Merge pull request #207242 from Yvonne-dQ/patch-3

Create Azure Percept folders

Author: JamesJBarnett

articles/azure-percept/deepstream/azure-percept-for-deepstream-overview.md

@@ -0,0 +1,58 @@
+---
+title: Azure Percept for DeepStream overview
+description: A description of Azure Percept for DeepStream developer tools that provide a custom developer experience. 
+author: MaxStrange
+ms.author: strangem
+manager: amiyouss
+ms.service: azure-percept
+ms.topic: overview
+ms.date: 08/10/2022
+---
+
+# Azure Percept for DeepStream overview
+
+Azure Percept for DeepStream includes developer tools that provide a custom developer experience. It enables you to create NVIDIA DeepStream containers using Microsoft-based images and guidance, supported models from NVIDIA out of the box, and/or bring your own models. 
+
+DeepStream is NVIDIA’s toolkit to develop and deploy Vision AI applications and services. It provides multi-platform, scalable, Transport Layer Security (TLS)-encrypted security that can be deployed on-premises, on the edge, and in the cloud. 
+
+## Azure Percept for DeepStream offers:
+
+- **Simplifying your development process** 
+
+  Auto selection of AI model execution and inference provider: One of several execution providers, such as ORT, CUDA, and TENSORT, are automatically selected to simplify your development process.
+
+- **Customizing Region of Interest (ROI) to enable your business scenario**
+
+  Region of Interest (ROI) configuration widget: Percept Player, a web app widget, is included for customizing ROIs to enable event detection for your business scenario.
+
+- **Simplifying the configuration for pre/post processing** 
+
+  You can add a Python-based model and parser using a configuration file, instead of hardcoding it into the pipeline.
+
+- **Offering a broad Pre-built AI model framework** 
+
+  This solution supports many of the most common CV models in use today, for example NVIDIA TAO, ONNX, CAFFE, UFF (TensorFlow), and Triton.
+
+- **Supporting bring your own model** 
+
+  Support for model and container customization, USB or RTSP camera and pre-recorded video streams, event-based video snippet storage in Azure Storage and Alerts, and AI model deployment via Azure IoT Module Twin update.
+
+## Azure Percept for DeepStream key components  
+
+The following table provides a list of Azure Percept for DeepStream’s key components and a description of each one.
+
+| Components              | Details                      | 
+|-------------------------|------------------------------|             
+| **Edge devices**            | Azure Percept for DeepStream is available on the following devices:<br> - [Azure Stack HCI](/azure-stack/hci/overview): Requires a NVIDIA GPU (T4 or A2)<br> - [NVIDIA Jetson Orin](https://www.nvidia.com/autonomous-machines/embedded-systems/jetson-orin/)<br> - [NVIDIA Jetson Xavier](https://www.nvidia.com/autonomous-machines/embedded-systems/jetson-agx-xavier/)<br><br>**Note**<br>You can use any of the listed devices with any of the development paths. Some implementation steps may differ depending on the architecture of your device. Azure Stack HCI uses AMD64. Jetson devices use ARM64.<br><br>                       | 
+| **Computer vision models**  | Azure Percept for DeepStream can work with many different computer vision (CV) models as outlined:<br><br> - **NVIDIA Models** <br>For example: Body Pose Estimation and License Plate Recognition. License Plate Recognition includes three models: traffic cam net, license plate detection, and license plate reading and other Nivida Models.<br><br> - **ONNX Models** <br>For example: SSD-MobileNetV1, YOLOv4, Tiny YOLOv3, EfficentNet-Lite.<br><br>                 | 
+| **Development Paths**       | Azure Percept for DeepStream offers three development paths:<br><br> - **Getting started path** <br>This path uses pre-trained models and pre-recorded videos of simulated manufacturing environment to demonstrate the steps required to create an Edge AI solution using Azure Percept for DeepStream.<br>If you're just getting started on your computer vision (CV) app journey or simply want to learn more about Azure Percept for DeepStream, we recommend this path.<br><br> - **Pre-built model path** <br>This path provides pre-built parsers in Python for the CV models outlined earlier. You can easily deploy one of these models and integrate your own video stream.<br>If you're familiar with Azure IoT Edge solutions and want to leverage one of the supported models with an existing video stream, we recommend this path. <br><br> - **Bring your own model (BYOM) path**<br>This path provides you with steps of how to integrate your own custom model and parser into your Azure Percept for DeepStream Edge AI solution.<br>If you're an experienced developer who is familiar with cloud-based CV solutions and want a simplified deployment experience Azure Percept for DeepStream, we recommend this path.<br><br>                   | 
+
+## Next steps
+
+Text to come.
+
+<!-- You're now ready to start using Azure Percept for DeepStream to create, manage, and deploy custom Edge AI solutions. We recommend the following resources to get started:  
+
+- [Getting started checklist for Azure Percept for DeepStream](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/EeWQwQ8T-LVDmTMqC62Gss0Bo_1Fbjj9I8mDSLYwlICd_Q?e=f9FajM)
+
+- [Tutorial: Deploy a supported model to your Azure Percept for DeepStream solution ](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/EQ9Wux4CkO5Iss8s82lcZj4B9XCwagaVoUEKyK0q2y-A1w?e=YfOaWn)  -->

