none" stroke="black"/>
                <path d="M 280,80 L 352,80" fill="none" stroke="black"/>
                <path d="M 24,144 L 96,144" fill="none" stroke="black"/>
                <path d="M 136,144 L 208,144" fill="none" stroke="black"/>
                <path d="M 280,144 L 352,144" fill="none" stroke="black"/>
                <path d="M 32,176 L 344,176" fill="none" stroke="black"/>
                <path d="M 32,208 L 344,208" fill="none" stroke="black"/>
                <path d="M 8,240 L 384,240" fill="none" stroke="black"/>
                <path d="M 32,272 L 208,272" fill="none" stroke="black"/>
                <path d="M 232,272 L 352,272" fill="none" stroke="black"/>
                <path d="M 32,320 L 208,320" fill="none" stroke="black"/>
                <path d="M 232,320 L 352,320" fill="none" stroke="black"/>
                <path d="M 8,352 L 384,352" fill="none" stroke="black"/>
                <path d="M 32,384 L 344,384" fill="none" stroke="black"/>
                <path d="M 32,416 L 344,416" fill="none" stroke="black"/>
                <path d="M 24,448 L 96,448" fill="none" stroke="black"/>
                <path d="M 136,448 L 208,448" fill="none" stroke="black"/>
                <path d="M 280,448 L 352,448" fill="none" stroke="black"/>
                <path d="M 24,512 L 96,512" fill="none" stroke="black"/>
                <path d="M 136,512 L 208,512" fill="none" stroke="black"/>
                <path d="M 280,512 L 352,512" fill="none" stroke="black"/>
                <path d="M 8,560 L 384,560" fill="none" stroke="black"/>
                <g class="text">
                  <text x="124" y="52">Edge</text>
                  <text x="184" y="52">Computing</text>
                  <text x="252" y="52">Domain</text>
                  <text x="56" y="100">Compute</text>
                  <text x="168" y="100">Compute</text>
                  <text x="312" y="100">Compute</text>
                  <text x="60" y="116">node/End</text>
                  <text x="172" y="116">node/End</text>
                  <text x="244" y="116">....</text>
                  <text x="316" y="116">node/End</text>
                  <text x="52" y="132">device</text>
                  <text x="88" y="132">1</text>
                  <text x="164" y="132">device</text>
                  <text x="200" y="132">2</text>
                  <text x="244" y="132">....</text>
                  <text x="308" y="132">device</text>
                  <text x="344" y="132">m</text>
                  <text x="136" y="196">IoT</text>
                  <text x="172" y="196">Edge</text>
                  <text x="224" y="196">Gateway</text>
                  <text x="84" y="292">Remote</text>
                  <text x="144" y="292">network</text>
                  <text x="288" y="292">Service</text>
                  <text x="60" y="308">(e.g.,</text>
                  <text x="112" y="308">cloud</text>
                  <text x="172" y="308">network)</text>
                  <text x="296" y="308">Operator(s)</text>
                  <text x="136" y="404">IoT</text>
                  <text x="172" y="404">Edge</text>
                  <text x="224" y="404">Gateway</text>
                  <text x="56" y="468">Compute</text>
                  <text x="168" y="468">Compute</text>
                  <text x="312" y="468">Compute</text>
                  <text x="60" y="484">node/End</text>
                  <text x="172" y="484">node/End</text>
                  <text x="244" y="484">....</text>
                  <text x="316" y="484">node/End</text>
                  <text x="52" y="500">device</text>
                  <text x="88" y="500">1</text>
                  <text x="164" y="500">device</text>
                  <text x="200" y="500">2</text>
                  <text x="244" y="500">....</text>
                  <text x="308" y="500">device</text>
                  <text x="344" y="500">n</text>
                  <text x="124" y="548">Edge</text>
                  <text x="184" y="548">Computing</text>
                  <text x="252" y="548">Domain</text>
                </g>
              </svg>
            </artwork>
            <artwork type="ascii-art" align="center"><![CDATA[
+----------------------------------------------+
|            Edge Computing Domain             |
|                                              |
| +--------+    +--------+        +--------+   |
| |Compute |    |Compute |        |Compute |   |
| |node/End|    |node/End|  ....  |node/End|   |
| |device 1|    |device 2|  ....  |device m|   |
| +----+---+    +----+---+        +----+---+   |
|      |             |                 |       |
|  +---+-------------+-----------------+--+    |
|  |           IoT Edge Gateway           |    |
|  +-----------+-------------------+------+    |
|              |                   |           |
+--------------+-------------------+-----------+
               |                   |
   +-----------+---------+  +------+-------+
   |   Remote network    |  |   Service    |
   |(e.g., cloud network)|  |  Operator(s) |
   +-----------+---------+  +------+-------+
               |                   |
+--------------+-------------------+-----------+
|              |                   |           |
|  +-----------+-------------------+------+    |
|  |           IoT Edge Gateway           |    |
|  +---+-------------+-----------------+--+    |
|      |             |                 |       |
| +----+---+    +----+---+        +----+---+   |
| |Compute |    |Compute |        |Compute |   |
| |node/End|    |node/End|  ....  |node/End|   |
| |device 1|    |device 2|  ....  |device n|   |
| +--------+    +--------+        +--------+   |
|                                              |
|            Edge Computing Domain             |
+----------------------------------------------+
]]></artwork>
          </artset>
        </figure>
        <t>In the following, we enumerate major edge computing domain components. They are here loosely organized into OAM (Operations, Administration, and Maintenance), functional, and application components, with the understanding that the distinction between these classes may not always be clear, depending on actual system architectures. Some representative research challenges are associated with those functions. We used input from co-authors, IRTF attendees, and some comprehensive reviews of the field (<xref target="Yousefpour"/>, <xref target="Zhang2"/>, <xref target="Khan"/>).</t>
      </section>
      <section anchor="sec-components-oam">
        <name>OAM Components</name>
        <t>Edge computing OAM extends beyond the network-related OAM functions listed in <xref target="RFC6291"/>. In addition to infrastructure (network, storage, and computing resources), edge computing systems can also include computing environments (for VMs, software containers, functions), IoT devices, data, and code.</t>
        <t>Operation-related functions include performance monitoring for service-level agreement measurements, fault management and provisioning for links, nodes, compute and storage resources, platforms, and services. Administration covers network/compute/storage resources, platforms and services discovery, configuration, and planning. Discovery during normal operation (e.g., discovery of compute or storage nodes by endpoints) is typically not included in OAM; however, in this document, we do not address it separately. Management covers the monitoring and diagnostics of failures, as well as means to minimize their occurrence and take corrective actions. This may include software update management and high service availability through redundancy and multipath communication. Centralized (e.g., SDN) and decentralized management systems can be used. Finally, we arbitrarily chose to address data management as an application component, however, in some systems, data management may be considered similar to a network management function.</t>
        <t>We further detail a few relevant OAM components.</t>
        <section anchor="sec-dis-auth">
          <name>Resource Discovery and Authentication</name>
          <t>Discovery and authentication may target platforms and , infrastructure resources, such as computing, networking, and storage, as well as other resources such as IoT devices, sensors, data, code units, services, applications, and users interacting with the system. Broker-based solutions can be used, for example, using an IoT gateway as a broker to discover IoT resources. More decentralized solutions can also be used in replacement or complement, for example, CoAP enables multicast discovery of an IoT device, and CoAP service discovery enables obtaining a list of resources made available by this device <xref target="RFC7252"/>. For device authentication, current centralized gateway-based systems rely on the installation of a secret on IoT devices and computing devices (e.g., a device certificate stored in a hardware security module, or a combination of code and data stored in a trusted execution environment).</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>Discovery, authentication, and trust establishment between IoT devices, compute nodes, and platforms, with regard to concerns such as mobility, heterogeneous devices and networks, scale, multiple trust domains, constrained devices, anonymity, and traceability.</li>
            <li>Intermittent connectivity to the Internet, removing the need to rely on a third-party authority <xref target="Echeverria"/>.</li>
            <li>Resiliency to failure <xref target="Harchol"/>, denial of service attacks, easier physical access for attackers.</li>
          </ul>
        </section>
        <section anchor="edge-organization-and-federation">
          <name>Edge Organization and Federation</name>
          <t>In a distributed system context, once edge devices have discovered and authenticated each other, they can be organized, or self-organized, into hierarchies or clusters. The organizational structure may range from centralized to peer-to-peer, or it may be closely tied to other systems. Such groups can also form federations with other edges or with remote clouds.</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>Support for scaling, and enabling fault-tolerance or self-healing <xref target="Jeong"/>. In addition to using a hierarchical organization to cope with scaling, another available and possibly complementary mechanism is multicast (<xref target="RFC7390"/> <xref target="I-D.ietf-core-groupcomm-bis"/>). Other approaches include relying on blockchains <xref target="Ali"/>.</li>
            <li>Integration of edge computing with virtualized Radio Access Networks (Fog RAN) <xref target="I-D.bernardos-sfc-fog-ran"/> and 5G access networks.</li>
            <li>Sharing resources in multi-vendor/operator scenarios, to optimize criteria such as profit <xref target="Anglano"/>, resource usage, latency, and energy consumption.</li>
            <li>Capacity planning, placement of infrastructure nodes to minimize delay <xref target="Fan"/>, cost, energy, etc.</li>
            <li>Incentives for participation, for example, in peer-to-peer federation schemes.</li>
            <li>Design of federated AI over IoT edge computing systems <xref target="Brecko"/>, for example, for anomaly detection.</li>
          </ul>
        </section>
        <section anchor="multi-tenancy-and-isolation">
          <name>Multi-Tenancy and Isolation</name>
