Previous generations of virtual machine sizes
Applies to: ✔️ Linux VMs ✔️ Windows VMs ✔️ Flexible scale sets ✔️ Uniform scale sets
Tip
Try the Virtual machines selector tool to find other sizes that best fit your workload.
This section provides information on previous generations of virtual machine sizes. These sizes can still be used, but there are newer generations available.
F-series
F-series is based on the 2.4 GHz Intel Xeon® E5-2673 v3 (Haswell) processor, which can achieve clock speeds as high as 3.1 GHz with the Intel Turbo Boost Technology 2.0. This is the same CPU performance as the Dv2-series of VMs.
F-series VMs are an excellent choice for workloads that demand faster CPUs but do not need as much memory or temporary storage per vCPU. Workloads such as analytics, gaming servers, web servers, and batch processing will benefit from the value of the F-series.
ACU: 210 - 250
Premium Storage: Not Supported
Premium Storage caching: Not Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max temp storage throughput: IOPS/Read MBps/Write MBps | Max data disks/throughput: IOPS | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|
| Standard_F1 | 1 | 2 | 16 | 3000/46/23 | 4/4x500 | 2/750 |
| Standard_F2 | 2 | 4 | 32 | 6000/93/46 | 8/8x500 | 2/1500 |
| Standard_F4 | 4 | 8 | 64 | 12000/187/93 | 16/16x500 | 4/3000 |
| Standard_F8 | 8 | 16 | 128 | 24000/375/187 | 32/32x500 | 8/6000 |
| Standard_F16 | 16 | 32 | 256 | 48000/750/375 | 64/64x500 | 8/12000 |
Fs-series 1
The Fs-series provides all the advantages of the F-series, in addition to Premium storage.
ACU: 210 - 250
Premium Storage: Supported
Premium Storage caching: Supported
Ephemeral OS Disks: Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max data disks | Max cached and temp storage throughput: IOPS/MBps (cache size in GiB) | Max uncached disk throughput: IOPS/MBps | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|
| Standard_F1s | 1 | 2 | 4 | 4 | 4000/32 (12) | 3200/48 | 2/750 |
| Standard_F2s | 2 | 4 | 8 | 8 | 8000/64 (24) | 6400/96 | 2/1500 |
| Standard_F4s | 4 | 8 | 16 | 16 | 16000/128 (48) | 12800/192 | 4/3000 |
| Standard_F8s | 8 | 16 | 32 | 32 | 32000/256 (96) | 25600/384 | 8/6000 |
| Standard_F16s | 16 | 32 | 64 | 64 | 64000/512 (192) | 51200/768 | 8/12000 |
MBps = 10^6 bytes per second, and GiB = 1024^3 bytes.
1 The maximum disk throughput (IOPS or MBps) possible with a Fs series VM may be limited by the number, size, and striping of the attached disk(s). For details, see Design for high performance.
NVv2-series
Newer size recommendation: NVv3-series
The NVv2-series virtual machines are powered by NVIDIA Tesla M60 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. NVv2 virtual machines support Premium Storage and come with twice the system memory (RAM) when compared with its predecessor NV-series.
Each GPU in NVv2 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.
Ephemeral OS Disks: Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs | Virtual Workstations | Virtual Applications |
|---|---|---|---|---|---|---|---|---|---|
| Standard_NV6s_v2 | 6 | 112 | 320 | 1 | 8 | 12 | 4 | 1 | 25 |
| Standard_NV12s_v2 | 12 | 224 | 640 | 2 | 16 | 24 | 8 | 2 | 50 |
| Standard_NV24s_v2 | 24 | 448 | 1280 | 4 | 32 | 32 | 8 | 4 | 100 |
Older generations of virtual machine sizes
This section provides information on older generations of virtual machine sizes. These sizes are still supported but will not receive additional capacity. There are newer or alternative sizes that are generally available. Please refer to Sizes virtual machines in Azure to choose the VM sizes that will best fit your need.
For more information on resizing a Linux VM, see Resize a VM.