articles/azure-percept/hci/azure-percept-on-azure-stack-hci-overview.md

@@ -0,0 +1,67 @@
+---
+title: Azure Percept on Azure Stack HCI overview
+description: A description of Azure Percept on Azure Stack HCI. 
+author: yvonne-deq
+ms.author: v-mdequadros
+manager: amiyouss
+ms.service: azure-percept
+ms.topic: overview
+ms.date: 08/15/2022 
+---
+
+# Azure Percept on Azure Stack HCI overview
+Azure Percept on Azure Stack HCI is a virtualized workload that enables you to extend the capabilities of your existing [Azure Stack HCI](https://azure.microsoft.com/products/azure-stack/hci/) deployments quickly and easily by adding sophisticated AI solutions at the Edge. It is available as a preconfigured virtual hard disk (VHDX) that functions as an Azure IoT Edge device with AI capabilities. 
+
+## Azure Percept on Azure Stack HCI enables you:
+
+### Maximize your investments easily
+Maximize your existing investments in the Azure Stack HCI computer infrastructure when you run Azure Percept on Azure Stack HCI. You can leverage [Windows Admin Center (WAC)](https://www.microsoft.com/windows-server/windows-admin-center) management expertise with Azure Percept for Azure Stack HCI extension to ingest and analyze data streams from your existing IP camera infrastructure. Using WAC also enables you to easily deploy, manage, scale, and secure your Azure Percept virtual machine (VM).
+
+### Bring data to storage and compute
+Use Azure Stack HCI’s robust storage and compute options to pre-process raw data at the Edge before sending it to Azure for further processing and training. Since artificial intelligence/machine learning (AI/ML) solutions at the edge generate and process a significant amount of data, using Azure Stack HCI reduces the amount of data transfer or bandwidth consumed into Azure.
+
+### Maintain device security
+Azure Percept on Azure Stack HCI provides multiple layers of security. Leverage security mechanisms and processes built into the solution, including virtual trusted platform module (TPM), secure boot, secure provisioning, trusted software, secure update, and [Microsoft Defender for IoT](https://www.microsoft.com/security/blog/2021/11/02/how-microsoft-defender-for-iot-can-secure-your-iot-devices/#:~:text=Microsoft%20Defender%20for%20IoT%20is%20an%20open%20platform,to%20enrich%20the%20information%20coming%20from%20multiple%20sources).
+
+## Key components of Azure Percept on Azure Stack HCI
+Azure Percept on Azure Stack HCI integrates with Azure Percept Studio, Azure IoT Edge, IoT Hub, and Spatial Analysis from Azure Cognitive Services to create an end-to-end intelligent solution that leverages your existing IP camera devices.
+
+The following diagram provides a high-level view of the Azure Percept on Azure Stack HCI architecture.
+
+![Architecture diagram for Azure Percept on Azure Stack HCI.](./media/azure-percept-component-diagram.png)
+
+**Azure Percept on Azure Stack HCI includes the following key components:**  
+
+### Azure Stack HCI