          <t>Some IoT edge computing systems make use of virtualized (compute, storage and networking) resources to address the need for secure multi-tenancy at the edge. This leads to "edge clouds" that share properties with remotes clouds and can reuse some of their ecosystems. Virtualization function management is largely covered by ETSI NFV and MEC standards and recommendations. Projects such as <xref target="LFEDGE-EVE"/> further cover virtualization and its management in distributed edge-computing settings.</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>Adapting cloud management platforms to the edge, to account for its distributed nature, e.g., using Conflict-free Replicated Data Types (CRDT) <xref target="Jeffery"/>, heterogeneity and customization, e.g., using intent-based management mechanisms <xref target="Cao"/>, and limited resources.</li>
            <li>Minimizing virtual function instantiation time and resource usage.</li>
          </ul>
        </section>
      </section>
      <section anchor="sec-components-functional">
        <name>Functional Components</name>
        <section anchor="in-network-computation">
          <name>In-Network Computation</name>
          <t>A core function of IoT edge computing is to enable local computation on a node at the network edge, typically for application-layer processing, such as processing input data from sensors, making local decisions, preprocessing data, offloading computation on behalf of a device, service, or user. Related functions include orchestrating computation (in a centralized or distributed manner) and managing application lifecycles. Support for in-network computation may vary in terms of capability, for example, computing nodes can host virtual machines, software containers, software actors, uni-kernels running stateful or stateless code, or a rule engine providing an API to register actions in response to conditions such as IoT device ID, sensor values to check, thresholds, etc.</t>
          <t>Edge offloading includes offloading to and from an IoT device, and to and from a network node. <xref target="Cloudlets"/> offer an example of offloading computation from an end device to a network node. In contrast, oneM2M is an example of a system that allows a cloud-based IoT platform to transfer resources and tasks to a target edge node <xref target="oneM2M-TR0052"/>. Once transferred, the edge node can directly support IoT devices that it serves with the service offloaded by the cloud (e.g., group management, location management, etc.).</t>
          <t>QoS can be provided in some systems through the combination of network QoS (e.g., traffic engineering or wireless resource scheduling) and compute/storage resource allocations. For example, in some systems, a bandwidth manager service can be exposed to enable allocation of the bandwidth to/from an edge-computing application instance.</t>
          <t>In-network computation can leverage the underlying services, provided using data generated by IoT devices and access networks. Such services include IoT device location, radio network information, bandwidth management and congestion management (e.g., the congestion management feature of oneM2M <xref target="oneM2M-TR0052"/>).</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>(Computation placement) Selecting, in a centralized or distributed/peer-to-peer manner, an appropriate compute device based on available resources, location of data input and data sinks, compute node properties, etc., and with varying goals including end-to-end latency, privacy, high availability, energy conservation, or network efficiency, for example, using load-balancing techniques to avoid congestion.</li>
            <li>Onboarding code on a platform or computing device, and invoking remote code execution, possibly as part of a distributed programming model and with respect to similar concerns of latency, privacy, etc.: For example, offloading can be included in a vehicular scenario <xref target="Grewe"/>. These operations should deal with heterogeneous compute nodes <xref target="Schafer"/>, and may also support end devices, including IoT devices, as compute nodes <xref target="Larrea"/>.</li>
            <li>Adapting Quality of Results (QoR) for applications where a perfect result is not necessary <xref target="Li"/>.</li>
            <li>Assisted or automatic partitioning of code: for example, for application programs <xref target="I-D.sarathchandra-coin-appcentres"/> or network programs <xref target="I-D.hsingh-coinrg-reqs-p4comp"/>.</li>
            <li>Supporting computation across trust domains: for example, verifying computation results.</li>
            <li>Support for computation mobility: relocating an instance from one compute node to another, while maintaining a given service level; session continuity when communicating with end devices that are mobile, possibly at high speed (e.g., in vehicular scenarios); defining lightweight execution environments for secure code mobility, for example, using WebAssembly <xref target="Nieke"/>.</li>
            <li>Defining, managing, and verifying Service Level Agreements (SLA) for edge-computing systems: pricing is a challenging task.</li>
          </ul>
        </section>
        <section anchor="edge-storage-and-caching">
          <name>Edge Storage and Caching</name>
          <t>Local storage or caching enables local data processing (e.g., preprocessing or analysis) as well as delayed data transfer to the cloud or delayed physical shipping.  An edge node may offer local data storage (in which persistence is subject to retention policies), caching, or both.  Caching generally refers to temporary storage to improve performance without persistence guarantees.  An edge-caching component manages data persistence, for example, it schedules the removal of data when it is no longer needed.  Other related aspects include the authentication and encryption of data.  Edge storage and caching can take the form of a distributed storage systems.</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>(Cache and data placement) Using cache positioning and data placement strategies to minimize data retrieval delay <xref target="Liu"/> and energy consumption. Caches may be positioned in the access network infrastructure or on end devices.</li>
            <li>Maintaining consistency, freshness, reliability, and privacy of stored/cached data in systems that are distributed, constrained, and dynamic (e.g., owing to end devices and computing nodes churn or mobility), and which can have additional data governance constraints on data storage location. For example, <xref target="Mortazavi"/> leverages a hierarchical storage organization. Freshness-related metrics include the age of information <xref target="Yates"/> that captures the timeliness of information received from a sender (e.g., an IoT device).</li>
          </ul>
        </section>
        <section anchor="communication">
          <name>Communication</name>
          <t>An edge cloud may provide a northbound data plane or management plane interface to a remote network, such as a cloud, home or enterprise network. This interface does not exist in stand-alone (local-only) scenarios. To support such an interface when it exists, an edge computing component needs to expose an API, deal with authentication and authorization, and support secure communication.</t>
          <t>An edge cloud may provide an API or interface to local or mobile users, for example, to provide access to services and applications, or to manage data published by local/mobile devices.</t>
          <t>Edge-computing nodes communicate with IoT devices over a southbound interface, typically for data acquisition and IoT device management.</t>
          <t>Communication brokering is a typical function of IoT edge computing that facilitates communication with IoT devices, enabling clients to register as recipients for data from devices, as well as forwarding/routing of traffic to or from IoT devices, enabling various data discovery and redistribution patterns, for example, north-south with clouds, east-west with other edge devices <xref target="I-D.mcbride-edge-data-discovery-overview"/>.  Another related aspect is dispatching alerts and notifications to interested consumers both inside and outside the edge-computing domain.  Protocol translation, analytics, and video transcoding can also be performed when necessary. Communication brokering may be centralized in some systems, for example, using a hub-and-spoke message broker, or distributed with message buses, possibly in a layered bus approach.  Distributed systems can leverage direct communication between end devices over device-to-device links.  A broker can ensure communication reliability and traceability and, in some cases, transaction management.</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>Defining edge computing abstractions, such as PaaS <xref target="Yangui"/>, suitable for users and cloud systems to interact with edge computing systems and dealing with interoperability issues such as data model heterogeneity.</li>
            <li>Enabling secure and resilient communication between IoT devices and remote cloud, for example, through multipath support.</li>
          </ul>
        </section>
      </section>
      <section anchor="sec-components-app">
        <name>Application Components</name>
        <t>IoT edge computing can host applications, such as those mentioned in <xref target="sec-uc"/>. While describing the components of individual applications is out of our scope, some of those applications share similar functions, such as IoT device management and data management, as described below.</t>
        <section anchor="iot-device-management">
          <name>IoT Device Management</name>
          <t>IoT device management includes managing information regarding IoT devices, including their sensors, and how to communicate with them. Edge computing addresses the scalability challenges of a large number of IoT devices by separating the scalability domain into edge/local networks and remote networks. For example, in the context of the oneM2M standard, a device management functionality (called "software campaign" in oneM2M) enables the installation, deletion, activation, and deactivation of software functions/services on a potentially large number of edge nodes <xref target="oneM2M-TR0052"/>. Using a dashboard or management software, a service provider issues these requests through an IoT cloud platform supporting the software campaign functionality.</t>
          <t>Challenges listed in <xref target="sec-dis-auth"/> may be applicable to IoT devices management as well.</t>
        </section>
        <section anchor="sec-data">
          <name>Data Management and Analytics</name>
          <t>Data storage and processing at the edge are major aspects of IoT edge computing, directly addressing the high-level IoT challenges listed in <xref target="sec-challenges"/>. Data analysis, for example, through AI/ML tasks performed at the edge, may benefit from specialized hardware support on the computing nodes.</t>
          <t>Related challenges include:</t>
          <ul spacing="normal">
            <li>Addressing concerns regarding resource usage, security, and privacy when sharing, processing, discovering, or managing data: for example presenting data in views composed of an aggregation of related data <xref target="Zhang"/>; protecting data communication between authenticated peers <xref target="Basudan"/>, classifying data (e.g., in terms of privacy, importance, validity), and compressing and encrypting data, for example, using homomorphic encryption to directly process encrypted data <xref target="Stanciu"/>.</li>
            <li>Other concerns regarding edge data discovery (e.g., streaming data, metadata, and events) include siloization and lack of standards in edge environments that can be dynamic (e.g., vehicular networks) and heterogeneous <xref target="I-D.mcbride-edge-data-discovery-overview"/>.</li>