Basic A
Newer size recommendation: Av2-series
Premium Storage: Not Supported
Premium Storage caching: Not Supported
The basic tier sizes are primarily for development workloads and other applications that don't require load balancing, auto-scaling, or memory-intensive virtual machines.
| Size – Size\Name | vCPU | Memory | NICs (Max) | Max temporary disk size | Max. data disks (1023 GB each) | Max. IOPS (300 per disk) |
|---|---|---|---|---|---|---|
| A0\Basic_A0 | 1 | 768 MB | 2 | 20 GB | 1 | 1x300 |
| A1\Basic_A1 | 1 | 1.75 GB | 2 | 40 GB | 2 | 2x300 |
| A2\Basic_A2 | 2 | 3.5 GB | 2 | 60 GB | 4 | 4x300 |
| A3\Basic_A3 | 4 | 7 GB | 2 | 120 GB | 8 | 8x300 |
| A4\Basic_A4 | 8 | 14 GB | 2 | 240 GB | 16 | 16x300 |
Standard A0 - A4 using CLI and PowerShell
In the classic deployment model, some VM size names are slightly different in CLI and PowerShell:
- Standard_A0 is ExtraSmall
- Standard_A1 is Small
- Standard_A2 is Medium
- Standard_A3 is Large
- Standard_A4 is ExtraLarge
A-series
Newer size recommendation: Av2-series
ACU: 50-100
Premium Storage: Not Supported
Premium Storage caching: Not Supported
| Size | vCPU | Memory: GiB | Temp storage (HDD): GiB | Max data disks | Max data disk throughput: IOPS | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|
| Standard_A0 1 | 1 | 0.768 | 20 | 1 | 1x500 | 2/100 |
| Standard_A1 | 1 | 1.75 | 70 | 2 | 2x500 | 2/500 |
| Standard_A2 | 2 | 3.5 | 135 | 4 | 4x500 | 2/500 |
| Standard_A3 | 4 | 7 | 285 | 8 | 8x500 | 2/1000 |
| Standard_A4 | 8 | 14 | 605 | 16 | 16x500 | 4/2000 |
| Standard_A5 | 2 | 14 | 135 | 4 | 4x500 | 2/500 |
| Standard_A6 | 4 | 28 | 285 | 8 | 8x500 | 2/1000 |
| Standard_A7 | 8 | 56 | 605 | 16 | 16x500 | 4/2000 |
1 The A0 size is over-subscribed on the physical hardware. For this specific size only, other customer deployments may impact the performance of your running workload. The relative performance is outlined below as the expected baseline, subject to an approximate variability of 15 percent.
A-series - compute-intensive instances
Newer size recommendation: Av2-series
ACU: 225
Premium Storage: Not Supported
Premium Storage caching: Not Supported
The A8-A11 and H-series sizes are also known as compute-intensive instances. The hardware that runs these sizes is designed and optimized for compute-intensive and network-intensive applications, including high-performance computing (HPC) cluster applications, modeling, and simulations. The A8-A11 series uses Intel Xeon E5-2670 @ 2.6 GHZ and the H-series uses Intel Xeon E5-2667 v3 @ 3.2 GHz.
| Size | vCPU | Memory: GiB | Temp storage (HDD): GiB | Max data disks | Max data disk throughput: IOPS | Max NICs |
|---|---|---|---|---|---|---|
| Standard_A8 1 | 8 | 56 | 382 | 32 | 32x500 | 2 |
| Standard_A9 1 | 16 | 112 | 382 | 64 | 64x500 | 4 |
| Standard_A10 | 8 | 56 | 382 | 32 | 32x500 | 2 |
| Standard_A11 | 16 | 112 | 382 | 64 | 64x500 | 4 |
1 For MPI applications, dedicated RDMA backend network is enabled by FDR InfiniBand network, which delivers ultra-low-latency and high bandwidth.
Note
The A8 – A11 VMs are planned for retirement on 3/2021. We strongly recommend not creating any new A8 – A11 VMs. Please migrate any existing A8 – A11 VMs to newer and powerful high-performance computing VM sizes such as H, HB, HC, HBv2 as well as general purpose compute VM sizes such as D, E, and F for better price-performance. For more information, see HPC Migration Guide.
