NERSC Enhances PDSF, Genepool Computing Capabilities
Linux cluster expansion speeds data access and analysis
January 3, 2014
Christmas came early for users of the Parallel Distributed Systems Facility (PDSF) and Genepool systems at Department of Energy’s National Energy Research Scientific Computer Center (NERSC).
Throughout November members of NERSC’s Computational Systems Group were busy expanding the Linux computing resources that support PDSF’s high-energy and nuclear physics computations and the bioinformatics data processing needs of the Joint Genome Institute’s (JGI) Genepool users.
While the expansion increases the number of processors on PDSF by 15%, the actual increase is larger because the new computers come with a faster network, so scientists can access their data much faster, noted Lisa Gerhardt, PDSF Computing Consultant in the NERSC User Services Group.
“Additionally, several nodes have been set aside to upgrade the interactive nodes,” she explained. “This should offer users faster access to their data and will greatly facilitate scientific analysis with PDSF.”
For JGI users, the new addition expands the cluster’s interactive capabilities, according to Doug Jacobsen, Bioinformatics Computing Consultant in the NERSC User Services Group.
“This is critical because Genepool is the principal computing resource for JGI and the interactive resources are used by their bioinformaticians to analyze calculation results and develop new algorithms and software,” he explained. “This enables JGI to be a strong innovative contributor to the scientific community, not only with production-level sequences but also ground-breaking methodologies.”
Initially implemented in 2012, the Linux cluster is designed to be flexible enough to meet the current needs of NERSC’s user constituencies and expandable enough to meet future needs. The layered cluster model provides a framework through which the workloads of multiple Linux computational systems and support servers can be processed on a unified hardware and software platform.
"With the speed at which we can deploy more compute power for less electrical power, this is a big win for PDSF and Genepool users," said NERSC's Larry Pezzaglia, who was instrumental in installing and integrating the new equipment. "Historically, we've managed our midrange systems separately. But with the expansion I worked to identify similarities in order to build a unified system that is more efficient to manage and still provides the same level of service to our customers."
About NERSC and Berkeley Lab
The National Energy Research Scientific Computing Center (NERSC) is a U.S. Department of Energy Office of Science User Facility that serves as the primary high-performance computing center for scientific research sponsored by the Office of Science. Located at Lawrence Berkeley National Laboratory, the NERSC Center serves more than 6,000 scientists at national laboratories and universities researching a wide range of problems in combustion, climate modeling, fusion energy, materials science, physics, chemistry, computational biology, and other disciplines. Berkeley Lab is a DOE national laboratory located in Berkeley, California. It conducts unclassified scientific research and is managed by the University of California for the U.S. DOE Office of Science. »Learn more about computing sciences at Berkeley Lab.