            <li>Data-driven programming models <xref target="Renart"/>, for example, event-based, including handling naming and data abstractions.</li>
            <li>Data integration in an environment that without data standardization, or where different sources use different ontologies <xref target="Farnbauer-Schmidt"/>.</li>
            <li>Addressing concerns such as limited resources, privacy, dynamic, and heterogeneous environments to deploy machine learning at the edge: for example, making machine learning more lightweight and distributed (e.g., enabling distributed inference at the edge), supporting shorter training times and simplified models, and supporting models that can be compressed for efficient communication <xref target="Murshed"/>.</li>
            <li>Although edge computing can support IoT services independently of cloud computing, it can also be connected to cloud computing. Thus, the relationship between IoT edge computing and cloud computing, with regard to data management, is another potential challenge <xref target="ISO_TR"/>.</li>
          </ul>
        </section>
      </section>
      <section anchor="simulation-and-emulation-environments">
        <name>Simulation and Emulation Environments</name>
        <t>IoT Edge Computing introduces new challenges to the simulation and emulation tools used by researchers and developers. A varied set of applications, networks, and computing technologies can coexist in a distributed system, making modeling difficult. Scale, mobility, and resource management are additional challenges <xref target="SimulatingFog"/>.</t>
        <t>Tools include simulators, where simplified application logic runs on top of a fog network model, and emulators, where actual applications can be deployed, typically in software containers, over a cloud infrastructure (e.g., Docker and Kubernetes) running over a network emulating network edge conditions such as variable delays, throughput and mobility events. To gain in scale, emulated and simulated systems can be used together in hybrid federation-based approaches <xref target="PseudoDynamicTesting"/>, whereas to gain in realism, physical devices can be interconnected with emulated systems. Examples of related work and platforms include the publicly accessible MEC sandbox work recently initiated in ETSI <xref target="ETSI_Sandbox"/>, and open source simulators and emulators (<xref target="AdvantEDGE"/> emulator and tools cited in <xref target="SimulatingFog"/>). EdgeNet <xref target="Senel"/> is a globally distributed edge cloud for Internet researchers, using nodes contributed by institutions, and based on Docker for containerization and Kubernetes for deployment and node management.</t>
        <t>Digital twins are virtual instances of a physical system (twin) that are continually updated with the latter's performance, maintenance, and health status data throughout the life cycle of the physical system.  <xref target="Madni"/>. In contrast to a traditional emulation or simulated environment, digital twins, once generated, are maintained in sync by their physical twin, which can be, among many other instances, an IoT device, edge device, an edge network. The benefits of digital twins go beyond those of emulation and include accelerated business processes, enhanced productivity, and faster innovation with reduced costs <xref target="I-D.irtf-nmrg-network-digital-twin-arch"/>.</t>
      </section>
    </section>
    <section anchor="security-considerations">
      <name>Security Considerations</name>
      <t>Privacy and security are drivers of the adoption of edge computing for the IoT (<xref target="sec-priv"/>). As discussed in <xref target="sec-dis-auth"/>, authentication and trust (among computing nodes, management nodes, and end devices) can be challenging as scale, mobility, and heterogeneity increase. The sometimes disconnected nature of edge resources can avoid reliance on third-party authorities. Distributed edge computing is exposed reliability and denial of service attacks. Personal or proprietary IoT data leakage is also a major threat, particularly because of the distributed nature of the systems (<xref target="sec-data"/>). Furthermore, blockchain-based distributed IoT edge computing must be designed for privacy, since public blockchain addressing does not guarantee absolute anonymity <xref target="Ali"/>.</t>
      <t>However, edge computing also offers solutions in the security space: maintaining privacy by computing sensitive data closer to data generators is a major use case for IoT edge computing.  An edge cloud can be used to perform actions based on sensitive data or to anonymize or aggregate data prior to transmission to a remote cloud server. Edge computing communication brokering functions can also be used to secure communication between edge and cloud networks.</t>
    </section>
    <section anchor="conclusion">
      <name>Conclusion</name>
      <t>IoT edge computing plays an essential role, complementary to the cloud, in enabling IoT systems in certain situations. In this document, we presented use cases and listing the core challenges faced by IoT that drive the need for IoT edge computing. The first part of this document may therefore help focus future research efforts on the aspects of IoT edge computing where it is most useful. The second part of this document presents a general system model and structured overview of the associated research challenges and related work. The structure, based on the system model, is not meant to be restrictive and exists for the purpose of having a link between individual research areas and where they are applicable in an IoT edge computing system.</t>
    </section>
    <section anchor="iana-considerations">
      <name>IANA Considerations</name>
      <t>This document has no IANA actions.</t>
    </section>
    <section anchor="acknowledgements">
      <name>Acknowledgements</name>
      <t>The authors would like to thank Joo-Sang Youn, Akbar Rahman, Michel Roy, Robert Gazda, Rute Sofia, Thomas Fossati, Chonggang Wang, <contact fullname="Marie-José Montpetit"/>, Carlos J. Bernardos, Milan Milenkovic, Dale Seed, JaeSeung Song, Roberto Morabito, Carsten Bormann and <contact fullname="Ari Keränen"/> for their valuable comments and suggestions on this document.</t>
    </section>
  </middle>
  <back>
    <references>
      <name>Informative References</name>
      <reference anchor="I-D.mcbride-edge-data-discovery-overview">
        <front>
          <title>Edge Data Discovery for COIN</title>
          <author fullname="Mike McBride" initials="M." surname="McBride">
            <organization>Futurewei</organization>
          </author>
          <author fullname="Dirk Kutscher" initials="D." surname="Kutscher">
            <organization>Emden University</organization>
          </author>
          <author fullname="Eve Schooler" initials="E." surname="Schooler">
            <organization>Intel</organization>
          </author>
          <author fullname="Carlos J. Bernardos" initials="C. J." surname="Bernardos">
            <organization>Universidad Carlos III de Madrid</organization>
          </author>
          <author fullname="Diego Lopez" initials="D." surname="Lopez">
            <organization>Telefonica</organization>
          </author>
          <author fullname="Xavier de Foy" initials="X." surname="de Foy">
            <organization>InterDigital Communications, LLC</organization>
          </author>
          <date day="1" month="November" year="2020"/>
          <abstract>
            <t>   This document describes the problem of distributed data discovery in
   edge computing, and in particular for computing-in-the-network
   (COIN), which may require both the marshalling of data at the outset
   of a computation and the persistence of the resultant data after the
   computation.  Although the data might originate at the network edge,
   as more and more distributed data is created, processed, and stored,
   it becomes increasingly dispersed throughout the network.  There
   needs to be a standard way to find it.  New and existing protocols
   will need to be developed to support distributed data discovery at
   the network edge and beyond.

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-mcbride-edge-data-discovery-overview-05"/>
      </reference>
      <reference anchor="RFC6291">
        <front>
          <title>Guidelines for the Use of the "OAM" Acronym in the IETF</title>
          <author fullname="L. Andersson" initials="L." surname="Andersson"/>
          <author fullname="H. van Helvoort" initials="H." surname="van Helvoort"/>
          <author fullname="R. Bonica" initials="R." surname="Bonica"/>
          <author fullname="D. Romascanu" initials="D." surname="Romascanu"/>
          <author fullname="S. Mansfield" initials="S." surname="Mansfield"/>
          <date month="June" year="2011"/>
          <abstract>
            <t>At first glance, the acronym "OAM" seems to be well-known and well-understood. Looking at the acronym a bit more closely reveals a set of recurring problems that are revisited time and again.</t>
            <t>This document provides a definition of the acronym "OAM" (Operations, Administration, and Maintenance) for use in all future IETF documents that refer to OAM. There are other definitions and acronyms that will be discussed while exploring the definition of the constituent parts of the "OAM" term. This memo documents an Internet Best Current Practice.</t>
          </abstract>
        </front>
        <seriesInfo name="BCP" value="161"/>
        <seriesInfo name="RFC" value="6291"/>
        <seriesInfo name="DOI" value="10.17487/RFC6291"/>
      </reference>
      <reference anchor="RFC8578">
        <front>
          <title>Deterministic Networking Use Cases</title>
          <author fullname="E. Grossman" initials="E." role="editor" surname="Grossman"/>
          <date month="May" year="2019"/>
          <abstract>
            <t>This document presents use cases for diverse industries that have in common a need for "deterministic flows". "Deterministic" in this context means that such flows provide guaranteed bandwidth, bounded latency, and other properties germane to the transport of time-sensitive data. These use cases differ notably in their network topologies and specific desired behavior, providing as a group broad industry context for Deterministic Networking (DetNet). For each use case, this document will identify the use case, identify representative solutions used today, and describe potential improvements that DetNet can enable.</t>
          </abstract>
        </front>
        <seriesInfo name="RFC" value="8578"/>
        <seriesInfo name="DOI" value="10.17487/RFC8578"/>
      </reference>
      <reference anchor="I-D.irtf-t2trg-rest-iot">
        <front>
          <title>Guidance on RESTful Design for Internet of Things Systems</title>
          <author fullname="Ari Keränen" initials="A." surname="Keränen">
            <organization>Ericsson</organization>
          </author>
          <author fullname="Matthias Kovatsch" initials="M." surname="Kovatsch">
            <organization>Siemens</organization>
          </author>
          <author fullname="Klaus Hartke" initials="K." surname="Hartke">
         </author>
          <date day="25" month="July" year="2023"/>
          <abstract>
            <t>   This document gives guidance for designing Internet of Things (IoT)
   systems that follow the principles of the Representational State
   Transfer (REST) architectural style.  This document is a product of
   the IRTF Thing-to-Thing Research Group (T2TRG).

