Skip to content

Commit 1e4c740

Browse files
mattmcinnesmattmcinnes
committed
Adding the rest of the N-series
1 parent 88e4435 commit 1e4c740

30 files changed

+498
-9
lines changed

‎articles/virtual-machines/sizes/gpu-accelerated/includes/nc-a100-v4-series-specs.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,4 +16,4 @@ ms.custom: include file
1616
| Memory | 220 - 880<sup>GiB | |
1717
| Data Disks | 8 - 32<sup>Disks | 30000 - 120000<sup>IOPS</sup> / 1000 - 4000<sup>MBps |
1818
| Network | 2 - 8 <sup>NICs | 20000 - 80000<sup>Mbps |
19-
| Accelerators | 1 - 4 NVIDIA A100 (PCIe) | 80<sup>GiB </sup>/ GPU |
19+
| Accelerators | 1 - 4<sup>GPUs</sup> | NVIDIA A100 (PCIe) 80<sup>GiB </sup> <br> 80 - 320<sup>GiB</sup> per VM|

‎articles/virtual-machines/sizes/gpu-accelerated/includes/nc-series-specs.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,4 +16,4 @@ ms.custom: include file
1616
| Memory | 56 - 224<sup>GiB | |
1717
| Data Disks | 24 - 64<sup>Disks | |
1818
| Network | 1 - 4<sup>NICs | |
19-
| Accelerators | 1 - 4 [NVIDIA Tesla K80](https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/tesla-product-literature/Tesla-K80-BoardSpec-07317-001-v05.pdf) | 12<sup>GiB </sup>/ GPU |
19+
| Accelerators | 1 - 4<sup>GPUs</sup> | [NVIDIA Tesla K80](https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/tesla-product-literature/Tesla-K80-BoardSpec-07317-001-v05.pdf) 12<sup>GiB </sup> <br> 12 - 48<sup>GiB</sup> per VM |

‎articles/virtual-machines/sizes/gpu-accelerated/includes/ncads-h100-v5-series-specs.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,4 +16,4 @@ ms.custom: include file
1616
| Memory | 320 - 640<sup>GiB | |
1717
| Data Disks | 8 - 16<sup>Disks | 100000 - 240000<sup>IOPS</sup> / 3000 - 7000<sup>MBps |
1818
| Network | 2 - 4<sup>NICs | 40000 - 80000<sup>Mbps </sup> |
19-
| Accelerators | 1 - 2 NVIDIA H100 NVL | 94<sup>GiB </sup>/ GPU |
19+
| Accelerators | 1 - 2<sup>GPUs</sup> | NVIDIA H100 NVL 94<sup>GiB </sup> <br> 94 - 188<sup>GiB</sup> per VM|

‎articles/virtual-machines/sizes/gpu-accelerated/includes/ncast4-v3-series-specs.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,4 +16,4 @@ ms.custom: include file
1616
| Memory | 28 - 440<sup>GiB | |
1717
| Data Disks | 8 - 32<sup>Disks | 20000 - 80000<sup>IOPS</sup> / 200 - 800<sup>MBps |
1818
| Network | 2 - 8 <sup>NICs | 8000 - 32000<sup>Mbps |
19-
| Accelerators | 1 - 4 NVIDIA Tesla T4 | 16<sup>GiB </sup>/ GPU |
19+
| Accelerators | 1 - 4<sup>GPUs</sup> | NVIDIA Tesla T4 16<sup>GiB </sup> <br> 16 - 64<sup>GiB</sup> per VM|

‎articles/virtual-machines/sizes/gpu-accelerated/includes/ncv2-series-specs.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,4 +16,4 @@ ms.custom: include file
1616
| Memory | 112 - 448<sup>GiB | |
1717
| Data Disks | 12 - 32<sup>Disks | 20000 - 80000<sup>IOPS</sup> / 200 - 800<sup>MBps |
1818
| Network | 4 - 8 <sup>NICs | |
19-
| Accelerators | 1 - 4 NVIDIA Tesla P100 | 16<sup>GiB </sup>/ GPU |
19+
| Accelerators | 1 - 4<sup>GPUs</sup> | NVIDIA Tesla P100 16<sup>GiB </sup> <br> 16 - 64<sup>GiB</sup> per VM|

‎articles/virtual-machines/sizes/gpu-accelerated/includes/ncv3-series-specs.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,4 +16,4 @@ ms.custom: include file
1616
| Memory | 112 - 448<sup>GiB | |
1717
| Data Disks | 12 - 32<sup>Disks | 20000 - 80000<sup>IOPS</sup> / 200 - 800<sup>MBps |
1818
| Network | 4 - 8 <sup>NICs | |
19-
| Accelerators | 1 - 4 NVIDIA Tesla V100 | 16<sup>GiB </sup>/ GPU |
19+
| Accelerators | 1 - 4 <sup>GPUs</sup> | NVIDIA Tesla V100 16<sup>GiB </sup> <br> 16 - 64<sup>GiB</sup> per VM |

‎articles/virtual-machines/sizes/gpu-accelerated/includes/nd-a100-v4-series-specs.md