+[Azure Stack HCI](https://azure.microsoft.com/products/azure-stack/hci/) is a hyperconverged infrastructure (HCI) cluster solution that hosts virtualized Windows and Linux workloads and their storage in a hybrid environment that combines on-premises infrastructure with Azure cloud services. It requires a minimum of two clustered compute nodes, scales to as many as 16 clustered nodes, and enables data pre-processing at the edge by providing robust storage and compute options. Azure Percept on Azure Stack HCI runs as a pre-configured VM on Azure Stack HCI and has failover capability to ensure continuous operation. For information about customizable solutions that you can configure to meet your needs, see [certified Azure Stack HCI systems](https://azurestackhcisolutions.azure.microsoft.com/#/catalog).
+
+### Azure Percept virtual machine (VM)
+The Azure Percept VM leverages a virtual hard disk (VHDX) that runs on the Azure Stack HCI device. It enables you to host your own AI models, communicate with the cloud via IoT Hub, and update the Azure Percept virtual machine (VM) so you can update containers, download models, and manage devices remotely.
+
+The Percept VM leverages Azure IoT Edge to communicate with [Azure IoT Hub](https://www.bing.com/aclk?ld=e8d3D-tqxgHU7f2fug-xNf9TVUCUyRhu5fu58-tWHmwhmAtKIzkXCQETOv1QnKdXCr1kFm6NQ4SA4K5mukLPrpKC5z7nTlhrXnaiTqPPGu2a47SnDq-aKylUzhYQLxKs1yyOtnDuD1DDg4q04CZdFUFwPani9jnp6DLiQPMoYBkhhEJ3FV6SFro1VVB67p_n_4De1B7A&u=aHR0cHMlM2ElMmYlMmZhenVyZS5taWNyb3NvZnQuY29tJTJmZW4tdXMlMmZmcmVlJTJmaW90JTJmJTNmT0NJRCUzZEFJRDIyMDAyNzdfU0VNX2VhM2NkYWExN2Y5MzFkNDE2NTkwYjgyMjdlMjk0ZjdmJTNhRyUzYXMlMjZlZl9pZCUzZGVhM2NkYWExN2Y5MzFkNDE2NTkwYjgyMjdlMjk0ZjdmJTNhRyUzYXMlMjZtc2Nsa2lkJTNkZWEzY2RhYTE3ZjkzMWQ0MTY1OTBiODIyN2UyOTRmN2Y&rlid=ea3cdaa17f931d416590b8227e294f7f&ntb=1). It runs locally and securely, performs AI inferencing at the Edge, and communicates with Azure services for security and updates. It includes [Defender for IoT](https://www.bing.com/ck/a?!&&p=4b4f5983a77f5d870170a12cd507a8d967bd32e10eab125544ac7aad1691be23JmltdHM9MTY1Mjc1MzE3OCZpZ3VpZD1mZmQyZGJiNi1iOWFlLTRiYjgtOTQ1MC1iM2FlNmQ1ZTBlNmUmaW5zaWQ9NTQ1Mg&ptn=3&fclid=f087fcb3-d585-11ec-b34a-9f80cb12a098&u=a1aHR0cHM6Ly9henVyZS5taWNyb3NvZnQuY29tL2VuLXVzL3NlcnZpY2VzL2lvdC1kZWZlbmRlci8&ntb=1) to provide a lightweight security agent that proactively monitors for security threats like botnets, brute force attempts, crypto miners, malware, and chatbots, that you can also integrate into your Azure Monitor infrastructure.
+
+### Azure Percept Windows Admin Center Extension (WAC)
+[Windows Admin Center (WAC)](https://www.microsoft.com/windows-server/windows-admin-center) is a locally deployed application accessed via your browser for managing Azure Stack HCI clusters, Windows Server, and more. Azure Percept on Azure Stack HCI is installed through a WAC extension that guides the user through configuring and deploying the Percept VM and related services. It creates a secure and performant AI video inferencing solution usable from the edge to the cloud.
+
+### Azure Percept Solution Development Paths 
+Whether you're a beginner, an expert, or anywhere in between, from zero to low code, to creating or bringing your own models, Azure Percept has a solution development path for you to build your Edge artificial intelligence (AI) solution. Azure Percept has three solution development paths that you can use to build Edge AI solutions: Azure Percept Studio, Azure Percept for DeepStream, and Azure Percept Open-Source Project. You aren't limited to one path; you can choose any or all of them depending on your business needs. For more information about the solution development paths, visit [Azure Percept solution development paths overview](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/EU92ZnNynDBGuVn3P5Xr5gcBFKS5HQguZm7O5sEENPUvPA?e=33T6Vi).