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-irtf-t2trg-rest-iot-12"/>
      </reference>
      <reference anchor="Ashton" target="http://www.itrco.jp/libraries/RFIDjournal-That%20Internet%20of%20Things%20Thing.pdf">
        <front>
          <title>That Internet of Things thing</title>
          <author initials="K." surname="Ashton">
            <organization/>
          </author>
          <date year="2009"/>
        </front>
        <refcontent>RFID J. vol. 22, no. 7, pp. 97-114</refcontent>
      </reference>
      <reference anchor="Lin">
        <front>
          <title>A Survey on Internet of Things: Architecture, Enabling Technologies, Security and Privacy, and Applications</title>
          <author initials="J." surname="Lin" fullname="Jie Lin">
            <organization/>
          </author>
          <author initials="W." surname="Yu" fullname="Wei Yu">
            <organization/>
          </author>
          <author initials="N." surname="Zhang" fullname="Nan Zhang">
            <organization/>
          </author>
          <author initials="X." surname="Yang" fullname="Xinyu Yang">
            <organization/>
          </author>
          <author initials="H." surname="Zhang" fullname="Hanlin Zhang">
            <organization/>
          </author>
          <author initials="W." surname="Zhao" fullname="Wei Zhao">
            <organization/>
          </author>
          <date year="2017" month="October"/>
        </front>
        <seriesInfo name="IEEE Internet of Things Journal" value="vol. 4, no. 5, pp. 1125-1142"/>
        <seriesInfo name="DOI" value="10.1109/jiot.2017.2683200"/>
      </reference>
      <reference anchor="NIST">
        <front>
          <title>The NIST definition of cloud computing</title>
          <author initials="P." surname="Mell" fullname="P M Mell">
            <organization/>
          </author>
          <author initials="T." surname="Grance" fullname="T Grance">
            <organization/>
          </author>
          <date year="2011"/>
        </front>
        <seriesInfo name="National Institute of Standards and Technology" value="report"/>
        <seriesInfo name="DOI" value="10.6028/nist.sp.800-145"/>
      </reference>
      <reference anchor="Botta">
        <front>
          <title>Integration of Cloud computing and Internet of Things: A survey</title>
          <author initials="A." surname="Botta" fullname="Alessio Botta">
            <organization/>
          </author>
          <author initials="W." surname="de Donato" fullname="Walter de Donato">
            <organization/>
          </author>
          <author initials="V." surname="Persico" fullname="Valerio Persico">
            <organization/>
          </author>
          <author initials="A." surname="Pescape" fullname="Antonio Pescape">
            <organization/>
          </author>
          <date year="2016" month="March"/>
        </front>
        <seriesInfo name="Future Generation Computer Systems" value="vol. 56, pp. 684-700"/>
        <seriesInfo name="DOI" value="10.1016/j.future.2015.09.021"/>
      </reference>
      <reference anchor="Shi">
        <front>
          <title>Edge Computing: Vision and Challenges</title>
          <author initials="W." surname="Shi" fullname="Weisong Shi">
            <organization/>
          </author>
          <author initials="J." surname="Cao" fullname="Jie Cao">
            <organization/>
          </author>
          <author initials="Q." surname="Zhang" fullname="Quan Zhang">
            <organization/>
          </author>
          <author initials="Y." surname="Li" fullname="Youhuizi Li">
            <organization/>
          </author>
          <author initials="L." surname="Xu" fullname="Lanyu Xu">
            <organization/>
          </author>
          <date year="2016" month="October"/>
        </front>
        <seriesInfo name="IEEE Internet of Things Journal" value="vol. 3, no. 5, pp. 637-646"/>
        <seriesInfo name="DOI" value="10.1109/jiot.2016.2579198"/>
      </reference>
      <reference anchor="Mahadev">
        <front>
          <title>The Emergence of Edge Computing</title>
          <author initials="M." surname="Satyanarayanan" fullname="Mahadev Satyanarayanan">
            <organization/>
          </author>
          <date year="2017" month="January"/>
        </front>
        <seriesInfo name="Computer" value="vol. 50, no. 1, pp. 30-39"/>
        <seriesInfo name="DOI" value="10.1109/mc.2017.9"/>
      </reference>
      <reference anchor="Chiang">
        <front>
          <title>Fog and IoT: An Overview of Research Opportunities</title>
          <author initials="M." surname="Chiang" fullname="Mung Chiang">
            <organization/>
          </author>
          <author initials="T." surname="Zhang" fullname="Tao Zhang">
            <organization/>
          </author>
          <date year="2016" month="December"/>
        </front>
        <seriesInfo name="IEEE Internet of Things Journal" value="vol. 3, no. 6, pp. 854-864"/>
        <seriesInfo name="DOI" value="10.1109/jiot.2016.2584538"/>
      </reference>
      <reference anchor="Weiner">
        <front>
          <title>Design of a low-latency, high-reliability wireless communication system for control applications</title>
          <author initials="M." surname="Weiner" fullname="Matthew Weiner">
            <organization/>
          </author>
          <author initials="M." surname="Jorgovanovic" fullname="Milos Jorgovanovic">
            <organization/>
          </author>
          <author initials="A." surname="Sahai" fullname="Anant Sahai">
            <organization/>
          </author>
          <author initials="B." surname="Nikolie" fullname="Borivoje Nikolie">
            <organization/>
          </author>
          <date year="2014" month="June"/>
        </front>
        <seriesInfo name="2014 IEEE International Conference on Communications" value="(ICC)"/>
        <seriesInfo name="DOI" value="10.1109/icc.2014.6883918"/>
      </reference>
      <reference anchor="Kelly" target="https://campustechnology.com/articles/2015/04/15/internet-of-things-data-to-top-1-6-zettabytes-by-2020.aspx">
        <front>
          <title>Internet of Things Data to Top 1.6 Zettabytes by 2022</title>
          <author initials="R." surname="Kelly">
            <organization/>
          </author>
          <date year="2015"/>
        </front>
        <refcontent>Retrieved on 2022-05-24</refcontent>
      </reference>
      <reference anchor="ISO_TR" target="https://www.iso.org/standard/53284.html">
        <front>
          <title>Internet of things (IoT) - Edge computing</title>
          <author>
            <organization/>
          </author>
          <date year="2020"/>
        </front>
        <seriesInfo name="ISO/IEC" value="TR 30164"/>
      </reference>
      <reference anchor="OpenFog" target="https://iiconsortium.org/pdf/OpenFog_Reference_Architecture_2_09_17.pdf">
        <front>
          <title>OpenFog Reference Architecture for Fog Computing</title>
          <author>
            <organization/>
          </author>
          <date year="2017"/>
        </front>
        <refcontent>OpenFog Consortium</refcontent>
      </reference>
      <reference anchor="ETSI_MEC_03" target="https://www.etsi.org/deliver/etsi_gs/MEC/001_099/003/02.01.01_60/gs_MEC003v020101p.pdf">
        <front>
          <title>Mobile Edge Computing (MEC); Framework and Reference Architecture</title>
          <author>
            <organization>ETSI</organization>
          </author>
          <date year="2019"/>
        </front>
        <refcontent>ETSI GS MEC 003</refcontent>
      </reference>
      <reference anchor="ETSI_MEC_01" target="https://www.etsi.org/deliver/etsi_gs/MEC/001_099/001/02.01.01_60/gs_MEC001v020101p.pdf">
        <front>
          <title>Multi-access Edge Computing (MEC); Terminology</title>
          <author initials="" surname="ETSI">
            <organization/>
          </author>
          <date year="2019"/>
        </front>
        <refcontent>ETSI GS MEC 001</refcontent>
      </reference>
      <reference anchor="NVIDIA" target="https://devblogs.nvidia.com/training-self-driving-vehicles-challenge-scale/">
        <front>
          <title>Training AI for Self-Driving Vehicles: the Challenge of Scale</title>
          <author initials="A." surname="Grzywaczewski">
            <organization/>
          </author>
          <date year="2017"/>
        </front>
        <refcontent>NVIDIA Developer Blog, retrieved on 2022-05-24</refcontent>
      </reference>
      <reference anchor="_60802" target="https://grouper.ieee.org/groups/802/1/files/public/docs2018/60802-industrial-use-cases-0918-v13.pdf">
        <front>
          <title>Use Cases IEC/IEEE 60802 V1.3</title>
          <author>
            <organization>IEC/IEEE</organization>
          </author>
          <date year="2018"/>
        </front>
        <seriesInfo name="IEC/IEEE" value="60802"/>
      </reference>
      <reference anchor="ENERGY">
        <front>
          <title>Revealing household characteristics from smart meter data</title>
          <author initials="C." surname="Beckel" fullname="Christian Beckel">
            <organization/>
          </author>
          <author initials="L." surname="Sadamori" fullname="Leyna Sadamori">
            <organization/>
          </author>
          <author initials="T." surname="Staake" fullname="Thorsten Staake">
            <organization/>
          </author>
          <author initials="S." surname="Santini" fullname="Silvia Santini">
            <organization/>
          </author>
          <date year="2014" month="December"/>
        </front>
        <seriesInfo name="Energy" value="vol. 78, pp. 397-410"/>
        <seriesInfo name="DOI" value="10.1016/j.energy.2014.10.025"/>
      </reference>
      <reference anchor="ETSI_Sandbox" target="https://portal.etsi.org/webapp/WorkProgram/Report_WorkItem.asp?WKI_ID=57671">
        <front>
          <title>Multi-access Edge Computing (MEC) MEC Sandbox Work Item</title>
          <author>
            <organization/>
          </author>
          <date year="2020"/>
        </front>
        <refcontent>Portal</refcontent>
      </reference>
      <reference anchor="AdvantEDGE" target="https://github.com/InterDigitalInc/AdvantEDGE">
        <front>
          <title>Mobile Edge Emulation Platform</title>
          <author>
            <organization/>
          </author>
          <date year="2020"/>
        </front>
        <refcontent>Source Code Repository</refcontent>
      </reference>
      <reference anchor="SimulatingFog">
        <front>
          <title>Simulating Fog and Edge Computing Scenarios: An Overview and Research Challenges</title>
          <author initials="S." surname="Svorobej" fullname="Sergej Svorobej">
            <organization/>
          </author>
          <author initials="P." surname="Takako Endo" fullname="Patricia Takako Endo">
            <organization/>
          </author>
          <author initials="M." surname="Bendechache" fullname="Malika Bendechache">
            <organization/>
          </author>
          <author initials="C." surname="Filelis-Papadopoulos" fullname="Christos Filelis-Papadopoulos">
            <organization/>
          </author>
          <author initials="K." surname="Giannoutakis" fullname="Konstantinos Giannoutakis">
            <organization/>
          </author>
          <author initials="G." surname="Gravvanis" fullname="George Gravvanis">
            <organization/>
          </author>
          <author initials="D." surname="Tzovaras" fullname="Dimitrios Tzovaras">
            <organization/>
          </author>
          <author initials="J." surname="Byrne" fullname="James Byrne">
            <organization/>
          </author>
          <author initials="T." surname="Lynn" fullname="Theo Lynn">
            <organization/>
          </author>
          <date year="2019" month="February"/>
        </front>
        <seriesInfo name="Future Internet" value="vol. 11, no. 3, pp. 55"/>
        <seriesInfo name="DOI" value="10.3390/fi11030055"/>
      </reference>
      <reference anchor="PseudoDynamicTesting">
        <front>
          <title>Pseudo-Dynamic Testing of Realistic Edge-Fog Cloud Ecosystems</title>
          <author initials="M." surname="Ficco" fullname="Massimo Ficco">
            <organization/>
          </author>
          <author initials="C." surname="Esposito" fullname="Christian Esposito">
            <organization/>
          </author>
          <author initials="Y." surname="Xiang" fullname="Yang Xiang">
            <organization/>
          </author>
          <author initials="F." surname="Palmieri" fullname="Francesco Palmieri">
            <organization/>
          </author>
          <date year="2017" month="November"/>
        </front>
        <seriesInfo name="IEEE Communications Magazine" value="vol. 55, no. 11, pp. 98-104"/>
        <seriesInfo name="DOI" value="10.1109/mcom.2017.1700328"/>
      </reference>
      <reference anchor="LFEDGE-EVE" target="https://www.lfedge.org/projects/eve">
        <front>
          <title>Project Edge Virtualization Engine (EVE)</title>
          <author initials="" surname="Linux Foundation">
            <organization/>
          </author>
          <date year="2020"/>
        </front>
        <refcontent>Portal, retrieved on 2022-05-24</refcontent>
      </reference>
      <reference anchor="Yousefpour">
        <front>
          <title>All one needs to know about fog computing and related edge computing paradigms: A complete survey</title>
          <author initials="A." surname="Yousefpour" fullname="Ashkan Yousefpour">
            <organization/>
          </author>
          <author initials="C." surname="Fung" fullname="Caleb Fung">
            <organization/>
          </author>
          <author initials="T." surname="Nguyen" fullname="Tam Nguyen">
            <organization/>
          </author>
          <author initials="K." surname="Kadiyala" fullname="Krishna Kadiyala">
            <organization/>
          </author>
          <author initials="F." surname="Jalali" fullname="Fatemeh Jalali">
            <organization/>
          </author>
          <author initials="A." surname="Niakanlahiji" fullname="Amirreza Niakanlahiji">
            <organization/>
          </author>
          <author initials="J." surname="Kong" fullname="Jian Kong">
            <organization/>
          </author>
          <author initials="J." surname="Jue" fullname="Jason P. Jue">
            <organization/>
          </author>
          <date year="2019" month="September"/>
        </front>
        <seriesInfo name="Journal of Systems Architecture" value="vol. 98, pp. 289-330"/>
        <seriesInfo name="DOI" value="10.1016/j.sysarc.2019.02.009"/>
      </reference>
      <reference anchor="I-D.bernardos-sfc-fog-ran">
        <front>
          <title>Service Function Chaining Use Cases in Fog RAN</title>
          <author fullname="Carlos J. Bernardos" initials="C. J." surname="Bernardos">
            <organization>UC3M</organization>
          </author>
          <author fullname="Alain Mourad" initials="A." surname="Mourad">
            <organization>InterDigital</organization>
          </author>
          <date day="22" month="October" year="2021"/>
          <abstract>
            <t>   Fog Radio Access Networks (RAN) refers to the part of the RAN that is
   virtualized at the very edge of the network, even at the end-user
   device.  Fog RAN support is considered critical for the 5G mobile
   network architectures currently being developed in various research,
   standardization and industry forums.  Since fog RAN builds on top of
   virtualization and can involve several virtual functions running on
   different virtualized resources, Service function chaining (SFC)
   support for the fog RAN will be critical.  This document describes
   the overall fog RAN approach and also gives some use cases.  Finally
   it proposes some requirements to be considered in the development of
   the SFC architecture and related protocols.