D-series
Newer size recommendation: Dav4-series, Dv4-series and Ddv4-series
ACU: 160-250 1
Premium Storage: Not Supported
Premium Storage caching: Not Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max temp storage throughput: IOPS/Read MBps/Write MBps | Max data disks/throughput: IOPS | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|
| Standard_D1 | 1 | 3.5 | 50 | 3000/46/23 | 4/4x500 | 2/500 |
| Standard_D2 | 2 | 7 | 100 | 6000/93/46 | 8/8x500 | 2/1000 |
| Standard_D3 | 4 | 14 | 200 | 12000/187/93 | 16/16x500 | 4/2000 |
| Standard_D4 | 8 | 28 | 400 | 24000/375/187 | 32/32x500 | 8/4000 |
1 VM Family can run on one of the following CPU's: 2.2 GHz Intel Xeon® E5-2660 v2, 2.4 GHz Intel Xeon® E5-2673 v3 (Haswell) or 2.3 GHz Intel XEON® E5-2673 v4 (Broadwell)
D-series - memory optimized
Newer size recommendation: Dav4-series, Dv4-series and Ddv4-series
ACU: 160-250 1
Premium Storage: Not Supported
Premium Storage caching: Not Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max temp storage throughput: IOPS/Read MBps/Write MBps | Max data disks/throughput: IOPS | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|
| Standard_D11 | 2 | 14 | 100 | 6000/93/46 | 8/8x500 | 2/1000 |
| Standard_D12 | 4 | 28 | 200 | 12000/187/93 | 16/16x500 | 4/2000 |
| Standard_D13 | 8 | 56 | 400 | 24000/375/187 | 32/32x500 | 8/4000 |
| Standard_D14 | 16 | 112 | 800 | 48000/750/375 | 64/64x500 | 8/8000 |
1 VM Family can run on one of the following CPU's: 2.2 GHz Intel Xeon® E5-2660 v2, 2.4 GHz Intel Xeon® E5-2673 v3 (Haswell) or 2.3 GHz Intel XEON® E5-2673 v4 (Broadwell)
Preview: DC-series
Newer size recommendation: DCsv2-series
Premium Storage: Supported
Premium Storage caching: Supported
Ephemeral OS Disks: Supported
The DC-series uses the latest generation of 3.7GHz Intel XEON E-2176G Processor with SGX technology, and with the Intel Turbo Boost Technology can go up to 4.7GHz.
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max data disks | Max cached and temp storage throughput: IOPS / MBps (cache size in GiB) | Max uncached disk throughput: IOPS / MBps | Max NICs / Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|
| Standard_DC2s | 2 | 8 | 100 | 2 | 4000 / 32 (43) | 3200 /48 | 2 / 1500 |
| Standard_DC4s | 4 | 16 | 200 | 4 | 8000 / 64 (86) | 6400 /96 | 2 / 3000 |
Important
DC-series VMs are generation 2 VMs and only support Gen2 images.
DS-series
Newer size recommendation: Dasv4-series, Dsv4-series and Ddsv4-series
ACU: 160-250 1
Premium Storage: Supported
Premium Storage caching: Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max data disks | Max cached and temp storage throughput: IOPS/MBps (cache size in GiB) | Max uncached disk throughput: IOPS/MBps | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|
| Standard_DS1 | 1 | 3.5 | 7 | 4 | 4000/32 (43) | 3200/32 | 2/500 |
| Standard_DS2 | 2 | 7 | 14 | 8 | 8000/64 (86) | 6400/64 | 2/1000 |
| Standard_DS3 | 4 | 14 | 28 | 16 | 16000/128 (172) | 12800/128 | 4/2000 |
| Standard_DS4 | 8 | 28 | 56 | 32 | 32000/256 (344) | 25600/256 | 8/4000 |
1 VM Family can run on one of the following CPU's: 2.2 GHz Intel Xeon® E5-2660 v2, 2.4 GHz Intel Xeon® E5-2673 v3 (Haswell) or 2.3 GHz Intel XEON® E5-2673 v4 (Broadwell)
DS-series - memory optimized
Newer size recommendation: Dasv4-series, Dsv4-series and Ddsv4-series
ACU: 160-250 1,2
Premium Storage: Supported
