Merge pull request #229496 from Padmalathas/Padmalathas-patch-1

Refactoring abandoned TOC

Author: ttorble

articles/virtual-machines/.openpublishing.redirection.virtual-machines.json

@@ -4918,6 +4918,46 @@
             "source_path_from_root": "/articles/virtual-machines/linux/azure-hybrid-benefit-byos-linux.md",
             "redirect_url": "/azure/virtual-machines/linux/azure-hybrid-benefit-linux/",
             "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hb-series-overview.md",
+            "redirect_url": "/azure/virtual-machines/hb-series-overview",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hb-series-performance.md",
+            "redirect_url": "/azure/virtual-machines/hb-series-performance",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hbv2-performance.md",
+            "redirect_url": "/azure/virtual-machines/hbv2-performance",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hbv2-series-overview.md",
+            "redirect_url": "/azure/virtual-machines/hbv2-series-overview",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hbv3-performance.md",
+            "redirect_url": "/azure/virtual-machines/hbv3-performance",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hbv3-series-overview.md",
+            "redirect_url": "/azure/virtual-machines/hbv3-series-overview",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hc-series-overview.md",
+            "redirect_url": "/azure/virtual-machines/hc-series-overview",
+            "redirect_document_id": false
+        },
+        {
+            "source_path_from_root": "/articles/virtual-machines/workloads/hpc/hc-series-performance.md",
+            "redirect_url": "/azure/virtual-machines/hc-series-performance",
+            "redirect_document_id": false
         }
     ]
 }

articles/virtual-machines/TOC.yml

@@ -339,36 +339,36 @@
         href: hb-series.md
         items:
         - name: Overview
-          href: ./workloads/hpc/hb-series-overview.md
+          href: hb-series-overview.md
         - name: Performance
-          href: ./workloads/hpc/hb-series-performance.md
+          href: hb-series-performance.md
         - name: HB-series retirement
           href: hb-series-retirement.md
       - name: HBv2-series
         href: hbv2-series.md
         items:
         - name: Overview
-          href: ./workloads/hpc/hbv2-series-overview.md
+          href: hbv2-series-overview.md
         - name: Performance
-          href: ./workloads/hpc/hbv2-performance.md
+          href: hbv2-performance.md
       - name: HBv3-series
         items:
         - name: Overview
-          href: ./workloads/hpc/hbv3-series-overview.md
+          href: hbv3-series-overview.md
         - name: HBv3-series
           href: hbv3-series.md
         - name: Performance
-          href: ./workloads/hpc/hbv3-performance.md
+          href: hbv3-performance.md
       - name: HBv4-series
         href: hbv4-series.md
       - name: HC-series
         items:
         - name: Overview
-          href: ./workloads/hpc/hc-series-overview.md
+          href: hc-series-overview.md
         - name: HC-series
           href: hc-series.md
         - name: Performance
-          href: ./workloads/hpc/hc-series-performance.md
+          href: hc-series-performance.md
       - name: HX-series
         href: hx-series.md
     - name: Previous generations

articles/virtual-machines/workloads/hpc/hb-series-overview.md renamed to articles/virtual-machines/hb-series-overview.md

@@ -4,7 +4,7 @@ description: Learn about the preview support for the HB-series VM size in Azure.
 ms.service: virtual-machines
 ms.subservice: hpc
 ms.topic: article
-ms.date: 08/19/2020
+ms.date: 03/04/2023
 ms.reviewer: cynthn
 ms.author: mamccrea
 author: mamccrea
@@ -16,7 +16,7 @@ author: mamccrea
 
 Maximizing high performance compute (HPC) application performance on AMD EPYC requires a thoughtful approach memory locality and process placement. Below we outline the AMD EPYC architecture and our implementation of it on Azure for HPC applications. We will use the term “pNUMA” to refer to a physical NUMA domain, and “vNUMA” to refer to a virtualized NUMA domain.
 