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-bernardos-sfc-fog-ran-10"/>
      </reference>
      <reference anchor="Schafer">
        <front>
          <title>Tasklets: Overcoming Heterogeneity in Distributed Computing Systems</title>
          <author initials="D." surname="Schafer" fullname="Dominik Schafer">
            <organization/>
          </author>
          <author initials="J." surname="Edinger" fullname="Janick Edinger">
            <organization/>
          </author>
          <author initials="S." surname="VanSyckel" fullname="Sebastian VanSyckel">
            <organization/>
          </author>
          <author initials="J." surname="Paluska" fullname="Justin Mazzola Paluska">
            <organization/>
          </author>
          <author initials="C." surname="Becker" fullname="Christian Becker">
            <organization/>
          </author>
          <date year="2016" month="June"/>
        </front>
        <seriesInfo name="2016 IEEE 36th International Conference on Distributed Computing Systems Workshops" value="(ICDCSW)"/>
        <seriesInfo name="DOI" value="10.1109/icdcsw.2016.22"/>
      </reference>
      <reference anchor="Echeverria">
        <front>
          <title>Establishing Trusted Identities in Disconnected Edge Environments</title>
          <author initials="S." surname="Echeverria" fullname="Sebastian Echeverria">
            <organization/>
          </author>
          <author initials="D." surname="Klinedinst" fullname="Dan Klinedinst">
            <organization/>
          </author>
          <author initials="K." surname="Williams" fullname="Keegan Williams">
            <organization/>
          </author>
          <author initials="G." surname="Lewis" fullname="Grace A. Lewis">
            <organization/>
          </author>
          <date year="2016" month="October"/>
        </front>
        <seriesInfo name="2016 IEEE/ACM Symposium on Edge Computing" value="(SEC)"/>
        <seriesInfo name="DOI" value="10.1109/sec.2016.27"/>
      </reference>
      <reference anchor="Harchol">
        <front>
          <title>CESSNA: Resilient Edge-Computing</title>
          <author initials="Y." surname="Harchol" fullname="Yotam Harchol">
            <organization/>
          </author>
          <author initials="A." surname="Mushtaq" fullname="Aisha Mushtaq">
            <organization/>
          </author>
          <author initials="J." surname="McCauley" fullname="James McCauley">
            <organization/>
          </author>
          <author initials="A." surname="Panda" fullname="Aurojit Panda">
            <organization/>
          </author>
          <author initials="S." surname="Shenker" fullname="Scott Shenker">
            <organization/>
          </author>
          <date year="2018" month="August"/>
        </front>
        <seriesInfo name="Proceedings of the 2018 Workshop on Mobile Edge" value="Communications"/>
        <seriesInfo name="DOI" value="10.1145/3229556.3229558"/>
      </reference>
      <reference anchor="Anglano">
        <front>
          <title>A game-theoretic approach to coalition formation in fog provider federations</title>
          <author initials="C." surname="Anglano" fullname="Cosimo Anglano">
            <organization/>
          </author>
          <author initials="M." surname="Canonico" fullname="Massimo Canonico">
            <organization/>
          </author>
          <author initials="P." surname="Castagno" fullname="Paolo Castagno">
            <organization/>
          </author>
          <author initials="M." surname="Guazzone" fullname="Marco Guazzone">
            <organization/>
          </author>
          <author initials="M." surname="Sereno" fullname="Matteo Sereno">
            <organization/>
          </author>
          <date year="2018" month="April"/>
        </front>
        <seriesInfo name="2018 Third International Conference on Fog and Mobile Edge Computing" value="(FMEC)"/>
        <seriesInfo name="DOI" value="10.1109/fmec.2018.8364054"/>
      </reference>
      <reference anchor="Jeong">
        <front>
          <title>Towards a distributed computing framework for Fog</title>
          <author initials="T." surname="Jeong" fullname="Taeyeol Jeong">
            <organization/>
          </author>
          <author initials="J." surname="Chung" fullname="Jaeyoon Chung">
            <organization/>
          </author>
          <author initials="J." surname="Hong" fullname="James Won-Ki Hong">
            <organization/>
          </author>
          <author initials="S." surname="Ha" fullname="Sangtae Ha">
            <organization/>
          </author>
          <date year="2017" month="October"/>
        </front>
        <seriesInfo name="2017 IEEE Fog World Congress" value="(FWC)"/>
        <seriesInfo name="DOI" value="10.1109/fwc.2017.8368528"/>
      </reference>
      <reference anchor="Fan">
        <front>
          <title>Cost Aware cloudlet Placement for big data processing at the edge</title>
          <author initials="Q." surname="Fan" fullname="Qiang Fan">
            <organization/>
          </author>
          <author initials="N." surname="Ansari" fullname="Nirwan Ansari">
            <organization/>
          </author>
          <date year="2017" month="May"/>
        </front>
        <seriesInfo name="2017 IEEE International Conference on Communications" value="(ICC)"/>
        <seriesInfo name="DOI" value="10.1109/icc.2017.7996722"/>
      </reference>
      <reference anchor="Yangui">
        <front>
          <title>A platform as-a-service for hybrid cloud/fog environments</title>
          <author initials="S." surname="Yangui" fullname="Sami Yangui">
            <organization/>
          </author>
          <author initials="P." surname="Ravindran" fullname="Pradeep Ravindran">
            <organization/>
          </author>
          <author initials="O." surname="Bibani" fullname="Ons Bibani">
            <organization/>
          </author>
          <author initials="R." surname="Glitho" fullname="Roch H. Glitho">
            <organization/>
          </author>
          <author initials="N." surname="Ben Hadj-Alouane" fullname="Nejib Ben Hadj-Alouane">
            <organization/>
          </author>
          <author initials="M." surname="Morrow" fullname="Monique J. Morrow">
            <organization/>
          </author>
          <author initials="P." surname="Polakos" fullname="Paul A. Polakos">
            <organization/>
          </author>
          <date year="2016" month="June"/>
        </front>
        <seriesInfo name="2016 IEEE International Symposium on Local and Metropolitan Area Networks" value="(LANMAN)"/>
        <seriesInfo name="DOI" value="10.1109/lanman.2016.7548853"/>
      </reference>
      <reference anchor="Li">
        <front>
          <title>MobiQoR: Pushing the Envelope of Mobile Edge Computing Via Quality-of-Result Optimization</title>
          <author initials="Y." surname="Li" fullname="Yongbo Li">
            <organization/>
          </author>
          <author initials="Y." surname="Chen" fullname="Yurong Chen">
            <organization/>
          </author>
          <author initials="T." surname="Lan" fullname="Tian Lan">
            <organization/>
          </author>
          <author initials="G." surname="Venkataramani" fullname="Guru Venkataramani">
            <organization/>
          </author>
          <date year="2017" month="June"/>
        </front>
        <seriesInfo name="2017 IEEE 37th International Conference on Distributed Computing Systems" value="(ICDCS)"/>
        <seriesInfo name="DOI" value="10.1109/icdcs.2017.54"/>
      </reference>
      <reference anchor="Liu">
        <front>
          <title>Cache Placement in Fog-RANs: From Centralized to Distributed Algorithms</title>
          <author initials="J." surname="Liu" fullname="Juan Liu">
            <organization/>
          </author>
          <author initials="B." surname="Bai" fullname="Bo Bai">
            <organization/>
          </author>
          <author initials="J." surname="Zhang" fullname="Jun Zhang">
            <organization/>
          </author>
          <author initials="K." surname="Letaief" fullname="Khaled B. Letaief">
            <organization/>
          </author>
          <date year="2017" month="November"/>
        </front>
        <seriesInfo name="IEEE Transactions on Wireless Communications" value="vol. 16, no. 11, pp. 7039-7051"/>
        <seriesInfo name="DOI" value="10.1109/twc.2017.2737015"/>
      </reference>
      <reference anchor="Zhang">
        <front>
          <title>Firework: Big Data Sharing and Processing in Collaborative Edge Environment</title>
          <author initials="Q." surname="Zhang" fullname="Quan Zhang">
            <organization/>
          </author>
          <author initials="X." surname="Zhang" fullname="Xiaohong Zhang">
            <organization/>
          </author>
          <author initials="Q." surname="Zhang" fullname="Qingyang Zhang">
            <organization/>
          </author>
          <author initials="W." surname="Shi" fullname="Weisong Shi">
            <organization/>
          </author>
          <author initials="H." surname="Zhong" fullname="Hong Zhong">
            <organization/>
          </author>
          <date year="2016" month="October"/>
        </front>
        <seriesInfo name="2016 Fourth IEEE Workshop on Hot Topics in Web Systems and Technologies" value="(HotWeb)"/>
        <seriesInfo name="DOI" value="10.1109/hotweb.2016.12"/>
      </reference>
      <reference anchor="Basudan">
        <front>
          <title>A Privacy-Preserving Vehicular Crowdsensing-Based Road Surface Condition Monitoring System Using Fog Computing</title>
          <author initials="S." surname="Basudan" fullname="Sultan Basudan">
            <organization/>
          </author>
          <author initials="X." surname="Lin" fullname="Xiaodong Lin">
            <organization/>
          </author>
          <author initials="K." surname="Sankaranarayanan" fullname="Karthik Sankaranarayanan">
            <organization/>
          </author>
          <date year="2017" month="June"/>
        </front>
        <seriesInfo name="IEEE Internet of Things Journal" value="vol. 4, no. 3, pp. 772-782"/>
        <seriesInfo name="DOI" value="10.1109/jiot.2017.2666783"/>
      </reference>
      <reference anchor="Renart">
        <front>
          <title>Data-Driven Stream Processing at the Edge</title>