+
+#### *Azure Percept Studio* 
+[Azure Percept Studio](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/EeyEj0dBcplEs9LSFaz95DsBApnmxRMdjZ9I3QinSgO0yA?e=cbIJkI) is a user-friendly portal for creating, deploying, and operating Edge artificial intelligence (AI) solutions. Using a low-code to no-code approach, you can discover and complete guided workflows and create an end-to-end Edge AI solution. This solution integrates Azure IoT and Azure AI cloud services like Azure IoT Hub, IoT Edge, Azure Storage, Log Analytics, and Spatial Analysis from Azure Cognitive Services.   
+
+#### *Azure Percept for DeepStream*
+[Azure Percept for DeepStream](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/ETDSdi6ruptBkwMqvLPRL90Bzv3ORhpmAZ1YLeGt1LvtVA?e=lY2Q4f&CID=DDDB383F-4BFE-4C97-86A7-70766B16EB93&wdLOR=cDA23C19C-5685-46EC-BA28-7C9DEC460A5B&isSPOFile=1&clickparams=eyJBcHBOYW1lIjoiVGVhbXMtRGVza3RvcCIsIkFwcFZlcnNpb24iOiIyNy8yMjA3MzEwMTAwNSIsIkhhc0ZlZGVyYXRlZFVzZXIiOmZhbHNlfQ%3D%3D) includes developer tools that provide a custom developer experience. It enables you to create NVIDIA DeepStream containers using Microsoft-based images and guidance, supported models from NVIDIA out of the box, and/or bring your own models (BYOM). DeepStream is NVIDIA’s toolkit to develop and deploy Vision AI applications and services. It provides multi-platform, scalable, Transport Layer Security (TLS)-encrypted security that can be deployed on-premises, on the edge, and in the cloud.  
+
+#### *Azure Percept Open-Source Project*
+[Azure Percept Open-Source Project](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/Eeoh0pZk5g1MqwJZUAZFEvEBMYmfAqdibII6Znm-PnnDIQ?e=4ZDfUT) is a framework for creating, deploying, and operating Edge artificial intelligence (AI) solutions at scale with the control and flexibility of open-source natively on your environment. Azure Percept Open-Source Project is fully open-sourced and leverages the open-source software (OSS) community to deliver enhanced experiences. It's a self-managed solution where you host the environment in your own cluster. 
+
+## Next steps
+
+Text to come.
+
+<!-- Before you start setting up your Azure Percept virtual machine (VM), we recommend the following articles: 
+- [Getting started checklist for Azure Percept on Azure Stack HCI](https://github.com/microsoft/santa-cruz-workload/blob/main/articles/getting-started-checklist-for-azure-percept.md)
+- [Azure Percept solution development paths overview](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/EU92ZnNynDBGuVn3P5Xr5gcBFKS5HQguZm7O5sEENPUvPA?e=DKZtr6) 
+
+If you’re ready to start setting up your Azure Percept virtual machine (VM), we recommend the following tutorial: 
+- [Tutorial: Setting up Azure Percept on Azure Stack HCI using WAC extension (Cluster server)](https://github.com/microsoft/santa-cruz-workload/blob/main/articles/tutorial-setting-up-azure-percept-using-wac-extension-cluster.md) -->

