What are you looking for ?
Infinidat
Articles_top

Mellanox HDR IB Solutions Accelerate Supercomputer “Great Lakes” at University of Michigan

Deployed by Dell EMC

Mellanox Technologies, Ltd. announced that its new generation of HDR IB solutions consisting of ConnectX-6 adapters, Quantum switches, LinkX cables and software accelerate the new supercomputer at the University of Michigan.

Mellanox HDR IB Solutions Accelerate  Supercomputer "Great Lakes"

The new supercomputer named Great Lakes will enhance the university research activities and provide scalable HPC and AI infrastructure for thousands of users across the university.

The HPC and AI supercomputer was deployed by Dell EMC and includes a nearly 13,000 processor-cores unit, and a 25,600 GPU-cores unit, built in a hybrid topology, interconnected by a Mellanox smart IB network. The combined performance of both units delivers more than one petaflop of performance.

Access to HPC resources is a critical component of the University of Michigan’s rich computing ecosystem,” said Ravi Pendse, VP for IT and CIO, University of Michigan. “Investigators in emerging fields like machine learning and precision health, among many others, will see improved computing performance with Great Lakes.

“Dell EMC is thrilled to collaborate with the University of Michigan and our technology partners to bring this innovative and powerful system to such a strong community of researchers,” said Thierry Pellegrino, VP, Dell EMC HPC. “This Great Lakes cluster will offer an exceptional boost in performance, throughput and response to reduce the time needed for University of Michigan researchers to make the next big discovery in a range of disciplines from artificial intelligence to genomics and bioscience.

We are proud to see the first Mellanox HDR IB ConnectX-6 adapters and Quantum switches based supercomputer in the world,” said Gilad Shainer, VP marketing, Mellanox. “The smart In-Network Computing acceleration engines that IB enables will deliver the highest performance, efficiency and scalability for the University of Michigan users, for both HPC and AI applications.

The Great Lakes supercomputer is planned to run in full capacity and be available to the University of Michigan community in the first half of 2019.

Articles_bottom
AIC
ATTO
OPEN-E