          <author initials="E." surname="Renart" fullname="Eduard Gibert Renart">
            <organization/>
          </author>
          <author initials="J." surname="Diaz-Montes" fullname="Javier Diaz-Montes">
            <organization/>
          </author>
          <author initials="M." surname="Parashar" fullname="Manish Parashar">
            <organization/>
          </author>
          <date year="2017" month="May"/>
        </front>
        <seriesInfo name="2017 IEEE 1st International Conference on Fog and Edge Computing" value="(ICFEC)"/>
        <seriesInfo name="DOI" value="10.1109/icfec.2017.18"/>
      </reference>
      <reference anchor="Zhang2">
        <front>
          <title>Data Security and Privacy-Preserving in Edge Computing Paradigm: Survey and Open Issues</title>
          <author initials="J." surname="Zhang" fullname="Jiale Zhang">
            <organization/>
          </author>
          <author initials="B." surname="Chen" fullname="Bing Chen">
            <organization/>
          </author>
          <author initials="Y." surname="Zhao" fullname="Yanchao Zhao">
            <organization/>
          </author>
          <author initials="X." surname="Cheng" fullname="Xiang Cheng">
            <organization/>
          </author>
          <author initials="F." surname="Hu" fullname="Feng Hu">
            <organization/>
          </author>
          <date year="2018"/>
        </front>
        <seriesInfo name="IEEE Access" value="vol. 6, pp. 18209-18237"/>
        <seriesInfo name="DOI" value="10.1109/access.2018.2820162"/>
      </reference>
      <reference anchor="Yates">
        <front>
          <title>The Age of Information: Real-Time Status Updating by Multiple Sources</title>
          <author initials="R." surname="Yates" fullname="Roy D. Yates">
            <organization/>
          </author>
          <author initials="S." surname="Kaul" fullname="Sanjit K. Kaul">
            <organization/>
          </author>
          <date year="2019" month="March"/>
        </front>
        <seriesInfo name="IEEE Transactions on Information Theory" value="vol. 65, no. 3, pp. 1807-1827"/>
        <seriesInfo name="DOI" value="10.1109/tit.2018.2871079"/>
      </reference>
      <reference anchor="Khan">
        <front>
          <title>Edge-Computing-Enabled Smart Cities: A Comprehensive Survey</title>
          <author initials="L." surname="Khan" fullname="Latif U. Khan">
            <organization/>
          </author>
          <author initials="I." surname="Yaqoob" fullname="Ibrar Yaqoob">
            <organization/>
          </author>
          <author initials="N." surname="Tran" fullname="Nguyen H. Tran">
            <organization/>
          </author>
          <author initials="S." surname="Kazmi" fullname="S. M. Ahsan Kazmi">
            <organization/>
          </author>
          <author initials="T." surname="Dang" fullname="Tri Nguyen Dang">
            <organization/>
          </author>
          <author initials="C." surname="Hong" fullname="Choong Seon Hong">
            <organization/>
          </author>
          <date year="2020" month="October"/>
        </front>
        <seriesInfo name="IEEE Internet of Things Journal" value="vol. 7, no. 10, pp. 10200-10232"/>
        <seriesInfo name="DOI" value="10.1109/jiot.2020.2987070"/>
      </reference>
      <reference anchor="RFC7390">
        <front>
          <title>Group Communication for the Constrained Application Protocol (CoAP)</title>
          <author fullname="A. Rahman" initials="A." role="editor" surname="Rahman"/>
          <author fullname="E. Dijk" initials="E." role="editor" surname="Dijk"/>
          <date month="October" year="2014"/>
          <abstract>
            <t>The Constrained Application Protocol (CoAP) is a specialized web transfer protocol for constrained devices and constrained networks. It is anticipated that constrained devices will often naturally operate in groups (e.g., in a building automation scenario, all lights in a given room may need to be switched on/off as a group). This specification defines how CoAP should be used in a group communication context. An approach for using CoAP on top of IP multicast is detailed based on existing CoAP functionality as well as new features introduced in this specification. Also, various use cases and corresponding protocol flows are provided to illustrate important concepts. Finally, guidance is provided for deployment in various network topologies.</t>
          </abstract>
        </front>
        <seriesInfo name="RFC" value="7390"/>
        <seriesInfo name="DOI" value="10.17487/RFC7390"/>
      </reference>
      <reference anchor="I-D.ietf-core-groupcomm-bis">
        <front>
          <title>Group Communication for the Constrained Application Protocol (CoAP)</title>
          <author fullname="Esko Dijk" initials="E." surname="Dijk">
            <organization>IoTconsultancy.nl</organization>
          </author>
          <author fullname="Chonggang Wang" initials="C." surname="Wang">
            <organization>InterDigital</organization>
          </author>
          <author fullname="Marco Tiloca" initials="M." surname="Tiloca">
            <organization>RISE AB</organization>
          </author>
          <date day="10" month="July" year="2023"/>
          <abstract>
            <t>   This document specifies the use of the Constrained Application
   Protocol (CoAP) for group communication, including the use of UDP/IP
   multicast as the default underlying data transport.  Both unsecured
   and secured CoAP group communication are specified.  Security is
   achieved by use of the Group Object Security for Constrained RESTful
   Environments (Group OSCORE) protocol.  The target application area of
   this specification is any group communication use cases that involve
   resource-constrained devices or networks that support CoAP.  This
   document replaces RFC 7390, while it updates RFC 7252 and RFC 7641.

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-ietf-core-groupcomm-bis-09"/>
      </reference>
      <reference anchor="Murshed">
        <front>
          <title>Machine Learning at the Network Edge: A Survey</title>
          <author initials="M." surname="Murshed" fullname="M. G. Sarwar Murshed">
            <organization/>
          </author>
          <author initials="C." surname="Murphy" fullname="Christopher Murphy">
            <organization/>
          </author>
          <author initials="D." surname="Hou" fullname="Daqing Hou">
            <organization/>
          </author>
          <author initials="N." surname="Khan" fullname="Nazar Khan">
            <organization/>
          </author>
          <author initials="G." surname="Ananthanarayanan" fullname="Ganesh Ananthanarayanan">
            <organization/>
          </author>
          <author initials="F." surname="Hussain" fullname="Faraz Hussain">
            <organization/>
          </author>
          <date year="2022" month="November"/>
        </front>
        <seriesInfo name="ACM Computing Surveys" value="vol. 54, no. 8, pp. 1-37"/>
        <seriesInfo name="DOI" value="10.1145/3469029"/>
      </reference>
      <reference anchor="I-D.sarathchandra-coin-appcentres">
        <front>
          <title>In-Network Computing for App-Centric Micro-Services</title>
          <author fullname="Dirk Trossen" initials="D." surname="Trossen">
            <organization>Huawei</organization>
          </author>
          <author fullname="Chathura Sarathchandra" initials="C." surname="Sarathchandra">
            <organization>InterDigital Inc.</organization>
          </author>
          <author fullname="Michael Boniface" initials="M." surname="Boniface">
            <organization>University of Southampton</organization>
          </author>
          <date day="26" month="January" year="2021"/>
          <abstract>
            <t>   The application-centric deployment of 'Internet' services has
   increased over the past ten years with many millions of applications
   providing user-centric services, executed on increasingly more
   powerful smartphones that are supported by Internet-based cloud
   services in distributed data centres, the latter mainly provided by
   large scale players such as Google, Amazon and alike. This draft
   outlines a vision for evolving those data centres towards executing
   app-centric micro-services; we dub this evolved data centre as an
   AppCentre. Complemented with the proliferation of such AppCentres at
   the edge of the network, they will allow for such micro-services to
   be distributed across many places of execution, including mobile
   terminals themselves, while specific micro-service chains equal
   today's applications in existing smartphones.

   We outline the key enabling technologies that needs to be provided
   for such evolution to be realized, including references to ongoing
   standardization efforts in key areas.


            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-sarathchandra-coin-appcentres-04"/>
      </reference>
      <reference anchor="I-D.defoy-t2trg-iot-edge-computing-background">
        <front>
          <title>IoT Edge Computing: Initiatives, Projects and Products</title>
          <author fullname="Xavier de Foy" initials="X." surname="de Foy">
            <organization>InterDigital Communications</organization>
          </author>
          <author fullname="Jungha Hong" initials="J." surname="Hong">
            <organization>ETRI</organization>
          </author>
          <author fullname="Yong-Geun Hong" initials="Y." surname="Hong">
            <organization>ETRI</organization>
          </author>
          <author fullname="Matthias Kovatsch" initials="M." surname="Kovatsch">
            <organization>Huawei Technologies Duesseldorf GmbH</organization>
          </author>
          <author fullname="Eve Schooler" initials="E." surname="Schooler">
            <organization>Intel</organization>
          </author>
          <author fullname="Dirk Kutscher" initials="D." surname="Kutscher">
            <organization>University of Applied Sciences Emden/Leer</organization>
          </author>
          <date day="25" month="May" year="2020"/>
          <abstract>
            <t>   Many IoT applications have requirements that cannot be met by
   the traditional Cloud.  As a result, the IoT is driving the Internet
   toward Edge computing.  This draft reviews initiatives, projects and
   products related to IoT Edge Computing.

