How Mellanox SHARP technology speeds Ai workloads

Mellanox Scalable Hierarchical Aggregation and Reduction Protocol (SHARP) technology improves upon the performance of MPI operations by offloading collective operations from the CPU to the switch network, and by eliminating the need to send data multiple times between endpoints. This innovative approach decreases the amount of data traversing the network as aggregation nodes are reached, and dramatically reduces the MPI operations time. Implementing collective communication algorithms in the network also has additional benefits, such as freeing up valuable CPU resources for computation rather than using them to process communication.”

Video: Mellanox HDR InfiniBand Speeds HPC and Ai Applications with SHARP Technologies

In this video, Gilad Shainer from Mellanox describes how the company’s newly available HDR 200 Gigabit/sec InfiniBand solutions can speed up HPC and Ai applications. “We are proud to see the first Mellanox HDR InfiniBand ConnectX-6 adapters and Quantum switches based supercomputer in the world,” said Gilad Shainer, vice president of marketing at Mellanox Technologies. “The smart In-Network Computing acceleration engines that InfiniBand enables will deliver the highest performance, efficiency and scalability for the University of Michigan users, for both HPC and AI applications.”