Todd Muirhead talks with Lan Vu about how NVIDIA vGPU allows vSPhere to share GPUs amount multiple VMs by using either time sliced vGPU or Multi Instance vGPU profiles. The differences in performance across a variety of workloads is discussed with recommendations about how to choose the right profile based on workload to maximize the benefits of vGPU and MIG vGPU.
Links to additional resources –
Blog and Whitepaper – https://blogs.vmware.com/performance/2022/06/vsphere7-vgpu-vs-mig-perf.html
Accelerate Machine Learning in vSphere Using GPUs Hands on Lab – https://pathfinder.vmware.com/v3/activity/vsphere_ml_gpu_hol
VMware Explore 2022 Session – NVIDIA+VMware Bare Metal Machine Learning Performance on vSphere – https://event.vmware.com/flow/vmware/explore2022us/content/page/catalog?tab.contentcatalogtabs=1627421929827002v7oA&search=uday
Extreme Performance Series 2022 Video Blogs – https://blogs.vmware.com/performance/2022/08/extreme-performance-series-2022-video-blogs.html
Extreme Performance Series 2021 Video Blogs – https://blogs.vmware.com/performance/2021/09/extreme-performance-video-blog-series.html