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-defoy-t2trg-iot-edge-computing-background-00"/>
      </reference>
      <reference anchor="Senel">
        <front>
          <title>EdgeNet: A Multi-Tenant and Multi-Provider Edge Cloud</title>
          <author initials="B." surname="Senel" fullname="Berat Can Senel">
            <organization/>
          </author>
          <author initials="M." surname="Mouchet" fullname="Maxime Mouchet">
            <organization/>
          </author>
          <author initials="J." surname="Cappos" fullname="Justin Cappos">
            <organization/>
          </author>
          <author initials="O." surname="Fourmaux" fullname="Olivier Fourmaux">
            <organization/>
          </author>
          <author initials="T." surname="Friedman" fullname="Timur Friedman">
            <organization/>
          </author>
          <author initials="R." surname="McGeer" fullname="Rick McGeer">
            <organization/>
          </author>
          <date year="2021" month="April"/>
        </front>
        <seriesInfo name="Proceedings of the 4th International Workshop on Edge Systems, Analytics and" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3434770.3459737"/>
      </reference>
      <reference anchor="Stanciu">
        <front>
          <title>Privacy-Preserving Crowd-Monitoring Using Bloom Filters and Homomorphic Encryption</title>
          <author initials="V." surname="Stanciu" fullname="Valeriu-Daniel Stanciu">
            <organization/>
          </author>
          <author initials="M." surname="Steen" fullname="Maarten van Steen">
            <organization/>
          </author>
          <author initials="C." surname="Dobre" fullname="Ciprian Dobre">
            <organization/>
          </author>
          <author initials="A." surname="Peter" fullname="Andreas Peter">
            <organization/>
          </author>
          <date year="2021" month="April"/>
        </front>
        <seriesInfo name="Proceedings of the 4th International Workshop on Edge Systems, Analytics and" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3434770.3459735"/>
      </reference>
      <reference anchor="Jeffery">
        <front>
          <title>Rearchitecting Kubernetes for the Edge</title>
          <author initials="A." surname="Jeffery" fullname="Andrew Jeffery">
            <organization/>
          </author>
          <author initials="H." surname="Howard" fullname="Heidi Howard">
            <organization/>
          </author>
          <author initials="R." surname="Mortier" fullname="Richard Mortier">
            <organization/>
          </author>
          <date year="2021" month="April"/>
        </front>
        <seriesInfo name="Proceedings of the 4th International Workshop on Edge Systems, Analytics and" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3434770.3459730"/>
      </reference>
      <reference anchor="Nieke">
        <front>
          <title>Edgedancer: Secure Mobile WebAssembly Services on the Edge</title>
          <author initials="M." surname="Nieke" fullname="Manuel Nieke">
            <organization/>
          </author>
          <author initials="L." surname="Almstedt" fullname="Lennart Almstedt">
            <organization/>
          </author>
          <author initials="R." surname="Kapitza" fullname="Rudiger Kapitza">
            <organization/>
          </author>
          <date year="2021" month="April"/>
        </front>
        <seriesInfo name="Proceedings of the 4th International Workshop on Edge Systems, Analytics and" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3434770.3459731"/>
      </reference>
      <reference anchor="Cao">
        <front>
          <title>eCaaS: A Management Framework of Edge Container as a Service for Business Workload</title>
          <author initials="L." surname="Cao" fullname="Lianjie Cao">
            <organization/>
          </author>
          <author initials="A." surname="Merican" fullname="Anu Merican">
            <organization/>
          </author>
          <author initials="D." surname="Tootaghaj" fullname="Diman Zad Tootaghaj">
            <organization/>
          </author>
          <author initials="F." surname="Ahmed" fullname="Faraz Ahmed">
            <organization/>
          </author>
          <author initials="P." surname="Sharma" fullname="Puneet Sharma">
            <organization/>
          </author>
          <author initials="V." surname="Saxena" fullname="Vinay Saxena">
            <organization/>
          </author>
          <date year="2021" month="April"/>
        </front>
        <seriesInfo name="Proceedings of the 4th International Workshop on Edge Systems, Analytics and" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3434770.3459741"/>
      </reference>
      <reference anchor="Larrea">
        <front>
          <title>The serverkernel operating system</title>
          <author initials="J." surname="Larrea" fullname="Jon Larrea">
            <organization/>
          </author>
          <author initials="A." surname="Barbalace" fullname="Antonio Barbalace">
            <organization/>
          </author>
          <date year="2020" month="April"/>
        </front>
        <seriesInfo name="Proceedings of the Third ACM International Workshop on Edge Systems, Analytics and" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3378679.3394537"/>
      </reference>
      <reference anchor="Mortazavi" target="https://www.usenix.org/conference/hotedge18/presentation/mortazavi">
        <front>
          <title>Toward Session Consistency for the Edge</title>
          <author initials="S." surname="Hossein Mortazavi">
            <organization/>
          </author>
          <author initials="B." surname="Balasubramanian">
            <organization/>
          </author>
          <author initials="E." surname="de Lara">
            <organization/>
          </author>
          <author initials="S. P." surname="Narayanan">
            <organization/>
          </author>
          <date year="2018"/>
        </front>
        <refcontent>USENIX, Workshop on Hot Topics in Edge Computing (HotEdge 18)</refcontent>
      </reference>
      <reference anchor="oneM2M-TR0001" target="https://member.onem2m.org/Application/documentapp/downloadLatestRevision/default.aspx?docID=28153">
        <front>
          <title>TR 0001, Use Cases Collection</title>
          <author initials="C." surname="Mladin">
            <organization/>
          </author>
          <date year="2018" month="October"/>
        </front>
        <refcontent>oneM2M</refcontent>
      </reference>
      <reference anchor="oneM2M-TR0018" target="https://member.onem2m.org/Application/documentapp/downloadLatestRevision/default.aspx?docID=29334">
        <front>
          <title>TR 0018, Industrial Domain Enablement</title>
          <author initials="C." surname="Lu">
            <organization/>
          </author>
          <author initials="M." surname="Jiang">
            <organization/>
          </author>
          <date year="2019" month="February"/>
        </front>
        <refcontent>oneM2M</refcontent>
      </reference>
      <reference anchor="oneM2M-TR0026" target="https://member.onem2m.org/Application/documentapp/downloadLatestRevision/default.aspx?docID=31410">
        <front>
          <title>TR 0026, Vehicular Domain Enablement</title>
          <author initials="K." surname="Yamamoto">
            <organization/>
          </author>
          <author initials="C." surname="Mladin">
            <organization/>
          </author>
          <author initials="V." surname="Kueh">
            <organization/>
          </author>
          <date year="2020" month="January"/>
        </front>
        <refcontent>oneM2M</refcontent>
      </reference>
      <reference anchor="oneM2M-TR0052" target="https://member.onem2m.org/Application/documentapp/downloadLatestRevision/default.aspx?docID=32633">
        <front>
          <title>TR 0052, Study on Edge and Fog Computing in oneM2M systems</title>
          <author initials="K." surname="Yamamoto">
            <organization/>
          </author>
          <author initials="C." surname="Mladin">
            <organization/>
          </author>
          <date year="2020" month="September"/>
        </front>
        <refcontent>oneM2M</refcontent>
      </reference>
      <reference anchor="oneM2M-TS0002" target="https://member.onem2m.org/Application/documentapp/downloadLatestRevision/default.aspx?docID=29274">
        <front>
          <title>TS 0002, Requirements</title>
          <author initials="S." surname="He">
            <organization/>
          </author>
          <date year="2019" month="February"/>
        </front>
        <refcontent>oneM2M</refcontent>
      </reference>
      <reference anchor="Chen">
        <front>
          <title>Edge Computing in IoT-Based Manufacturing</title>
          <author initials="B." surname="Chen" fullname="Baotong Chen">
            <organization/>
          </author>
          <author initials="J." surname="Wan" fullname="Jiafu Wan">
            <organization/>
          </author>
          <author initials="A." surname="Celesti" fullname="Antonio Celesti">
            <organization/>
          </author>
          <author initials="D." surname="Li" fullname="Di Li">
            <organization/>
          </author>
          <author initials="H." surname="Abbas" fullname="Haider Abbas">
            <organization/>
          </author>
          <author initials="Q." surname="Zhang" fullname="Qin Zhang">
            <organization/>
          </author>
          <date year="2018" month="September"/>
        </front>
        <seriesInfo name="IEEE Communications Magazine" value="vol. 56, no. 9, pp. 103-109"/>
        <seriesInfo name="DOI" value="10.1109/mcom.2018.1701231"/>
      </reference>
      <reference anchor="Jones">
        <front>
          <title>Characterising the Digital Twin: A systematic literature review</title>
          <author initials="D." surname="Jones" fullname="David Jones">
            <organization/>
          </author>
          <author initials="C." surname="Snider" fullname="Chris Snider">
            <organization/>
          </author>
          <author initials="A." surname="Nassehi" fullname="Aydin Nassehi">
            <organization/>
          </author>
          <author initials="J." surname="Yon" fullname="Jason Yon">
            <organization/>
          </author>
          <author initials="B." surname="Hicks" fullname="Ben Hicks">
            <organization/>
          </author>
          <date year="2020" month="May"/>
        </front>
        <seriesInfo name="CIRP Journal of Manufacturing Science and Technology" value="vol. 29, pp. 36-52"/>
        <seriesInfo name="DOI" value="10.1016/j.cirpj.2020.02.002"/>
      </reference>
      <reference anchor="RFC7252">
        <front>
          <title>The Constrained Application Protocol (CoAP)</title>
          <author fullname="Z. Shelby" initials="Z." surname="Shelby"/>
          <author fullname="K. Hartke" initials="K." surname="Hartke"/>
          <author fullname="C. Bormann" initials="C." surname="Bormann"/>
          <date month="June" year="2014"/>
          <abstract>
            <t>The Constrained Application Protocol (CoAP) is a specialized web transfer protocol for use with constrained nodes and constrained (e.g., low-power, lossy) networks. The nodes often have 8-bit microcontrollers with small amounts of ROM and RAM, while constrained networks such as IPv6 over Low-Power Wireless Personal Area Networks (6LoWPANs) often have high packet error rates and a typical throughput of 10s of kbit/s. The protocol is designed for machine- to-machine (M2M) applications such as smart energy and building automation.</t>
