Authors: Mark Nossokoff, Bob Sorensen
Publication Date: November 2020
Length: 4 pages
The growing proliferation of GPUs and related HPDA and HPC-based AI workloads is challenging the HPC storage architectures of conventional CPU-based designs and traditional HPC modeling/simulation workloads. Conventional HPC storage systems manage the well-understood needs of largely independent and segregated home directories and scratch files, keeping the system fully utilized for optimal performance. However, data-intensive HPDA and AI workloads, with much greater variety of heterogenous I/O profiles, are stressing the performance capabilities of the conventional storage systems and related architectures.
Hyperion Research recently closed the books on the 2018 year for the high performance computing market in our market tracking database, the QView. 2018 was another high growth year, with the server market reaching $13.7 billion, and the total market (servers, storage, software and technical support service) totaling more than $27.6 billion. The market has continued to grow at a steady rate of around 6% for the past 5 years.
May 2019 | Quick Take
DOE's recent announcement of the operational Summit supercomputer at Oak Ridge National Laboratory (ORNL) marks a shift in the global lead for supercomputer performance from China back to the United States.
June 2018 | Quick Take