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,19 @@
1+
---
2+
title: ND-A100_v4-series specs include
3+
description: Include file containing specifications of ND-A100_v4-series VM sizes.
4+
services: virtual-machines
5+
author: mattmcinnes
6+
ms.topic: include
7+
ms.service: virtual-machines
8+
ms.subservice: sizes
9+
ms.date: 04/18/2024
10+
ms.author: mattmcinnes
11+
ms.custom: include file
12+
---
13+
| Part | Quantity <br><sup>Count <sup>Units | Specs <br><sup>SKU ID, Performance <sup>Units</sup>, etc. |
14+
|---|---|---|
15+
| Processor | 96<sup>vCores | AMD Epyc™ 7V12 (Rome) |
16+
| Memory | 900<sup>GiB | |
17+
| Data Disks | 32<sup>Disks | 80000<sup>IOPS</sup> / 800<sup>MBps |
18+
| Network | 8 <sup>NICs | 24000<sup>Mbps |
19+
| Accelerators | 8<sup>GPUs</sup> | NVIDIA A100 (NVLink) 40<sup>GiB </sup> <br> 320<sup>GiB</sup> per VM|

‎articles/virtual-machines/sizes/gpu-accelerated/includes/nd-a100-v4-series-summary.md

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
---
2+
title: ND-A100_v4-series summary include
3+
description: Include file containing a summary of the ND-A100_v4-series size family.
4+
services: virtual-machines
5+
author: mattmcinnes
6+
ms.topic: include
7+
ms.service: virtual-machines
8+
ms.subservice: sizes
9+
ms.date: 04/18/2024
10+
ms.author: mattmcinnes
11+
ms.custom: include file
12+
---
13+
The ND A100 v4 series virtual machine(VM) is a new flagship addition to the Azure GPU family. It's designed for high-end Deep Learning training and tightly coupled scale-up and scale-out HPC workloads. The ND A100 v4 series starts with a single VM and eight NVIDIA Ampere A100 40GB Tensor Core GPUs. ND A100 v4-based deployments can scale up to thousands of GPUs with an 1.6 TB/s of interconnect bandwidth per VM. Each GPU within the VM is provided with its own dedicated, topology-agnostic 200 GB/s NVIDIA Mellanox HDR InfiniBand connection. These connections are automatically configured between VMs occupying the same VM scale set, and support GPUDirect RDMA. Each GPU features NVLINK 3.0 connectivity for communication within the VM, and the instance is backed by 96 physical 2nd-generation AMD Epyc™ 7V12 (Rome) CPU cores. These instances provide excellent performance for many AI, ML, and analytics tools that support GPU acceleration 'out-of-the-box,' such as TensorFlow, Pytorch, Caffe, RAPIDS, and other frameworks. Additionally, the scale-out InfiniBand interconnect is supported by a large set of existing AI and HPC tools that are built on NVIDIA's NCCL2 communication libraries for seamless clustering of GPUs.

‎articles/virtual-machines/sizes/gpu-accelerated/includes/nd-h100-v5-series-specs.md

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,19 @@
1+
---
2+
title: ND H100 v5-series specs include
3+
description: Include file containing specifications of ND H100 v5-series VM sizes.
4+
services: virtual-machines
5+
author: mattmcinnes
6+
ms.topic: include
7+
ms.service: virtual-machines
8+
ms.subservice: sizes
9+
ms.date: 04/18/2024
10+
ms.author: mattmcinnes
11+
ms.custom: include file
12+
---
13+
| Part | Quantity <br><sup>Count <sup>Units | Specs <br><sup>SKU ID, Performance <sup>Units</sup>, etc. |
14+
|---|---|---|
15+
| Processor | 96<sup>vCores | Intel® Xeon® Scalable (Sapphire Rapids) |
16+
| Memory | 1900<sup>GiB | |
17+
| Data Disks | 32<sup>Disks | 40800<sup>IOPS</sup> / 612<sup>MBps |
18+
| Network | 8<sup>NICs | 80000<sup>Mbps |
19+
| Accelerators | 8<sup>GPUs</sup> | NVIDIA H100 (NVLink) 80<sup>GiB </sup> <br> 640<sup>GiB</sup> per VM |

‎articles/virtual-machines/sizes/gpu-accelerated/includes/nd-h100-v5-series-summary.md

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
---
2+
title: ND H100 v5-series summary include
3+
description: Include file containing a summary of the ND H100 v5-series size family.
4+
services: virtual-machines
5+
author: mattmcinnes
6+
ms.topic: include
7+
ms.service: virtual-machines
8+
ms.subservice: sizes
9+
ms.date: 04/18/2024
10+
ms.author: mattmcinnes
11+
ms.custom: include file
12+
---
13+
The ND H100 v5 series virtual machine (VM) is a new flagship addition to the Azure GPU family. It’s designed for high-end Deep Learning training and tightly coupled scale-up and scale-out Generative AI and HPC workloads. The ND H100 v5 series starts with a single VM and eight NVIDIA H100 Tensor Core GPUs. ND H100 v5-based deployments can scale up to thousands of GPUs with 3.2Tb/s of interconnect bandwidth per VM. Each GPU within the VM is provided with its own dedicated, topology-agnostic 400 Gb/s NVIDIA Quantum-2 CX7 InfiniBand connection. These connections are automatically configured between VMs occupying the same virtual machine scale set, and support GPUDirect RDMA. Each GPU features NVLINK 4.0 connectivity for communication within the VM, and the instance is backed by 96 physical 4th Gen Intel Xeon Scalable processor cores. These instances provide excellent performance for many AI, ML, and analytics tools that support GPU acceleration ‘out-of-the-box,’ such as TensorFlow, Pytorch, Caffe, RAPIDS, and other frameworks. Additionally, the scale-out InfiniBand interconnect is supported by a large set of existing AI and HPC tools that are built on NVIDIA’s NCCL communication libraries for seamless clustering of GPUs.

0 commit comments

Comments
 (0)