            <t>CoAP provides a request/response interaction model between application endpoints, supports built-in discovery of services and resources, and includes key concepts of the Web such as URIs and Internet media types. CoAP is designed to easily interface with HTTP for integration with the Web while meeting specialized requirements such as multicast support, very low overhead, and simplicity for constrained environments.</t>
          </abstract>
        </front>
        <seriesInfo name="RFC" value="7252"/>
        <seriesInfo name="DOI" value="10.17487/RFC7252"/>
      </reference>
      <reference anchor="Cloudlets">
        <front>
          <title>The Case for VM-Based Cloudlets in Mobile Computing</title>
          <author initials="M." surname="Satyanarayanan" fullname="M. Satyanarayanan">
            <organization/>
          </author>
          <author initials="P." surname="Bahl" fullname="P. Bahl">
            <organization/>
          </author>
          <author initials="R." surname="Caceres" fullname="R. Caceres">
            <organization/>
          </author>
          <author initials="N." surname="Davies" fullname="N. Davies">
            <organization/>
          </author>
          <date year="2009" month="October"/>
        </front>
        <seriesInfo name="IEEE Pervasive Computing" value="vol. 8, no. 4, pp. 14-23"/>
        <seriesInfo name="DOI" value="10.1109/mprv.2009.82"/>
      </reference>
      <reference anchor="ETSI_MEC_33" target="https://www.etsi.org/deliver/etsi_gs/MEC/001_099/033/03.01.01_60/gs_MEC033v030101p.pdf">
        <front>
          <title>Multi-access Edge Computing (MEC); IoT API</title>
          <author>
            <organization>ETSI</organization>
          </author>
          <date year="2022"/>
        </front>
        <refcontent>ETSI GS MEC 033</refcontent>
      </reference>
      <reference anchor="Madni">
        <front>
          <title>Leveraging digital twin technology in model-based systems engineering</title>
          <author initials="A. M." surname="Madni" fullname="Azad M. Madni">
            <organization/>
          </author>
          <author initials="C." surname="Madni" fullname="Carla C. Madni">
            <organization/>
          </author>
          <author initials="S. D." surname="Lucero" fullname="Scott D. Lucero">
            <organization/>
          </author>
          <date year="2019"/>
        </front>
        <seriesInfo name="Systems 7, no. 1" value="7"/>
        <seriesInfo name="DOI" value="10.3390/systems7010007"/>
      </reference>
      <reference anchor="I-D.irtf-nmrg-network-digital-twin-arch">
        <front>
          <title>Digital Twin Network: Concepts and Reference Architecture</title>
          <author fullname="Cheng Zhou" initials="C." surname="Zhou">
            <organization>China Mobile</organization>
          </author>
          <author fullname="Hongwei Yang" initials="H." surname="Yang">
            <organization>China Mobile</organization>
          </author>
          <author fullname="Xiaodong Duan" initials="X." surname="Duan">
            <organization>China Mobile</organization>
          </author>
          <author fullname="Diego Lopez" initials="D." surname="Lopez">
            <organization>Telefonica I+D</organization>
          </author>
          <author fullname="Antonio Pastor" initials="A." surname="Pastor">
            <organization>Telefonica I+D</organization>
          </author>
          <author fullname="Qin Wu" initials="Q." surname="Wu">
            <organization>Huawei</organization>
          </author>
          <author fullname="Mohamed Boucadair" initials="M." surname="Boucadair">
            <organization>Orange</organization>
          </author>
          <author fullname="Christian Jacquenet" initials="C." surname="Jacquenet">
            <organization>Orange</organization>
          </author>
          <date day="27" month="April" year="2023"/>
          <abstract>
            <t>   Digital Twin technology has been seen as a rapid adoption technology
   in Industry 4.0.  The application of Digital Twin technology in the
   networking field is meant to develop various rich network
   applications and realize efficient and cost effective data driven
   network management and accelerate network innovation.

   This document presents an overview of the concepts of Digital Twin
   Network, provides the basic definitions and a reference architecture,
   lists a set of application scenarios, and discusses the benefits and
   key challenges of such technology.

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-irtf-nmrg-network-digital-twin-arch-03"/>
      </reference>
      <reference anchor="I-D.hsingh-coinrg-reqs-p4comp">
        <front>
          <title>Requirements for P4 Program Splitting for Heterogeneous Network Nodes</title>
          <author fullname="Hemant Singh" initials="H." surname="Singh">
            <organization>MNK Labs and Consulting</organization>
          </author>
          <author fullname="Marie-Jose Montpetit" initials="M." surname="Montpetit">
            <organization>Concordia Univeristy</organization>
          </author>
          <date day="18" month="February" year="2021"/>
          <abstract>
            <t>   For distributed computing, the P4 research community has published a
   paper to show how to split a P4 program into sub-programs which run
   on heterogeneous network nodes in a network.  Examples of nodes are a
   network switch, a smartNIC, or a host machine.  The paper has
   developed artifacts to split program based on latency, data rate,
   cost, etc.  However, the paper does not mention any requirements.  To
   provide guidance, this document covers requirements for splitting P4
   programs for heterogeneous network nodes.

            </t>
          </abstract>
        </front>
        <seriesInfo name="Internet-Draft" value="draft-hsingh-coinrg-reqs-p4comp-03"/>
      </reference>
      <reference anchor="Farnbauer-Schmidt">
        <front>
          <title>Combining the Concepts of Semantic Data Integration and Edge Computing</title>
          <author initials="M." surname="Farnbauer-Schmidt" fullname="Matthias Farnbauer-Schmidt">
            <organization/>
          </author>
          <author initials="J." surname="Lindner" fullname="Julian Lindner">
            <organization/>
          </author>
          <author initials="C." surname="Kaffenberger" fullname="Christopher Kaffenberger">
            <organization/>
          </author>
          <author initials="J." surname="Albrecht" fullname="Jens Albrecht">
            <organization/>
          </author>
          <date year="2019"/>
        </front>
        <seriesInfo name="INFORMATIK 2019" value="50 Jahre Gesellschaft fur Informatik - Informatik fur Gesellschaft, pp. 139-152"/>
        <seriesInfo name="DOI" value="10.18420/inf2019_19"/>
      </reference>
      <reference anchor="Brecko">
        <front>
          <title>Federated Learning for Edge Computing: A Survey</title>
          <author initials="A." surname="Brecko" fullname="Alexander Brecko">
            <organization/>
          </author>
          <author initials="E." surname="Kajati" fullname="Erik Kajati">
            <organization/>
          </author>
          <author initials="J." surname="Koziorek" fullname="Jiri Koziorek">
            <organization/>
          </author>
          <author initials="I." surname="Zolotova" fullname="Iveta Zolotova">
            <organization/>
          </author>
          <date year="2022"/>
        </front>
        <seriesInfo name="Applied Sciences 12, no. 18" value="9124"/>
        <seriesInfo name="DOI" value="10.3390/app12189124"/>
      </reference>
      <reference anchor="Grewe">
        <front>
          <title>Information-Centric Mobile Edge Computing for Connected Vehicle Environments: Challenges and Research Directions</title>
          <author initials="D." surname="Grewe" fullname="Dennis Grewe">
            <organization/>
          </author>
          <author initials="M." surname="Wagner" fullname="Marco Wagner">
            <organization/>
          </author>
          <author initials="M." surname="Arumaithurai" fullname="Mayutan Arumaithurai">
            <organization/>
          </author>
          <author initials="I." surname="Psaras" fullname="Ioannis Psaras">
            <organization/>
          </author>
          <author initials="D." surname="Kutscher" fullname="Dirk Kutscher">
            <organization/>
          </author>
          <date year="2017"/>
        </front>
        <seriesInfo name="Proceedings of the Workshop on Mobile Edge Communications" value="pp. 7-12"/>
        <seriesInfo name="DOI" value="10.1145/3098208.3098210"/>
      </reference>
      <reference anchor="Ali">
        <front>
          <title>Enabling a Blockchain-Based IoT Edge</title>
          <author initials="M. S." surname="Ali" fullname="Muhammad Salek Ali">
            <organization/>
          </author>
          <author initials="M." surname="Vecchio" fullname="Massimo Vecchio">
            <organization/>
          </author>
          <author initials="F." surname="Antonelli" fullname="Fabio Antonelli">
            <organization/>
          </author>
          <date year="2018"/>
        </front>
        <seriesInfo name="IEEE Internet of Things Magazine" value="pp. 24-29"/>
        <seriesInfo name="DOI" value="10.1109/IOTM.2019.1800024"/>
      </reference>
      <reference anchor="chipmunk">
        <front>
          <title>Chipmunk: Distributed Object Storage for NDN</title>
          <author fullname="Yong Yoon Shin" initials="Y." surname="Shin">
            <organization>ETRI, GurumNetworks, Republic of Korea</organization>
          </author>
          <author fullname="Sae Hyong Park" initials="S." surname="Park">
            <organization>ETRI, GurumNetworks, Republic of Korea</organization>
          </author>
          <author fullname="Namseok Ko" initials="N." surname="Ko">
            <organization>ETRI, GurumNetworks, Republic of Korea</organization>
          </author>
          <author fullname="Arm Jeong" initials="A." surname="Jeong">
            <organization>ETRI, GurumNetworks, Republic of Korea</organization>
          </author>
          <date month="September" year="2020"/>
        </front>
        <seriesInfo name="Proceedings of the 7th ACM Conference on Information-Centric" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3405656.3420231"/>
        <refcontent>ACM</refcontent>
      </reference>
      <reference anchor="kua">
        <front>
          <title>Kua: a distributed object store over named data networking</title>
          <author fullname="Varun Patil" initials="V." surname="Patil">
            <organization>UCLA</organization>
          </author>
          <author fullname="Hemil Desai" initials="H." surname="Desai">
            <organization>UCLA</organization>
          </author>
          <author fullname="Lixia Zhang" initials="L." surname="Zhang">
            <organization>UCLA</organization>
          </author>
          <date month="September" year="2022"/>
        </front>
        <seriesInfo name="Proceedings of the 9th ACM Conference on Information-Centric" value="Networking"/>
        <seriesInfo name="DOI" value="10.1145/3517212.3558083"/>
        <refcontent>ACM</refcontent>
      </reference>
      <reference anchor="mqtt5" target="https://docs.oasis-open.org/mqtt/mqtt/v5.0/mqtt-v5.0.html">
        <front>
          <title>MQTT Version 5.0</title>
          <author initials="" surname="OASIS Message Queuing Telemetry Transport (MQTT) TC">
            <organization/>
          </author>
          <date year="2019" month="March"/>
        </front>
        <refcontent>OASIS</refcontent>
      </reference>
    </references>
  </back>
  <!-- ##markdown-source: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-->

</rfc>