Skip to content

Commit 6fd3aaa

Browse files
v-dirichardsv-dirichards
authored
Merge pull request #211917 from LouisBerner/v-lberner-ADO_1989216
New PR for VM size updates
2 parents 0ca9f3e + 1718f08 commit 6fd3aaa

File tree

2 files changed

+52
-14
lines changed

2 files changed

+52
-14
lines changed

‎articles/databox-online/azure-stack-edge-gpu-virtual-machine-sizes.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@ author: alkohli
77
ms.service: databox
88
ms.subservice: edge
99
ms.topic: conceptual
10-
ms.date: 08/09/2022
10+
ms.date: 09/21/2022
1111
ms.author: alkohli
1212
#Customer intent: As an IT admin, I need to understand how to create and manage virtual machines (VMs) on my Azure Stack Edge Pro device by using APIs, so that I can efficiently manage my VMs.
1313
---

‎includes/azure-stack-edge-gateway-supported-vm-sizes.md

Lines changed: 51 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22
author: alkohli
33
ms.service: databox
44
ms.topic: include
5-
ms.date: 08/09/2022
5+
ms.date: 09/22/2022
66
ms.author: alkohli
77
---
88

@@ -46,7 +46,7 @@ For more information, see [Dv2 and DSv2-series](../articles/virtual-machines/dv2
4646

4747
These sizes are supported for GPU VMs on your device and are optimized for compute-intensive GPU-accelerated applications, for example, inferencing workloads. The GPU VM that you deploy should match the GPU type on your Azure Stack Edge device.
4848

49-
#### For Nvidia's Tesla T4 GPU.
49+
#### NVIDIA Tesla T4 GPU
5050

5151
| Size  | vCPU  | Memory (GiB)  | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | GPU  | GPU memory (GiB)  | Max NICs  |
5252
|-----------------------|-------|---------------|---------------------|--------------------------------|-------------------------------------|----------------------------------|------|-------------------|-----------|
@@ -56,7 +56,7 @@ These sizes are supported for GPU VMs on your device and are optimized for compu
5656

5757
For more information, see [NCasT4_v3-series](../articles/virtual-machines/nct4-v3-series.md).
5858

59-
#### For Nvidia's A2 Tensor Core GPU
59+
#### NVIDIA A2 Tensor Core GPU
6060

6161

6262
| Size  | vCPU  | Memory (GiB)  | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | GPU  | GPU memory (GiB)  | Max NICs  |
@@ -83,6 +83,7 @@ These series are optimized for computational workloads and run on Intel Xeon pro
8383
| **Standard_F4s**  | 4  | 8  | 16  | 2000  | 16000  | 2300  | 16  | 4  |
8484
| **Standard_F8s**  | 8  | 16  | 32  | 2000  | 32000  | 2300  | 32  | 8  |
8585
| **Standard_F16s**  | 16  | 32  | 64  | 2000  | 64000  | 2300  | 64  | 8 |
86+
| **Standard_F4s_v1**| 4 | 4 | 32 | 6400 | 8000 | 8000 | 8 | 2 |
8687

8788
For more information, see [Fsv2-series](../articles/virtual-machines/fsv2-series.md).
8889

@@ -92,18 +93,55 @@ The high-performance network (HPN) virtual machines are tailored for workloads t
9293

9394
#### HPN DSv2-series
9495

95-
| Size  | vCPU  | Memory (GiB)  | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | Max data disks  | Max NICs  |
96+
| Size  | vCPU  | Memory (GiB)  | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | Max data disks  | Max NICs<sup>1</sup>  |
9697
|-------------------|-------|---------------|---------------------|--------------------------------|-------------------------------------|----------------------------------|-----------------|-----------|
97-
| **Standard_DS2_v2_HPN**  | 2  | 7  | 14  | 2000  | 8000  | 2300  | 8  | 2  |
98-
| **Standard_DS3_v2_HPN**  | 4  | 14  | 28  | 2000  | 16000  | 2300  | 16  | 4  |
99-
| **Standard_DS4_v2_HPN**  | 8  | 28  | 56  | 2000  | 32000  | 2300  | 32  | 8  |
98+
| **Standard_DS2_v2_HPN**  | 2  | 7  | 14  | 2000  | 8000  | 2300  | 8  | 14  |
99+
| **Standard_DS3_v2_HPN**  | 4  | 14  | 28  | 2000  | 16000  | 2300  | 16  | 14  |
100+
| **Standard_DS4_v2_HPN**  | 8  | 28  | 56  | 2000  | 32000  | 2300  | 32  | 14  |
101+
102+
<sup>1</sup>Windows Server 2016 Datacenter VHD has a limit of 8 NICs for all HPN VM sizes.
100103