-Physically, an [HB-series](../../hb-series.md) server is 2 * 32-core EPYC 7551 CPUs for a total of 64 physical cores. These 64 cores are divided into 16 pNUMA domains (8 per socket), each of which is four cores and known as a “CPU Complex” (or “CCX”). Each CCX has its own L3 cache, which is how an OS will see a pNUMA/vNUMA boundary. A pair of adjacent CCXs shares access to two channels of physical DRAM (32 GB of DRAM in HB-series servers).
+Physically, an [HB-series](hb-series.md) server is 2 * 32-core EPYC 7551 CPUs for a total of 64 physical cores. These 64 cores are divided into 16 pNUMA domains (8 per socket), each of which is four cores and known as a “CPU Complex” (or “CCX”). Each CCX has its own L3 cache, which is how an OS will see a pNUMA/vNUMA boundary. A pair of adjacent CCXs shares access to two channels of physical DRAM (32 GB of DRAM in HB-series servers).
 
 To provide room for the Azure hypervisor to operate without interfering with the VM, we reserve physical pNUMA domain 0 (the first CCX). We then assign pNUMA domains 1-15 (the remaining CCX units) for the VM. The VM will see:
 
@@ -28,7 +28,7 @@ Process pinning will work on HB-series VMs because we expose the underlying sili
 
 The following diagram shows the segregation of cores reserved for Azure Hypervisor and the HB-series VM.
 
-![Segregation of cores reserved for Azure Hypervisor and HB-series VM](./media/architecture/hb-segregation-cores.png)
+![Segregation of cores reserved for Azure Hypervisor and HB-series VM](./media/hpc/architecture/hb-segregation-cores.png)
 
 ## Hardware specifications
 
@@ -51,7 +51,7 @@ The following diagram shows the segregation of cores reserved for Azure Hypervis
 | Additional Frameworks       | UCX, libfabric, PGAS |
 | Azure Storage Support       | Standard and Premium Disks (maximum 4 disks) |
 | OS Support for SRIOV RDMA   | CentOS/RHEL 7.6+, Ubuntu 16.04+, SLES 12 SP4+, WinServer 2016+  |
-| Orchestrator Support        | CycleCloud, Batch, AKS; [cluster configuration options](../../sizes-hpc.md#cluster-configuration-options) |
+| Orchestrator Support        | CycleCloud, Batch, AKS; [cluster configuration options](sizes-hpc.md#cluster-configuration-options) |
 
 ## Next steps
 

articles/virtual-machines/workloads/hpc/hb-series-performance.md renamed to articles/virtual-machines/hb-series-performance.md

@@ -4,7 +4,7 @@ description: Learn about performance testing results for HB-series VM sizes in A
 ms.service: virtual-machines
 ms.subservice: hpc
 ms.topic: article
-ms.date: 09/09/2020
+ms.date: 03/04/2023
 ms.reviewer: cynthn
 ms.author: mamccrea
 author: mamccrea
@@ -33,7 +33,7 @@ MPI latency test from the OSU microbenchmark suite is run. Sample scripts are on
 ./bin/mpirun_rsh -np 2 -hostfile ~/hostfile MV2_CPU_MAPPING=[INSERT CORE #] ./osu_latency 
 ```
 
-:::image type="content" source="./media/latency-hb.png" alt-text="MPI latency on Azure HB.":::
+![Screenshot of MPI latency.](./media/hpc/latency-hb.png)
 
 ## MPI bandwidth
 
@@ -43,7 +43,7 @@ MPI bandwidth test from the OSU microbenchmark suite is run. Sample scripts are
 ./mvapich2-2.3.install/bin/mpirun_rsh -np 2 -hostfile ~/hostfile MV2_CPU_MAPPING=[INSERT CORE #] ./mvapich2-2.3/osu_benchmarks/mpi/pt2pt/osu_bw
 ```
 
-:::image type="content" source="./media/bandwidth-hb.png" alt-text="MPI bandwidth on Azure HB.":::
+![Screenshot of MPI bandwidth.](./media/hpc/bandwidth-hb.png)
 
 
 ## Mellanox Perftest

articles/virtual-machines/hb-series.md

@@ -4,7 +4,7 @@ description: Specifications for the HB-series VMs.
 ms.service: virtual-machines
 ms.subservice: sizes
 ms.topic: conceptual
-ms.date: 12/19/2022
+ms.date: 03/04/2023
 ms.reviewer: jushiman
 ---
 
