The ND v5 H200 series virtual machine (VM) is designed to deliver exceptional performance for AI and high-performance computing (HPC) workloads. These VMs leverage the power of the NVIDIA H200 Tensor Core GPU, which offer a 76% increase in High Bandwidth Memory over the H100 GPUs to deliver higher performance on state-of-the-art Generative AI models. With 141 GB of high-speed memory, and 4.8 TB/s of memory bandwidth, the H200 GPU can handle larger datasets and more complex models, making it ideal for generative AI and scientific computing.
The ND H200 v5 series starts with a single VM and eight NVIDIA H200 Tensor Core GPUs, interconnected with 900 GB/s NVLink. ND H200 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.
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.
1Temp disk speed often differs between RR (Random Read) and RW (Random Write) operations. RR operations are typically faster than RW operations. The RW speed is usually slower than the RR speed on series where only the RR speed value is listed.
Storage capacity is shown in units of GiB or 1024^3 bytes. When you compare disks measured in GB (1000^3 bytes) to disks measured in GiB (1024^3), remember that capacity numbers given in GiB may appear smaller. For example, 1023 GiB = 1098.4 GB.
Disk throughput is measured in input/output operations per second (IOPS) and MBps where MBps = 10^6 bytes/sec.
Storage capacity is shown in units of GiB or 1024^3 bytes. When you compare disks measured in GB (1000^3 bytes) to disks measured in GiB (1024^3), remember that capacity numbers given in GiB may appear smaller. For example, 1023 GiB = 1098.4 GB.
Disk throughput is measured in input/output operations per second (IOPS) and MBps where MBps = 10^6 bytes/sec.
Data disks can operate in cached or uncached modes. For cached data disk operation, the host cache mode is set to ReadOnly or ReadWrite. For uncached data disk operation, the host cache mode is set to None.
Expected network bandwidth is the maximum aggregated bandwidth allocated per VM type across all NICs, for all destinations. For more information, see Virtual machine network bandwidth.
Upper limits aren't guaranteed. Limits offer guidance for selecting the right VM type for the intended application. Actual network performance depends on several factors including network congestion, application loads, and network settings. For information on optimizing network throughput, see Optimize network throughput for Azure virtual machines.
To achieve the expected network performance on Linux or Windows, you may need to select a specific version or optimize your VM. For more information, see Bandwidth/Throughput testing (NTTTCP).
Accelerator (GPUs, FPGAs, etc.) information for each size