ND96asr_v4 by Microsoft Azure
(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | azure |
Server ID | Standard_ND96asr_v4 |
Name | ND96asr_v4 |
API Reference | Standard_ND96asr_v4 |
Description | NDASv4_A100 family [AMD processor] [Premium Storage capable] [RDMA capable] 96 vCPUs |
Family | NDASv4_A100 |
Hw Virt | - |
Status | active |
Observed At | 2026-07-28T00:16:49.158283 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
East US (US) / eastus | 5.9833 USD/h | 27.197 USD/h |
West US 2 (US) / westus2 | 5.9833 USD/h | 27.197 USD/h |
South Central US (US) / southcentralus | 7.1786 USD/h | 32.63 USD/h |
Italy North (IT) / italynorth | 7.7785 USD/h | 35.357 USD/h |
West Europe (NL) / westeurope | 7.7785 USD/h | 35.357 USD/h |
Processor
vCPUs | 96 |
Hypervisor | Microsoft Hyper-V |
CPU Allocation | Dedicated |
CPU Architecture | x86_64 |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 900 GiB |
| GPU | |
|---|---|
GPU Count | 8 |
GPU Memory Min | 0 MiB |
GPU Memory Total | 0 MiB |
GPUs |
| STORAGE | |
|---|---|
Storage Size | 10650 GB |
Storages |
|
| NETWORK | |
|---|---|
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
A high-density GPU accelerated platform with dedicated AMD processors, massive local storage, and low-latency RDMA networking for parallel workloads.
Microsoft Azure ND96asr_v4 is a specialized GPU-accelerated virtual machine from the NDASv4_A100 family, designed for high-performance computing and deep learning. Built on the x86_64 architecture with AMD processors, it features 96 dedicated vCPUs, 900.0 GB of system memory, and 8.0 integrated GPUs. The instance is equipped with 10,650 GB of local storage, supports Premium Storage, and features RDMA networking capabilities for low-latency cluster communication. Managed via Microsoft Hyper-V, this server provides a dense hardware profile optimized for multi-node scaling. It is best suited for large-scale artificial intelligence training, complex simulations, and intensive data science workloads.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | GPUs |
|---|
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | GPUs |
|---|