Merge pull request #190668 from ianwhytez/main

RAC changes

Author: laurabren

articles/baremetal-infrastructure/workloads/oracle/high-availability-features.md

@@ -22,22 +22,6 @@ You can use this feature instead of a time-delayed redo on the standby database.
 
 The Oracle Database keeps flashback logs in the fast recovery area (FRA). These logs are separate from the redo logs and require more space within the FRA. By default, 24 hours of flashback logs are kept, but you can change this setting per your requirements.
 
-## Oracle Real Application Clusters
-
-[Oracle Real Application Clusters (RAC)](https://docs.oracle.com/en/database/oracle/oracle-database/19/racad/introduction-to-oracle-rac.html#GUID-5A1B02A2-A327-42DD-A1AD-20610B2A9D92) allows multiple interconnected servers to appear as one database service to end users and applications. This feature removes many points of failure and is a recognized high availability active/active solution for Oracle databases.
-
-As shown in the following figure from Oracle's [High Availability Overview and Best Practices](https://docs.oracle.com/en/database/oracle/oracle-database/19/haovw/ha-features.html), a single RAC database is presented to the application layer. The applications connect to the SCAN listener, which directs traffic to a specific database instance. RAC controls access from multiple instances to maintain data consistency across separate compute nodes.
-
-![Diagram showing an overview of the architecture of Oracle RAC.](media/oracle-high-availability/oracle-real-application-clusters.png)
-
-If one instance fails, the service continues on all other remaining instances. Each database deployed on the solution will be in a RAC configuration of n+1, where n is the minimum processing power required to support the service.
-
-Oracle Database services are used to allow connections to fail over between nodes when an instance fails transparently. Such failures may be planned or unplanned. Working with Oracle RAC Fast Application Notification, when an instance is made unavailable, the service is moved to a surviving node. The service moves to a node specified in the service configuration as either preferred or available.
-
-Another key feature of Oracle Database services is only starting a service depending on its role. This feature is used when there's a Data Guard failover. All patterns deployed using Data Guard are required to link a database service to a Data Guard role.
-
-For example, two services could be created, MY\_DB\_APP and MY\_DB\_AS. The MY\_DB\_APP service is started only when the database instance is started with the Data Guard role of PRIMARY. MY\_DB\_AS is only started when the Data Guard role is PHYSICAL\_STANDBY. This configuration allows for applications to point to the \_APP service, while also reporting, which can be offloaded to Active Standby and pointed to the \_AS service.
-
 ## Oracle Data Guard
 
 With Data Guard, you can maintain a copy of a database on separate physical hardware. Ideally, that hardware should be geographically removed from the primary database. Data Guard places no limit on the distance, although distance has a bearing on modes of protection. Increased distance adds latency between sites, which can cause some options (such as synchronous replication) to become untenable.

articles/baremetal-infrastructure/workloads/oracle/oracle-baremetal-architecture.md

@@ -16,40 +16,6 @@ This topology supports a single instance of Oracle Database with Oracle Data Gua
 
 [![Diagram showing the architecture of a single instance of Oracle Database with Oracle Data Guard.](media/oracle-baremetal-architecture/single-instance-architecture.png)](media/oracle-baremetal-architecture/single-instance-architecture.png#lightbox)
 
-## Oracle Real Application Clusters (RAC) One Node
-
-This topology supports a RAC configuration with shared storage and GRID cluster. Database instances run only on one node (active-passive configuration).
-
-Features include:
-
-- Active-passive with Oracle RAC One Node
-
-    - Automatic fail-over
-
-    - Fast restart on second node
-
-- Real-time fail-over and scalability with Oracle RAC
-
-- Zero downtime rolling maintenance
-
-[![Diagram showing the architecture of an Oracle RAC One Node active-passive configuration.](media/oracle-baremetal-architecture/one-node-rac-architecture.png)](media/oracle-baremetal-architecture/one-node-rac-architecture.png#lightbox)
-
-## RAC
-
-This topology supports an Oracle RAC configuration with shared storage and Grid cluster while multiple instances per database run concurrently (active-active configuration).
-
-- Performance is easy to scale through online provisioning of added servers. 
--  Users are active on all servers, and all servers share access to the same Oracle Database. 
--  All types of database maintenance can be performed either online or in rolling fashion for minimal or zero downtime. 
-- Oracle Active Data Guard (ADG) standby systems can easily serve a dual-purpose as test systems. 
-
-This configuration allows you to test all changes on an exact copy of the production database before they're applied to the production environment.
-
-> [!NOTE]
-> If you intend to use Active Data Guard Far Sync (synchronous mode), you'll need to consider the regional zones where this feature is supported. For geographical distributed regions only, we recommend using Data Guard with asynchronous mode.
-
-[![Diagram showing the architecture of an Oracle RAC active-active configuration.](media/oracle-baremetal-architecture/rac-architecture.png)](media/oracle-baremetal-architecture/rac-architecture.png#lightbox)
-
 ## Next steps
 
 Learn about provisioning your BareMetal instances for Oracle workloads.

articles/baremetal-infrastructure/workloads/oracle/oracle-baremetal-overview.md

@@ -17,7 +17,6 @@ These instances are for running mission critical applications requiring an Oracl
 Other features of BareMetal Infrastructure for Oracle include:
 
 - Oracle certified UCS blades - UCSB200-M5, UCSB460-M4, UCSB480-M5
-- Oracle Real Application Clusters (RAC) node-to-node (multi-cast) communication using private virtual LAN (VLAN) -40 Gb.
 - Microsoft-managed hardware
   - Redundant storage, network, power, management
   - Monitoring for Infra, repairs, and replacement

articles/baremetal-infrastructure/workloads/oracle/oracle-baremetal-provision.md

@@ -10,7 +10,7 @@ ms.date: 04/14/2021
 
 In this article, we'll look at how to provision your BareMetal Infrastructure instances for Oracle workloads. 
 
-The first step to provision your BareMetal instances is to work with your Microsoft CSA. They'll help you based on your specific workload needs and the architecture you're deploying, whether single instance, One Node RAC, or RAC. For more information on these topologies, see [Architecture of BareMetal Infrastructure for Oracle](oracle-baremetal-architecture.md).
+The first step to provision your BareMetal instances is to work with your Microsoft CSA. They'll help you based on your specific workload needs and the architecture you're deploying. For more information on these topologies, see [Architecture of BareMetal Infrastructure for Oracle](oracle-baremetal-architecture.md).
 
 ## Prerequisites
 

articles/baremetal-infrastructure/workloads/oracle/oracle-baremetal-storage.md

@@ -10,8 +10,7 @@ ms.date: 04/14/2021
 
 In this article, we'll give an overview of the storage offered by the BareMetal Infrastructure for Oracle workloads.
 
-BareMetal Infrastructure for Oracle offers NetApp Network File System (NFS) storage. NFS storage does not require Oracle Real Application Clusters (RAC) certification. For more information, see [Oracle RAC Technologies Matrix for Linux Clusters](https://www.oracle.com/database/technologies/tech-generic-linux-new.html).
-
+BareMetal Infrastructure for Oracle offers NetApp Network File System (NFS) storage. 
 This storage offering includes Tier 3 support from an OEM partner, using either A700s or A800s storage controllers.
 
 BareMetal Infrastructure storage offers these premium storage capabilities: