In this video from the Mellanox booth at SC13, DK Panda from Ohio State presents: High Performance Scalable and Fault Resilient MPI and GPUDirect RDMA.
See more talks from the Mellanox Booth.
[SPONSORED GUEST CONTENT] Scheduled Ethernet is emerging as a viable alternative to InfiniBand for AI networking. Why? Because of its ability to offer comparable performance but with greater flexibility and cost-effectiveness.
ANSYS® FLUENT® ON MICROSOFT® AZURE® HBv3 VIRTUAL MACHINES FEATURING AMD 3D V-CACHE™ TECHNOLOGY COMPUTATIONAL FLUID DYNAMICS AMD EPYC™7003 Series Processors with AMD 3D V-Cache deliver outstanding scale-out performance running Ansys® Fluent® on Microsoft® Azure® HBv3 virtual machines. This performance brief concludes that Azure HBv3 VMs powered by AMD EPYC 7003 Series processors with AMD […]
Dr. Panda’s lectures/tutorials are worth attending!