SC19: Pavilion Provides NVMe-oF Storage for SCinet’19 HPC Simulations
Hyperparallel flash array shown at event
This is a Press Release edited by StorageNewsletter.com on November 25, 2019 at 2:19 pmPavilion Data Systems, Inc. will be part of a project called SCinet’19 to provide the high-performance storage infrastructure with NVMe and NVMe-oF of HPC organizations.
Spearheaded by Caltech Professor Harvey Newman, SCinet’19 showcases how people and organizations gain insights and knowledge in gene sequencing, social and economic analysis, and other big data applications. The work was showed at SC19, taking place in Denver, CO, November 17-21, 2019.
More about SCinet’19 project in this blog post
Pavilion HFA relies on 72 Western Digital Ultrastar DC SN200 NVMe SSDs
Pavilion’s Hyperparallel HFA was selected by Professor Newman and team to meet the IO/s and bandwidth requirements of NVMe-oF for SCinet’19.
“As we searched for a storage system that could accommodate hundreds of gigabytes per second of throughput at microsecond latencies that could accommodate our space and power requirements, Pavilion’s HFA was an ideal fit,” said Harvey B. Newman, Marvin L. Goldberger Professor of Physics, Caltech.
“We are looking forward to our work with Caltech and the broader HPC community to advance networking and storage technologies,” added VR Satish, CTO, Pavilion. “Some of the most sophisticated AI, ML and deep learning projects take place through collaboration at events such as Supercomputing 2019 and with platforms like SCinet’19. We are pleased to be selected by Caltech as their storage platform for this event as we push the envelope of performance, density and low-latency.”
SCinet’19 features the latest advances in Tb/s networks, intelligent global operations and monitoring systems, workflow optimization methodologies with real-time analytics, and long-distance data transfer methods, with tools and storage designs to meet the challenges faced by programs in high energy physics, astrophysics, climate and other fields of data-intensive science.