articles/azure-percept/hci/media/azure-percept-component-diagram.png

@@ -0,0 +1,51 @@
+---
+title: Azure Percept Open-Source Project overview
+description: An overview of the Azure Percept Open-Source project
+author: yvonne-deq
+ms.author: v-mdequadros
+manager: amiyouss
+ms.service: azure-percept
+ms.topic: overview
+ms.date: 08/17/2022 
+---
+
+# Azure Percept Open-Source Project overview
+
+Azure Percept Open-Source Project is a framework for creating, deploying, and operating Edge artificial intelligence (AI) solutions at scale with the control and flexibility of open-source natively on your environment. It's fully open-sourced and leverages the open-source software (OSS) community to deliver enhanced experiences. And, as a self-managed solution, you can host the experience on your own Kubernetes clusters.
+
+Azure Percept Open-Source Project has a no- to low-code portal experience as well as APIs that can be used to build custom Edge AI applications. It supports running Edge AI apps by utilizing cameras and Edge devices with different Edge runtimes and accelerators across multiple locations at scale. Since it's designed with machine learning operations (MLOps) in mind, it provides support for active learning, continuous training, and data gathering using your machine learning (ML) models running at the edge.
+
+## Azure Percept Open-Source Project offers
+
+- **An integrated developer experience** 
+
+  You can easily build camera-based Edge AI apps using first- and third-party ML models. In one seamless flow, you can leverage pre-built models from our partner’s Model Zoo and create your own ML models with Azure Custom Vision.
+
+- **Solution deployment and management experience at scale**
+
+  Azure Percept Open-Source Project is Kubernetes native, so you can run the experience wherever Kubernetes runs; on-premises, hybrid, cloud, or multicloud environments. You can manage your experience using Kubernetes native tools such as Kubectl, our unique command line interface (CLI), and/or our no- to low-code native web portal. Edge AI apps and assets you create are projected and managed as Kubernetes objects, which allows you to rely on the Kubernetes control plane to manage the state of your Edge AI assets across many environments at scale.
+
+- **Standard-based**
+
+  Azure Percept Open-Source Project is built on and supports popular industrial standards, protocols, and frameworks like Open Platform Communications Unified Architecture (OPC-UA), Open Network Video Interface Forum (ONVIF), OpenTelemetry, CloudEvents, Distributed Application Runtime (Dapr), Message Queuing Telemetry Transport (MQTT), Open Neural Network Exchange (ONNX), Akri, Kubectl, Helm, and many others. 
+
+- **Zero-friction adoption**
+
+  Even without any Edge hardware, you can get started with a few commands, then seamlessly transit from prototype/pilot to production at scale. Azure Percept Open-Source Project has an easy-to-use no- to low-code portal experience that allows developers to create and manage Edge AI solutions in minutes instead of days or months. 
+
+- **Azure powered and platform agnostic**
+
+  Azure Percept Open-Source Project natively uses and supports Azure Edge and AI Services like Azure IoT Hub, Azure IoT Edge, Azure Cognitive Services, Azure Storage Server, Azure ML, and so on. At the same time, it also allows you to modify the experience for use cases that require the use of other services (Azure or non-Azure) or other Open-Source Software (OSS) tools. 
+
+## Next steps
+
+Text to come.
+
+<!-- You're now ready to start using Azure Percept Open-Source Project. We recommend the following resources to get started. 
+
+ - TBD (getting started) How to get started and setup Azure Percept Open-Source Project 
+
+- [Introduction to Azure Percept for Open-Source Project core concepts](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/EQwRE6w96T1OiO_kstWw1lMBs1yZFUow_ik3kx3rV12EVg?e=bactOi) 
+
+- [Tutorial: Create an Edge AI solution with Azure Percept for Open-Source Project](https://microsoft.sharepoint-df.com/:w:/t/AzurePerceptHCIDocumentation/ERF8mxgtOqhIt2YJWFafuZoBC6kZ6hC-iRAMuCJeyZjD-w?e=BS4cN5)
+-->

articles/azure-percept/open-source/azure-percept-open-source-project-overview.md

@@ -1,5 +1,5 @@
 ---
-title: Azure Percept Studio overview
+title: Azure Percept Studio overview v1
 description: Learn more about Azure Percept Studio
 author: nkhuyent
 ms.author: nbabar
@@ -9,7 +9,7 @@ ms.date: 03/23/2021
 ms.custom: template-concept #Required; leave this attribute/value as-is.
 ---
 
-# Azure Percept Studio overview
+# Azure Percept Studio overview v1
 
 [Azure Percept Studio](https://go.microsoft.com/fwlink/?linkid=2135819) is the single launch point for creating edge AI models and solutions. Azure Percept Studio allows you to discover and complete guided workflows that make it easy to integrate edge AI-capable hardware and powerful Azure AI and IoT cloud services.