@@ -32,11 +32,11 @@ HB-series VMs feature 100 Gb/sec Mellanox EDR InfiniBand. These VMs are connecte
 | Standard_HB60rs | 60 | AMD EPYC 7551 | 228 | 263 | 2.0 | 2.55 | 2.55 | 100 | All | 700 | 4 | 8 |
 
 Learn more about the:
-- [Architecture and VM topology](./workloads/hpc/hb-series-overview.md)
-- Supported [software stack](./workloads/hpc/hb-series-overview.md#software-specifications) including supported OS
-- Expected [performance](./workloads/hpc/hb-series-performance.md) of the HB-series VM
+- [Architecture and VM topology](hb-series-overview.md)
+- Supported [software stack](hb-series-overview.md#software-specifications) including supported OS
+- Expected [performance](hb-series-performance.md) of the HB-series VM
 
-[!INCLUDE [hpc-include.md](./workloads/hpc/includes/hpc-include.md)]
+[!INCLUDE [hpc-include.md](./includes/hpc-include.md)]
 
 [!INCLUDE [virtual-machines-common-sizes-table-defs](../../includes/virtual-machines-common-sizes-table-defs.md)]
 

articles/virtual-machines/workloads/hpc/hbv2-performance.md renamed to articles/virtual-machines/hbv2-performance.md

@@ -6,7 +6,7 @@ ms.service: virtual-machines
 ms.subservice: hpc
 ms.workload: infrastructure-services 
 ms.topic: article 
-ms.date: 09/28/2020 
+ms.date: 03/04/2023 
 ms.reviewer: cynthn
 ms.author: mamccrea
 author: mamccrea
@@ -16,7 +16,7 @@ author: mamccrea
 
 **Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets
 
-Several performance tests have been run on [HBv2-series](../../hbv2-series.md) size VMs. The following are some of the results of this performance testing.
+Several performance tests have been run on [HBv2-series](hbv2-series.md) size VMs. The following are some of the results of this performance testing.
 
 
 | Workload                                        | HBv2                                                              |
@@ -36,7 +36,7 @@ MPI latency test from the OSU microbenchmark suite is run. Sample scripts are on
 ./bin/mpirun_rsh -np 2 -hostfile ~/hostfile MV2_CPU_MAPPING=[INSERT CORE #] ./osu_latency
 ```
 
-:::image type="content" source="./media/latency-hbv2.png" alt-text="MPI latency on Azure HB.":::
+![Screenshot of MPI latency.](./media/hpc/latency-hbv2.png)
 
 ## MPI bandwidth
 
@@ -46,7 +46,7 @@ MPI bandwidth test from the OSU microbenchmark suite is run. Sample scripts are
 ./mvapich2-2.3.install/bin/mpirun_rsh -np 2 -hostfile ~/hostfile MV2_CPU_MAPPING=[INSERT CORE #] ./mvapich2-2.3/osu_benchmarks/mpi/pt2pt/osu_bw
 ```
 
-:::image type="content" source="./media/bandwidth-hbv2.png" alt-text="MPI bandwidth on Azure HB.":::
+![Screenshot of MPI bandwidth.](./media/hpc/bandwidth-hbv2.png)
 
 ## Mellanox Perftest
 
@@ -61,4 +61,4 @@ numactl --physcpubind=[INSERT CORE #]  ib_send_lat -a
 ## Next steps
 
 - Read about the latest announcements, HPC workload examples, and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute).
-- For a higher-level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/).
+- For a higher-level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/).

articles/virtual-machines/workloads/hpc/hbv2-series-overview.md renamed to articles/virtual-machines/hbv2-series-overview.md

@@ -7,7 +7,7 @@ ms.service: virtual-machines
 ms.subservice: hpc
 ms.workload: infrastructure-services 
 ms.topic: article 
-ms.date: 12/15/2020 
+ms.date: 03/04/2023 
 ms.reviewer: cynthn
 ms.author: mamccrea
 author: mamccrea
@@ -18,17 +18,17 @@ author: mamccrea
 
 **Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets
 
-Maximizing high performance compute (HPC) application performance on AMD EPYC requires a thoughtful approach memory locality and process placement. Below we outline the AMD EPYC architecture and our implementation of it on Azure for HPC applications. We will use the term **pNUMA** to refer to a physical NUMA domain, and **vNUMA** to refer to a virtualized NUMA domain. 
+Maximizing high performance compute (HPC) application performance on AMD EPYC requires a thoughtful approach memory locality and process placement. Below we outline the AMD EPYC architecture and our implementation of it on Azure for HPC applications. We use the term **pNUMA** to refer to a physical NUMA domain, and **vNUMA** to refer to a virtualized NUMA domain. 
 
