12 Sep 2022

UH joins Jetstream2 in immensely fast computing for research, AI

University of Hawaiʻi researchers and students will soon have access to some of the most powerful computing power at their fingertips thanks to a joint project with several universities around the country and the National Science Foundation (NSF).

UH is part of a newly available national cloud computing resource that allows researchers to access powerful cloud-based environments on demand to explore and understand immense amounts of data from a laptop or tablet.

The NSF supported project Jetstream2 is an update to the widely used Jetstream cloud computing system. Jetstream2 provides eight petaFLOPS of virtual supercomputing power, allowing researchers to simplify data analysis, boost scientific discovery and leverage artificial intelligence (AI) tools.

“A 1 petaFLOPS (PFLOPS) computer system is capable of performing one quadrillion (1015) floating-point operations per second. The rate 1 PFLOPS is equivalent to 1,000 TFLOPS. To match what a 1 PFLOPS computer system can do in just one second, you’d have to perform one calculation every second for 31,688,765 years,” according to Indiana University.

“Our participation in Jetstream2 ensures our researchers, faculty and students have access to a national-scale cloud computing resource platform to effectively collaborate with their peers and research partners and leverage state-of-the-art technology and techniques to drive science forward,” said Garret Yoshimi, UH vice president for information technology services. “We are grateful to NSF for their generous support in making Jetstream2 a reality for our UH community.”

Related: UH part of $10M cloud computing coalition, June 4, 2020

With five computational systems, Jetstream2 spans the nation with partners at UH, Indiana University, Arizona State University, Cornell University and the Texas Advanced Computing Center. The system is designed to be easily used by researchers who have limited experience with cloud computing and to serve smaller academic communities with no previous access to such resources.

The 24/7 user-friendly cloud computing environment provides virtual infrastructure that allows for more access to high-end technologies that enable deep learning and AI techniques. The Jetstream2 system is easy to expand and reconfigure and supports diverse modes of on-demand access. In addition, it extends a broad range of hardware and services that include larger and faster storage systems, graphics processing units, large memory nodes, virtual clusters and much more.

Jetstream2 also supports experimentation and teaching. The system will serve more students than any other NSF-funded cyberinfrastructure resource, better equipping them to fully participate in the evolving STEM workforce.

Additional partnerships with the University of Arizona, Johns Hopkins University and University Corporation for Atmospheric Research will contribute to Jetstream2’s unparalleled usability and support for a broad range of scientific efforts.

Researchers may request several allocation types through NSF’s Xtreme Science and Engineering Discovery Environment.