GPU optimized virtual machine sizes

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 as well as storage throughput and network bandwidth for each size in this grouping.

  • NC, NCv2, and ND sizes are optimized for compute-intensive and network-intensive applications and algorithms, including CUDA- and OpenCL-based applications and simulations, AI, and Deep Learning.
  • NV sizes are optimized and designed for remote visualization, streaming, gaming, encoding, and VDI scenarios utilizing frameworks such as OpenGL and DirectX.

NC instances

The NC instances are powered by the NVIDIA Tesla K80 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.

Size vCPU Memory: GiB Temp storage (SSD) GiB GPU Maximum data disks
Standard_NC6 6 56 380 1 24
Standard_NC12 12 112 680 2 48
Standard_NC24 24 224 1440 4 64
Standard_NC24r* 24 224 1440 4 64

1 GPU = one-half K80 card.

*RDMA capable

NCv2 instances

NCv2 instances are the next generation of the NC-series machines, powered by NVIDIA Tesla P100 GPUs. These GPUs can provide more than 2x the computational performance of the current 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. Like the NC-series, the NCv2-series offers a configuration with 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 for this family in an available region.

Size vCPU Memory: GiB Temp storage (SSD) GiB GPU Maximum data disks
Standard_NC6_v2 6 112 336 1 12
Standard_NC12_v2 12 224 672 2 24
Standard_NC24_v2 24 448 1344 4 32
Standard_NC24r_v2* 24 1448 1344 4 32

1 GPU = one P100 card.

*RDMA capable

ND instances

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 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.

Important

For this size family, the vCPU (core) quota per region in your subscription is initially set to 0. Request a vCPU quota increase for this family in an available region.

Size vCPU Memory: GiB Temp storage (SSD) GiB GPU Maximum data disks
Standard_ND6 6 112 336 1 12
Standard_ND12 12 224 672 2 24
Standard_ND24 24 448 1344 4 32
Standard_ND24r* 24 1448 1344 4 32

1 GPU = one P40 card.

*RDMA capable

NV instances

The NV instances are powered by NVIDIA Tesla M60 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.

Size vCPU Memory: GiB Temp storage (SSD) GiB GPU Maximum data disks
Standard_NV6 6 56 380 1 24
Standard_NV12 12 112 680 2 48
Standard_NV24 24 224 1440 4 64

1 GPU = one-half M60 card.

Size table definitions

  • Storage capacity is shown in units of GiB or 1024^3 bytes. When comparing disks measured in GB (1000^3 bytes) to disks measured in GiB (1024^3) remember that capacity numbers given in GiB may appear smaller. For example, 1023 GiB = 1098.4 GB
  • Disk throughput is measured in input/output operations per second (IOPS) and MBps where MBps = 10^6 bytes/sec.
  • Data disks can operate in cached or uncached modes. For cached data disk operation, the host cache mode is set to ReadOnly or ReadWrite. For uncached data disk operation, the host cache mode is set to None.
  • If you want to get the best performance for your VMs, you should limit the number of data disks to 2 disks per vCPU.
  • Expected network bandwidth is the maximum aggregated bandwidth allocated per VM type across all NICs, for all destinations. Upper limits are not guaranteed, but are intended to provide guidance for selecting the right VM type for the intended application. Actual network performance will depend on a variety of factors including network congestion, application loads, and network settings. For information on optimizing network throughput, see Optimizing network throughput for Windows and Linux. To achieve the expected network performance on Linux or Windows, it may be necessary to select a specific version or optimize your VM. For more information, see How to reliably test for virtual machine throughput.

  • † 16 vCPU performance will consistently reach the upper limit in an upcoming release.

Supported operating systems and drivers

NC, NCv2, and ND instances - NVIDIA Tesla drivers

OS Driver
Windows Server 2016 385.54 (.exe)
Windows Server 2012 R2 385.54 (.exe)

Note

Tesla driver download links are current at time of publication. For the latest drivers, visit the NVIDIA website.

NV instances - NVIDIA GRID drivers

OS Driver
Windows Server 2016 385.41 (.exe)
Windows Server 2012 R2 385.41 (.exe)

Note

Microsoft redistributes NVIDIA GRID driver installers for NV VMs. Install only these GRID drivers on Azure NV VMs. These drivers include licensing for GRID Virtual GPU Software in Azure.

For driver installation and verification steps, see N-series driver setup for Windows.

Deployment considerations

  • For availability of N-series VMs, see Products available by region.

  • N-series VMs can only be deployed in the Resource Manager deployment model.

  • When creating an N-series VM using the Azure portal, on the Basics blade, select a VM disk type of HDD. To choose an available N-series size, on the Size blade, click View all.

  • NC and NV VMs do not support VM disks that are backed by Azure Premium storage.

  • If you want to deploy more than a few N-series VMs, consider a pay-as-you-go subscription or other purchase options. If you're using an Azure free account, you can use only a limited number of Azure compute cores.

  • You might need to increase the cores quota (per region) in your Azure subscription, and increase the separate quota for NC, NCv2, ND, or NV cores. To request a quota increase, open an online customer support request at no charge. Default limits may vary depending on your subscription category.

  • One VM image you can deploy on N-series VMs is the Azure Data Science Virtual Machine. The Data Science Virtual Machine preinstalls and configures many popular data science and deep learning tools. It also preinstalls NVIDIA Tesla GPU drivers.

Other sizes

Next steps

Learn more about how Azure compute units (ACU) can help you compare compute performance across Azure SKUs.