-Physically, an [HBv2-series](../../hbv2-series.md) server is 2 * 64-core EPYC 7742 CPUs for a total of 128 physical cores. These 128 cores are divided into 32 pNUMA domains (16 per socket), each of which is 4 cores and termed by AMD as a **Core Complex** (or **CCX**). Each CCX has its own L3 cache, which is how an OS will see a pNUMA/vNUMA boundary. Four adjacent CCXs share access to 2 channels of physical DRAM. 
+Physically, an [HBv2-series](hbv2-series.md) server is 2 * 64-core EPYC 7742 CPUs for a total of 128 physical cores. These 128 cores are divided into 32 pNUMA domains (16 per socket), each of which is 4 cores and termed by AMD as a **Core Complex** (or **CCX**). Each CCX has its own L3 cache, which is how an OS sees a pNUMA/vNUMA boundary. Four adjacent CCXs share access to two channels of physical DRAM. 
 
-To provide room for the Azure hypervisor to operate without interfering with the VM, we reserve physical pNUMA domains 0 and 16 (i.e. the first CCX of each CPU socket). All remaining 30 pNUMA domains are assigned to the VM at which point they become vNUMA. Thus, the VM will see:
+To provide room for the Azure hypervisor to operate without interfering with the VM, we reserve physical pNUMA domains 0 and 16 (that is, the first CCX of each CPU socket). All remaining 30 pNUMA domains are assigned to the VM at which point they become vNUMA. Thus, the VM sees:
 
 `(30 vNUMA domains) * (4 cores/vNUMA) = 120` cores per VM 
 
 The VM itself has no awareness that pNUMA 0 and 16 are reserved. It enumerates the vNUMA it sees as 0-29, with 15 vNUMA per socket symmetrically, vNUMA 0-14 on vSocket 0, and vNUMA 15-29 on vSocket 1.
 
-Process pinning will work on HBv2-series VMs because we expose the underlying silicon as-is to the guest VM. We strongly recommend process pinning for optimal performance and consistency. 
+Process pinning works on HBv2-series VMs because we expose the underlying silicon as-is to the guest VM. We strongly recommend process pinning for optimal performance and consistency. 
 
 
 ## Hardware specifications 
@@ -53,7 +53,7 @@ Process pinning will work on HBv2-series VMs because we expose the underlying si
 | Additional Frameworks       | UCX, libfabric, PGAS |
 | Azure Storage Support       | Standard and Premium Disks (maximum 8 disks) |
 | OS Support for SRIOV RDMA   | CentOS/RHEL 7.6+, Ubuntu 16.04+, SLES 12 SP4+, WinServer 2016+  |
-| Orchestrator Support        | CycleCloud, Batch, AKS; [cluster configuration options](../../sizes-hpc.md#cluster-configuration-options)  |
+| Orchestrator Support        | CycleCloud, Batch, AKS; [cluster configuration options](sizes-hpc.md#cluster-configuration-options)  |
 
 > [!NOTE] 
 > Windows Server 2012 R2 is not supported on HBv2 and other VMs with more than 64 (virtual or physical) cores. See [Supported Windows guest operating systems for Hyper-V on Windows Server](/windows-server/virtualization/hyper-v/supported-windows-guest-operating-systems-for-hyper-v-on-windows) for more details.

articles/virtual-machines/hbv2-series.md

@@ -4,7 +4,7 @@ description: Specifications for the HBv2-series VMs.
 ms.service: virtual-machines
 ms.subservice: sizes
 ms.topic: conceptual
-ms.date: 12/19/2022
+ms.date: 03/04/2023
 ms.reviewer: jushiman
 ---
 
@@ -35,11 +35,11 @@ HBv2-series VMs feature 200 Gb/sec Mellanox HDR InfiniBand. These VMs are connec
 | Standard_HB120-16rs_v2 | 16 | AMD EPYC 7V12 | 456 | 350 | 2.45 | 3.1 | 3.3 | 200 | All | 480 + 960 | 8 | 8 |
 