Premium Storage caching: Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max data disks | Max cached and temp storage throughput: IOPS/MBps (cache size in GiB) | Max uncached disk throughput: IOPS/MBps | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|
| Standard_DS11 | 2 | 14 | 28 | 8 | 8000/64 (72) | 6400/64 | 2/1000 |
| Standard_DS12 | 4 | 28 | 56 | 16 | 16000/128 (144) | 12800/128 | 4/2000 |
| Standard_DS13 | 8 | 56 | 112 | 32 | 32000/256 (288) | 25600/256 | 8/4000 |
| Standard_DS14 | 16 | 112 | 224 | 64 | 64000/512 (576) | 51200/512 | 8/8000 |
1 The maximum disk throughput (IOPS or MBps) possible with a DS series VM may be limited by the number, size and striping of the attached disk(s). For details, see Design for high performance. 2 VM Family can run on one of the following CPU's: 2.2 GHz Intel Xeon® E5-2660 v2, 2.4 GHz Intel Xeon® E5-2673 v3 (Haswell) or 2.3 GHz Intel XEON® E5-2673 v4 (Broadwell)
Ls-series
Newer size recommendation: Lsv2-series
The Ls-series offers up to 32 vCPUs, using the Intel® Xeon® processor E5 v3 family. The Ls-series gets the same CPU performance as the G/GS-Series and comes with 8 GiB of memory per vCPU.
The Ls-series does not support the creation of a local cache to increase the IOPS achievable by durable data disks. The high throughput and IOPS of the local disk makes Ls-series VMs ideal for NoSQL stores such as Apache Cassandra and MongoDB which replicate data across multiple VMs to achieve persistence in the event of the failure of a single VM.
ACU: 180-240
Premium Storage: Supported
Premium Storage caching: Not Supported
Ephemeral OS Disks: Supported
| Size | vCPU | Memory (GiB) | Temp storage (GiB) | Max data disks | Max temp storage throughput (IOPS/MBps) | Max uncached disk throughput (IOPS/MBps) | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|
| Standard_L4s | 4 | 32 | 678 | 16 | 20000/200 | 5000/125 | 2/4000 |
| Standard_L8s | 8 | 64 | 1388 | 32 | 40000/400 | 10000/250 | 4/8000 |
| Standard_L16s | 16 | 128 | 2807 | 64 | 80000/800 | 20000/500 | 8/16000 |
| Standard_L32s 1 | 32 | 256 | 5630 | 64 | 160000/1600 | 40000/1000 | 8/20000 |
The maximum disk throughput possible with Ls-series VMs may be limited by the number, size, and striping of any attached disks. For details, see Design for high performance.
1 Instance is isolated to hardware dedicated to a single customer.
GS-series
Newer size recommendation: Easv4-series, Esv4-series, Edsv4-series and M-series
ACU: 180 - 240 1
Premium Storage: Supported
Premium Storage caching: Supported
Ephemeral OS Disks: Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max data disks | Max cached and temp storage throughput: IOPS / MBps (cache size in GiB) | Max uncached disk throughput: IOPS/MBps | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|
| Standard_GS1 | 2 | 28 | 56 | 8 | 10000/100 (264) | 5000/ 125 | 2/2000 |
| Standard_GS2 | 4 | 56 | 112 | 16 | 20000/200 (528) | 10000/ 250 | 2/4000 |
| Standard_GS3 | 8 | 112 | 224 | 32 | 40000/400 (1056) | 20000/ 500 | 4/8000 |
| Standard_GS4 3 | 16 | 224 | 448 | 64 | 80000/800 (2112) | 40000/1000 | 8/16000 |
| Standard_GS5 2, 3 | 32 | 448 | 896 | 64 | 160000/1600 (4224) | 80000/2000 | 8/20000 |
1 The maximum disk throughput (IOPS or MBps) possible with a GS series VM may be limited by the number, size and striping of the attached disk(s). For details, see Design for high performance.