101104
#### HPN F-series
102105

103-
| Size  | vCPU  | Memory (GiB) | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | Max data disks  | Max NICs  |
106+
| Size  | vCPU  | Memory (GiB) | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | Max data disks  | Max NICs<sup>1</sup>  |
104107
|----------------|---------|--------------|---------------------|--------------------------------|-------------------------------------|----------------------------------|-----------------|-----------|
105-
| **Standard_F2s_HPN**  | 2  | 4  | 8  | 2000  | 8000  | 2300  | 8  | 4  |
106-
| **Standard_F4s_HPN**  | 4  | 8  | 16  | 2000  | 16000  | 2300  | 16  | 4  |
107-
| **Standard_F8s_HPN**  | 8  | 16  | 32  | 2000  | 32000  | 2300  | 32  | 8  |
108-
| **Standard_F16s_HPN**  | 16  | 32  | 64  | 2000  | 64000  | 2300  | 64  | 8 |
109-
| **Standard_F12_HPN**  | 12 | 64  | 64 | 1000  | 48000  | 500  | 64  | 8  |
108+
| **Standard_F2s_HPN**  | 2  | 4  | 8  | 2000  | 8000  | 2300  | 8  | 14  |
109+
| **Standard_F4s_HPN**  | 4  | 8  | 16  | 2000  | 16000  | 2300  | 16  | 14  |
110+
| **Standard_F8s_HPN**  | 8  | 16  | 32  | 2000  | 32000  | 2300  | 32  | 14  |
111+
| **Standard_F12s_HPN** | 12 | 24 | 48 | 2000 | 48000 | 2300 | 64 | 14
112+
| **Standard_F16s_HPN**  | 16  | 32  | 64  | 2000  | 64000  | 2300  | 64  | 14 |
113+
| **Standard_F12_HPN**  | 12 | 24  | 64 | 1000  | 48000  | 500  | 64  | 14  |
114+
115+
<sup>1</sup>Windows Server 2016 Datacenter VHD has a limit of 8 NICs for all HPN VM sizes.
116+
117+
#### HPN and NVIDIA Tesla T4 GPU series
118+
119+
| Size  | vCPU  | Memory (GiB) | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | Max data disks  | Max NICs<sup>1</sup>  | GPU | GPU memory (GiB) |
120+
|----------------|---------|--------------|---------------------|--------------------------------|-------------------------------------|----------------------------------|-----------------|-----------|------|------|
121+
| **Standard_DS2_v2_HPN_T4_v3**  | 2  | 7  | 14  | 2000  | 8000  | 2300  | 8  | 14  |1 |16 |
122+
| **Standard_DS3_v2_HPN_T4_v3**  | 4  | 14  | 28  | 2000  | 16000  | 2300  | 16  | 14  |1 |16 |
123+
| **Standard_DS4_v2_HPN_T4_v3**  | 8  | 28  | 56  | 2000  | 32000  | 2300  | 32  | 14  |1 |16 |
124+
| **Standard_F2s_HPN_T4_v3**  | 2  | 4  | 8  | 2000  | 8000  | 2300  | 8  | 14  |1 |16 |
125+
| **Standard_F4s_HPN_T4_v3**  | 4  | 8  | 16  | 2000  | 16000  | 2300  | 16  | 14  |1 |16 |
126+
| **Standard_F8s_HPN_T4_v3**  | 8  | 16  | 32  | 2000  | 32000  | 2300  | 32  | 14  |1 |16 |
127+
| **Standard_F12s_HPN_T4_v3**  | 12  | 24  | 48  | 2000  | 48000  | 2300  | 64  | 14  |1 |16 |
128+
| **Standard_F16s_HPN_T4_v3**  | 16  | 32  | 64  | 2000  | 64000  | 2300  | 64  | 14  |1 |16 |
129+
| **Standard_F12_HPN_T4_v3**  | 12  | 24  | 64  | 1000  | 48000  | 500  | 64  | 14  |1 |16 |
130+
131+
<sup>1</sup>Windows Server 2016 Datacenter VHD has a limit of 8 NICs for all HPN VM sizes.
132+
133+
#### HPN and NVIDIA Tesla A2 Tensor Core GPU series
134+
135+
| Size  | vCPU  | Memory (GiB) | Temp storage (GiB)  | Max OS disk throughput (IOPS)  | Max temp storage throughput (IOPS)  | Max data disk throughput (IOPS)  | Max data disks  | Max NICs<sup>1</sup>  | GPU | GPU memory (GiB) |
136+
|----------------|---------|--------------|---------------------|--------------------------------|-------------------------------------|----------------------------------|-----------------|-----------|------|------|
137+
| **Standard_DS2_v2_HPN_A2**  | 2  | 7  | 14  | 2000  | 8000  | 2300  | 8  | 14  |1 |16 |
138+
| **Standard_DS3_v2_HPN_A2**  | 4  | 14  | 28  | 2000  | 16000  | 2300  | 16  | 14  |1 |16 |
139+
| **Standard_DS4_v2_HPN_A2**  | 8  | 28  | 56  | 2000  | 32000  | 2300  | 32  | 14  |1 |16 |
140+
| **Standard_F2s_HPN_A2**  | 2  | 4  | 8  | 2000  | 8000  | 2300  | 8  | 14  |1 |16 |
141+
| **Standard_F4s_HPN_A2**  | 4  | 8  | 16  | 2000  | 16000  | 2300  | 16  | 14  |1 |16 |
142+
| **Standard_F8s_HPN_A2**  | 8  | 16  | 32  | 2000  | 32000  | 2300  | 32  | 14  |1 |16 |
143+
| **Standard_F12s_HPN_A2**  | 12  | 24  | 48  | 2000  | 48000  | 2300  | 64  | 14  |1 |16 |
144+
| **Standard_F16s_HPN_A2**  | 16  | 32  | 64  | 2000  | 64000  | 2300  | 64  | 14  |1 |16 |
145+
| **Standard_F12_HPN_A2**  | 12  | 64  | 64  | 1000  | 48000  | 500  | 64  | 14  |1 |16 |
146+
147+
<sup>1</sup>Windows Server 2016 Datacenter VHD has a limit of 8 NICs for all HPN VM sizes.

0 commit comments

Comments
 (0)