At the SC19 supercomputing show in Denver, Microsoft and NVIDIA announced the new NDv2, the world’s largest GPU-accelerated cloud-based supercomputer on Azure. NDv2 VMs are designed specifically for most demanding distributed HPC, AI, and machine learning workloads. These VMs come with 8 NVIDIA Tesla V100 NVLink interconnected GPUs each with 32 GB of HBM2 memory, 40 non-hyperthreaded cores from the Intel Xeon Platinum 8168 processor, and 672 GiB of system memory. They also feature 100 Gigabit EDR InfiniBand from Mellanox with support for standard OFED drivers and all MPI types and versions.
Microsoft mentioned that they were able to train BERT, a popular conversational AI model, in just three hour using these new VMs.
“Now you can open up an instance, you grab one of the stacks … in the container, you launch it, on Azure, and you’re doing science,” said Jensen Huang, NVIDIA CEO. “It’s really quite fantastic, This puts a supercomputer in the hands of every scientist in the world,” Huang added.