By Rob Mitchum // December 2, 2014
A new, $6.6 million project to ease the introduction of researchers to high-performance computing resources will incorporate Globus services for research data management and publication. Jetstream, a National Science Foundation-funded cloud-computing infrastructure, hopes to serve as a new computational “on ramp” for the growing number of scientists in fields such as biology, the humanities, and social sciences that increasingly require advanced computing power for their research.
Researchers will use Jetstream to create “virtual machines” that look and feel like their lab workstation or home computer, but are capable of harnessing thousands of times the computing power. These virtual machines will use Globus for transferring and sharing data among researchers and between campus data storage and national computing resources, and the new Globus data publication service to preserve virtual machines with unique identifiers, enabling the sharing of results, reproducibility of analyses, and new analyses of published research data.
Hosted at the Indiana University Pervasive Technology Institute and the Texas Advanced Computing Center, Jetstream is part of a new push for “interactive” supercomputing, with lower barriers of entry and improved flexibility for new HPC users. In addition to providing virtual machines, Jetstream will allow for more “on-demand” use of NSF computing resources, avoiding the long wait times created by traditional batch queuing systems.
“In the atmosphere, a jet stream is the border between two different masses of air,” said Craig Stewart, PTI executive director and associate dean for research technologies at Indiana University, in an NSF press release. “The new Jetstream cloud system will operate at the border between the existing NSF-funded cyberinfrastructure and thousands of researchers and research students who will be new to use of NSF XD program resources. Jetstream will give researchers access to cloud computing and data analysis resources interactively, when they need them.”
To meet exponentially increasing demand, the National Science Foundation has spent nearly $1 billion in the last five years to build new resources for researchers across the country. In the last 15 months alone, NSF grants have laid the foundation for two major research computing projects involving the Computation Institute: the experimental cloud computing testbed Chameleon and the open science data resource Wrangler, which will also use Globus services.
“We are gratified to see the growing adoption of Globus by innovative projects such as Jetstream and Wrangler,” said Ian Foster, director of the Computation Institute at the University of Chicago and Argonne National Laboratory. “Researchers in all disciplines are increasingly turning to cloud computing as a cost-effective means of scaling their work. Globus provides the connective tissue that bridges campus computing and national cyberinfrastructure, simplifying the tedious but necessary data management tasks required to take advantage of this hybrid environment.”
In addition to data transfer and publication, Globus will also provide other services to make Jetstream more user-friendly and collaborative. Globus provides an easy authentication system that allows researchers to access Jetstream and related resources using only their campus ID. Users can also form Globus groups to organize research teams spread across multiple institutions, facilitating collaboration and the shared use of workflows built with Jetstream capabilities.
“As high performance computing has evolved over the years, its importance to science and engineering has grown,” said Irene Qualters, division director for Advanced Cyberinfrastructure at NSF. “Bridges and Jetstream will help researchers today, but also point to the rich future of computational science. With these systems, we continue to push the boundaries of computing so that researchers in all fields can solve critical, and previously intractable, problems.”