Breyta

Deila með


HBv2 sizes series

HBv2-series VMs are optimized for applications that are driven by memory bandwidth, such as fluid dynamics, finite element analysis, and reservoir simulation. HBv2 VMs feature 120 AMD EPYC 7V12 processor cores, 4 GB of RAM per CPU core, and no simultaneous multithreading. Each HBv2 VM provides up to 350 GB/s of memory bandwidth, and up to 4 teraFLOPS of FP64 compute. HBv2-series VMs feature 200 Gb/sec Mellanox HDR InfiniBand. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. These VMs support Adaptive Routing and the Dynamic Connected Transport (DCT, in addition to standard RC and UD transports). These features enhance application performance, scalability, and consistency, and their usage is recommended.

Host specifications

Part Quantity
Count Units
Specs
SKU ID, Performance Units, etc.
Processor 120 - 16 vCPUs AMD EPYC 7V12 (Genoa) [x86-64]
Memory 456 GiB
Local Storage 1 Temp Disk
1 NVMe Disk
480 GiB
960 GiB
Remote Storage 8 Disks
Network 8 NICs
Accelerators None

Feature support

Premium Storage: Supported
Premium Storage caching: Supported
Live Migration: Not Supported
Memory Preserving Updates: Not Supported
Generation 2 VMs: Supported
Generation 1 VMs: Supported
Accelerated Networking: Supported
Ephemeral OS Disk: Supported
Nested Virtualization: Not Supported

Sizes in series

vCPUs (Qty.) and Memory for each size

Size Name vCPUs (Qty.) Memory (GB) Memory Bandwidth (GB/s) Base CPU Frequency (GHz) Single-core Frequency Peak (GHz) All-core Frequency Peak (GHz)
Standard_HB120rs_v2 120 456 350 2.45 3.3 3.1
Standard_HB120-96rs_v2 96 456 350 2.45 3.3 3.1
Standard_HB120-64rs_v2 64 456 350 2.45 3.3 3.1
Standard_HB120-32rs_v2 32 456 350 2.45 3.3 3.1
Standard_HB120-16rs_v2 16 456 350 2.45 3.3 3.1

VM Basics resources

Other size information

List of all available sizes: Sizes

Pricing Calculator: Pricing Calculator

Information on Disk Types: Disk Types

Next steps

Read about the latest announcements, HPC workload examples, and performance results at the Azure Compute Tech Community Blogs.

For a higher level architectural view of running HPC workloads, see High Performance Computing (HPC) on Azure.