2 Isolation feature retired on 2/28/2022. For information, see the retirement announcement.
3 Constrained core sizes available.
G-series
Newer size recommendation: Eav4-series, Ev4-series and Edv4-series and M-series
ACU: 180 - 240
Premium Storage: Not Supported
Premium Storage caching: Not Supported
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Max temp storage throughput: IOPS/Read MBps/Write MBps | Max data disks/throughput: IOPS | Max NICs/Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|
| Standard_G1 | 2 | 28 | 384 | 6000/93/46 | 8/8x500 | 2/2000 |
| Standard_G2 | 4 | 56 | 768 | 12000/187/93 | 16/16x500 | 2/4000 |
| Standard_G3 | 8 | 112 | 1536 | 24000/375/187 | 32/32x500 | 4/8000 |
| Standard_G4 | 16 | 224 | 3072 | 48000/750/375 | 64/64x500 | 8/16000 |
| Standard_G5 1 | 32 | 448 | 6144 | 96000/1500/750 | 64/64x500 | 8/20000 |
1 Isolation feature retired on 2/28/2022. For information, see the retirement announcement.
NV-series
Newer size recommendation: NVv3-series and NVv4-series
The NV-series virtual machines 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. 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.
Premium Storage: Not Supported
Premium Storage caching: Not Supported
Live Migration: Not Supported
Memory Preserving Updates: Not Supported
| 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 |
1 GPU = one-half M60 card.
NC series
Newer size recommendation: NC T4 v3-series
NC-series VMs are powered by the NVIDIA Tesla K80 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.
Premium Storage: Not Supported
Premium Storage caching: Not Supported
Live Migration: Not Supported
Memory Preserving Updates: Not Supported
VM Generation Support: Generation 1
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs |
|---|---|---|---|---|---|---|---|
| Standard_NC6 | 6 | 56 | 340 | 1 | 12 | 24 | 1 |
| Standard_NC12 | 12 | 112 | 680 | 2 | 24 | 48 | 2 |
| Standard_NC24 | 24 | 224 | 1440 | 4 | 48 | 64 | 4 |
| Standard_NC24r* | 24 | 224 | 1440 | 4 | 48 | 64 | 4 |
1 GPU = one-half K80 card.
*RDMA capable
NCv2 series
Newer size recommendation: NC T4 v3-series and NC V100 v3-series
NCv2-series VMs are powered by NVIDIA 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.
Premium Storage: Supported
Premium Storage caching: Supported
Live Migration: Not Supported
Memory Preserving Updates: Not Supported
VM Generation Support: Generation 1 and 2
Ephemeral OS Disks: Supported
For this VM series, the vCPU (core) quota in your subscription is initially set to 0 in each region. Request a vCPU quota increase for this series in an available region.
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs |
|---|---|---|---|---|---|---|---|---|
| Standard_NC6s_v2 | 6 | 112 | 736 | 1 | 16 | 12 | 20000/200 | 4 |
| Standard_NC12s_v2 | 12 | 224 | 1474 | 2 | 32 | 24 | 40000/400 | 8 |
| Standard_NC24s_v2 | 24 | 448 | 2948 | 4 | 64 | 32 | 80000/800 | 8 |
| Standard_NC24rs_v2* | 24 | 448 | 2948 | 4 | 64 | 32 | 80000/800 | 8 |
1 GPU = one P100 card.
*RDMA capable
ND series
Newer size recommendation: NDv2-series and NC V100 v3-series
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 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.
Premium Storage: Supported
Premium Storage caching: Supported
Live Migration: Not Supported
Memory Preserving Updates: Not Supported
VM Generation Support: Generation 1 and 2
Ephemeral OS Disks: Supported
For this VM series, the vCPU (core) quota per region in your subscription is initially set to 0. Request a vCPU quota increase for this series in an available region.
| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs |
|---|---|---|---|---|---|---|---|---|
| Standard_ND6s | 6 | 112 | 736 | 1 | 24 | 12 | 20000/200 | 4 |
| Standard_ND12s | 12 | 224 | 1474 | 2 | 48 | 24 | 40000/400 | 8 |
| Standard_ND24s | 24 | 448 | 2948 | 4 | 24 | 32 | 80000/800 | 8 |
| Standard_ND24rs* | 24 | 448 | 2948 | 4 | 96 | 32 | 80000/800 | 8 |
1 GPU = one P40 card.
*RDMA capable
Next steps
Learn more about how Azure compute units (ACU) can help you compare compute performance across Azure SKUs.
Feedback
Submit and view feedback for