Merge pull request #79297 from laurenhughes/cpu-types

v-albemi · web-flow · commit be58387add85 · 2019-06-11T12:55:36.000-07:00
Add CPU info
diff --git a/articles/virtual-machines/linux/sizes-gpu.md b/articles/virtual-machines/linux/sizes-gpu.md
@@ -14,12 +14,12 @@ ms.devlang: na
 ms.topic: article
 ms.tgt_pltfrm: vm-linux
 ms.workload: infrastructure-services
-ms.date: 05/14/2019
+ms.date: 06/11/2019
 ms.author: jonbeck
 
 ---
 
-# Accelerated compute virtual machine sizes
+# GPU optimized virtual machine sizes
 
 [!INCLUDE [virtual-machines-common-sizes-gpu](../../../includes/virtual-machines-common-sizes-gpu.md)]
 
diff --git a/articles/virtual-machines/windows/sizes-gpu.md b/articles/virtual-machines/windows/sizes-gpu.md
@@ -14,7 +14,7 @@ ms.devlang: na
 ms.topic: article
 ms.tgt_pltfrm: vm-windows
 ms.workload: infrastructure-services
-ms.date: 09/24/2018
+ms.date: 06/11/2019
 ms.author: jonbeck
 
 ---
diff --git a/includes/virtual-machines-common-sizes-gpu.md b/includes/virtual-machines-common-sizes-gpu.md
@@ -5,30 +5,26 @@
  author: cynthn
  ms.service: multiple
  ms.topic: include
- ms.date: 05/14/2019
+ ms.date: 06/11/2019
  ms.author: cynthn;azcspmt;jonbeck
  ms.custom: include file
 ---
 
-GPU optimized VM sizes are specialized virtual machines available with single or multiple NVIDIA GPUs. These sizes are designed for compute-intensive, graphics-intensive, and visualization workloads. This article provides information about the number and type of GPUs, vCPUs, data disks, and NICs. Storage throughput and network bandwidth are also included for each size in this grouping. 
+GPU optimized VM sizes are specialized virtual machines available with single or multiple NVIDIA GPUs. These sizes are designed for compute-intensive, graphics-intensive, and visualization workloads. This article provides information about the number and type of GPUs, vCPUs, data disks, and NICs. Storage throughput and network bandwidth are also included for each size in this grouping.
 
-* **NC, NCv2, NCv3, ND, and NDv2** sizes are optimized for compute-intensive and network-intensive applications and algorithms. Some examples are CUDA- and OpenCL-based applications and simulations, AI, and Deep Learning. The NCv3-series is focused on high-performance computing workloads featuring NVIDIA’s Tesla V100 GPU.  The ND-series is focused on training and inference scenarios for deep learning. It uses the NVIDIA Tesla P40 GPU.
+* **NC, NCv2, NCv3** sizes are optimized for compute-intensive and network-intensive applications and algorithms. Some examples are CUDA- and OpenCL-based applications and simulations, AI, and Deep Learning. The NCv3-series is focused on high-performance computing workloads featuring NVIDIA’s Tesla V100 GPU. The NC-series uses the Intel Xeon E5-2690 v3 2.60GHz v3 (Haswell) processor, and the NCv2-series and NCv3-series VMs use the Intel Xeon E5-2690 v4 (Broadwell) processor.
 
-* The **NC-series** features an	Intel Xeon® E5-2690 v3 2.60GHz processor.
+* **ND, and NDv2** The ND-series is focused on training and inference scenarios for deep learning. It uses the NVIDIA Tesla P40 GPU and the Intel Xeon E5-2690 v4 (Broadwell) processor. The NDv2-series uses the Intel Xeon Platinum 8168 (Skylake) processor.
 
-* The **NCSv3**, **NCSv2**, and **ND** sizes feature an	Intel Xeon® E5-2690 v4 2.60GHz processor.
-                      
 * **NV and NVv3** sizes are optimized and designed for remote visualization, streaming, gaming, encoding, and VDI scenarios using frameworks such as OpenGL and DirectX.  These VMs are backed by the NVIDIA Tesla M60 GPU.
 
-
 ## NC-series
 
 Premium Storage:  Not Supported
 
 Premium Storage caching:  Not Supported
 
-NC-series VMs are powered by the [NVIDIA Tesla K80](https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/tesla-product-literature/Tesla-K80-BoardSpec-07317-001-v05.pdf) card. Users can crunch through data faster by leveraging CUDA for energy exploration applications, crash simulations, ray traced rendering, deep learning, and more. The NC24r configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads.
-
+NC-series VMs are powered by the [NVIDIA Tesla K80](https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/tesla-product-literature/Tesla-K80-BoardSpec-07317-001-v05.pdf) card and the Intel Xeon E5-2690 v3 (Haswell) processor. Users can crunch through data faster by leveraging CUDA for energy exploration applications, crash simulations, ray traced rendering, deep learning, and more. The NC24r configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads.
 
 | Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs |
 | --- | --- | --- | --- | --- | --- | --- | ---- |
@@ -47,7 +43,9 @@ Premium Storage:  Supported
 
 Premium Storage caching:  Supported
 
-NCv2-series VMs are powered by [NVIDIA Tesla P100](https://www.nvidia.com/en-us/data-center/tesla-p100/) GPUs. These GPUs can provide more than 2x the computational performance of the NC-series. Customers can take advantage of these updated GPUs for traditional HPC workloads such as reservoir modeling, DNA sequencing, protein analysis, Monte Carlo simulations, and others. The NC24rs v2 configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads.
+NCv2-series VMs are powered by [NVIDIA Tesla P100](https://www.nvidia.com/en-us/data-center/tesla-p100/) GPUs. These GPUs can provide more than 2x the computational performance of the NC-series. Customers can take advantage of these updated GPUs for traditional HPC workloads such as reservoir modeling, DNA sequencing, protein analysis, Monte Carlo simulations, and others. In addition to the GPUs, the NCv2-series VMs are also powered by Intel Xeon E5-2690 v4 (Broadwell) CPUs.
+
+The NC24rs v2 configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads.
 
 > [!IMPORTANT]
 > For this size family, the vCPU (core) quota in your subscription is initially set to 0 in each region. [Request a vCPU quota increase](../articles/azure-supportability/resource-manager-core-quotas-request.md) for this family in an [available region](https://azure.microsoft.com/regions/services/).
@@ -70,7 +68,7 @@ Premium Storage:  Supported
 
 Premium Storage caching:  Supported
 
-NCv3-series VMs are powered by [NVIDIA Tesla V100](https://www.nvidia.com/en-us/data-center/tesla-v100/) GPUs. These GPUs can provide 1.5x the computational performance of the NCv2-series. Customers can take advantage of these updated GPUs for traditional HPC workloads such as reservoir modeling, DNA sequencing, protein analysis, Monte Carlo simulations, and others. The NC24rs v3 configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads.
+NCv3-series VMs are powered by [NVIDIA Tesla V100](https://www.nvidia.com/en-us/data-center/tesla-v100/) GPUs. These GPUs can provide 1.5x the computational performance of the NCv2-series. Customers can take advantage of these updated GPUs for traditional HPC workloads such as reservoir modeling, DNA sequencing, protein analysis, Monte Carlo simulations, and others. The NC24rs v3 configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads. In addition to the GPUs, the NCv3-series VMs are also powered by Intel Xeon E5-2690 v4 (Broadwell) CPUs.
 
 > [!IMPORTANT]
 > For this size family, the vCPU (core) quota in your subscription is initially set to 0 in each region. [Request a vCPU quota increase](../articles/azure-supportability/resource-manager-core-quotas-request.md) for this family in an [available region](https://azure.microsoft.com/regions/services/).
@@ -89,15 +87,13 @@ NCv3-series VMs are powered by [NVIDIA Tesla V100](https://www.nvidia.com/en-us/
 
 ## NDv2-series (Preview)
 
-
 Premium Storage:  Supported
 
 Premium Storage caching:  Supported
 
 Infiniband: Not supported
 
-
-NDv2-series virtual machine is a new addition to the GPU family designed for the needs of the HPC, AI, and machine learning workloads. It’s powered by 8 NVIDIA Tesla V100 NVLINK interconnected GPUs and 40 Intel Skylake cores and 672 GiB of system memory. NDv2 instance provides excellent FP32 and FP64 performance for HPC and AI workloads utilizing Cuda, TensorFlow, Pytorch, Caffe, and other frameworks.
+NDv2-series virtual machine is a new addition to the GPU family designed for the needs of the HPC, AI, and machine learning workloads. It’s powered by 8 NVIDIA Tesla V100 NVLINK interconnected GPUs and 40 Intel Xeon Platinum 8168 (Skylake) cores and 672 GiB of system memory. NDv2 instance provides excellent FP32 and FP64 performance for HPC and AI workloads utilizing Cuda, TensorFlow, Pytorch, Caffe, and other frameworks.
 
 [Sign-up and get access to these machines during preview](https://aka.ms/ndv2signup).
 <br>
@@ -112,7 +108,7 @@ Premium Storage:  Supported
 