 Learn more about the:
-- [Architecture and VM topology](./workloads/hpc/hbv2-series-overview.md)
-- Supported [software stack](./workloads/hpc/hbv2-series-overview.md#software-specifications) including supported OS
-- Expected [performance](./workloads/hpc/hbv2-performance.md) of the HBv2-series VM
+- [Architecture and VM topology](hbv2-series-overview.md)
+- Supported [software stack](hbv2-series-overview.md#software-specifications) including supported OS
+- Expected [performance](hbv2-performance.md) of the HBv2-series VM
 
-[!INCLUDE [hpc-include](./workloads/hpc/includes/hpc-include.md)]
+[!INCLUDE [hpc-include](./includes/hpc-include.md)]
 
 [!INCLUDE [virtual-machines-common-sizes-table-defs](../../includes/virtual-machines-common-sizes-table-defs.md)]
 

articles/virtual-machines/workloads/hpc/hbv3-performance.md renamed to articles/virtual-machines/hbv3-performance.md

@@ -6,7 +6,7 @@ ms.service: virtual-machines
 ms.subservice: hpc
 ms.workload: infrastructure-services 
 ms.topic: article 
-ms.date: 03/25/2021 
+ms.date: 03/04/2023 
 ms.reviewer: cynthn
 ms.author: mamccrea
 author: mamccrea
@@ -22,16 +22,16 @@ Performance expectations using common HPC microbenchmarks are as follows:
 |-------------------------------------------------|-------------------------------------------------------------------|
 | STREAM Triad                                    | 330-350 GB/s (amplified up to 630 GB/s)                                     |
 | High-Performance Linpack (HPL)                  | 4 TF (Rpeak, FP64), 8 TF (Rpeak, FP32) for 120-core VM size               |
-| RDMA latency & bandwidth                        | 1.2 microseconds (1-byte), 192 Gb/s (one-way)                                        |
+| RDMA latency & bandwidth                        | 1.2 microseconds (1 byte), 192 GB/s (one-way)                                        |
 | FIO on local NVMe SSDs (RAID0)                  | 7 GB/s reads, 3 GB/s writes; 186k IOPS reads, 201k IOPS writes |
 
 ## Process pinning
 
-[Process pinning](compiling-scaling-applications.md#process-pinning) works well on HBv3-series VMs because we expose the underlying silicon as-is to the guest VM. We strongly recommend process pinning for optimal performance and consistency.
+[Process pinning](./workloads/hpc/compiling-scaling-applications.md#process-pinning) works well on HBv3-series VMs because we expose the underlying silicon as-is to the guest VM. We strongly recommend process pinning for optimal performance and consistency.
 
 ## MPI latency
 
-The MPI latency test from the OSU microbenchmark suite can be executed per below. Sample scripts are on [GitHub](https://github.com/Azure/azhpc-images/blob/04ddb645314a6b2b02e9edb1ea52f079241f1297/tests/run-tests.sh).
+The MPI latency test from the OSU microbenchmark suite can be executed as shown. Sample scripts are on [GitHub](https://github.com/Azure/azhpc-images/blob/04ddb645314a6b2b02e9edb1ea52f079241f1297/tests/run-tests.sh).
 
 ```bash
 ./bin/mpirun_rsh -np 2 -hostfile ~/hostfile MV2_CPU_MAPPING=[INSERT CORE #] ./osu_latency
@@ -48,7 +48,7 @@ The [Mellanox Perftest package](https://community.mellanox.com/s/article/perftes
 numactl --physcpubind=[INSERT CORE #]  ib_send_lat -a
 ```
 ## Next steps
-- Learn about [scaling MPI applications](compiling-scaling-applications.md).
+- Learn about [scaling MPI applications](./workloads/hpc/compiling-scaling-applications.md).
 - Review the performance and scalability results of HPC applications on the HBv3 VMs at the [TechCommunity article](https://techcommunity.microsoft.com/t5/azure-compute/hpc-performance-and-scalability-results-with-azure-hbv3-vms/bc-p/2235843).
 - Read about the latest announcements, HPC workload examples, and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute).
 - For a higher-level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/).