 Premium Storage caching:  Supported
 
-The ND-series virtual machines are a new addition to the GPU family designed for AI, and Deep Learning workloads. They offer excellent performance for training and inference. ND instances are powered by [NVIDIA Tesla P40](http://images.nvidia.com/content/pdf/tesla/184427-Tesla-P40-Datasheet-NV-Final-Letter-Web.pdf) GPUs. These instances provide excellent performance for single-precision floating point operations, for AI workloads utilizing Microsoft Cognitive Toolkit, TensorFlow, Caffe, and other frameworks. The ND-series also offers a much larger GPU memory size (24 GB), enabling to fit much larger neural net models. Like the NC-series, the ND-series offers a configuration with a secondary low-latency, high-throughput network through RDMA, and InfiniBand connectivity so you can run large-scale training jobs spanning many GPUs.
+The ND-series virtual machines are a new addition to the GPU family designed for AI, and Deep Learning workloads. They offer excellent performance for training and inference. ND instances are powered by [NVIDIA Tesla P40](http://images.nvidia.com/content/pdf/tesla/184427-Tesla-P40-Datasheet-NV-Final-Letter-Web.pdf) GPUs and Intel Xeon E5-2690 v4 (Broadwell) CPUs. These instances provide excellent performance for single-precision floating point operations, for AI workloads utilizing Microsoft Cognitive Toolkit, TensorFlow, Caffe, and other frameworks. The ND-series also offers a much larger GPU memory size (24 GB), enabling to fit much larger neural net models. Like the NC-series, the ND-series offers a configuration with a secondary low-latency, high-throughput network through RDMA, and InfiniBand connectivity so you can run large-scale training jobs spanning many GPUs.
 
 > [!IMPORTANT]
 > For this size family, the vCPU (core) quota per region in your subscription is initially set to 0. [Request a vCPU quota increase](../articles/azure-supportability/resource-manager-core-quotas-request.md) for this family in an [available region](https://azure.microsoft.com/regions/services/).
@@ -135,12 +131,12 @@ Premium Storage:  Not Supported
 
 Premium Storage caching:  Not Supported
 
-The NV-series virtual machines are powered by [NVIDIA Tesla M60](http://images.nvidia.com/content/tesla/pdf/188417-Tesla-M60-DS-A4-fnl-Web.pdf) GPUs and NVIDIA GRID technology for desktop accelerated applications and virtual desktops where customers are able to visualize their data or simulations. Users are able to visualize their graphics intensive workflows on the NV instances to get superior graphics capability and additionally run single precision workloads such as encoding and rendering. 
+The NV-series virtual machines are powered by [NVIDIA Tesla M60](http://images.nvidia.com/content/tesla/pdf/188417-Tesla-M60-DS-A4-fnl-Web.pdf) GPUs and NVIDIA GRID technology for desktop accelerated applications and virtual desktops where customers are able to visualize their data or simulations. Users are able to visualize their graphics intensive workflows on the NV instances to get superior graphics capability and additionally run single precision workloads such as encoding and rendering. NV-series VMs are also powered by Intel Xeon E5-2690 v3 (Haswell) CPUs.
 
 Each GPU in NV instances comes with a GRID license. This license gives you the flexibility to use an NV instance as a virtual workstation for a single user, or 25 concurrent users can connect to the VM for a virtual application scenario.
 
-| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs | Virtual Workstations | Virtual Applications | 
-| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | 
+| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs | Virtual Workstations | Virtual Applications |
+| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
 | Standard_NV6 |6 |56 |340 | 1 | 8 | 24 | 1 | 1 | 25 |
 | Standard_NV12 |12 |112 |680 | 2 | 16 | 48 | 2 | 2 | 50 |
 | Standard_NV24 |24 |224 |1440 | 4 | 32 | 64 | 4 | 4 | 100 |
@@ -153,7 +149,7 @@ Premium Storage:  Supported
 
 Premium Storage caching:  Supported
 
-The NVv3-series virtual machines are powered by [NVIDIA Tesla M60](http://images.nvidia.com/content/tesla/pdf/188417-Tesla-M60-DS-A4-fnl-Web.pdf) GPUs and NVIDIA GRID technology with Intel Broadwell CPUs. These virtual machines are targeted for GPU accelerated graphics applications and virtual desktops where customers want to visualize their data, simulate results to view, work on CAD, or render and stream content. Additionally, these virtual machines can run single precision workloads such as encoding and rendering. NVv3 virtual machines support Premium Storage and come with twice the system memory (RAM) when compared with its predecessor NV-series.  
+The NVv3-series virtual machines are powered by [NVIDIA Tesla M60](http://images.nvidia.com/content/tesla/pdf/188417-Tesla-M60-DS-A4-fnl-Web.pdf) GPUs and NVIDIA GRID technology with Intel E5-2690 v4 (Broadwell) CPUs. These virtual machines are targeted for GPU accelerated graphics applications and virtual desktops where customers want to visualize their data, simulate results to view, work on CAD, or render and stream content. Additionally, these virtual machines can run single precision workloads such as encoding and rendering. NVv3 virtual machines support Premium Storage and come with twice the system memory (RAM) when compared with its predecessor NV-series.  
 
 Each GPU in NVv3 instances comes with a GRID license. This license gives you the flexibility to use an NV instance as a virtual workstation for a single user, or 25 concurrent users can connect to the VM for a virtual application scenario.
 
@@ -165,6 +161,4 @@ Each GPU in NVv3 instances comes with a GRID license. This license gives you the
 
 1 GPU = one-half M60 card.
 
-<sup>1</sup> NVv3-series VM’s feature Intel® Hyper-Threading Technology
-
- 
+<sup>1</sup> NVv3-series VMs feature Intel Hyper-Threading Technology
