SC2000 — Platform Computing Corporation
announced today that NASA has acquired Platform’s LSF MultiCluster software to
manage the combined scientific and engineering computing workloads at the
Glenn Research Center in Cleveland, OH and the Langley Research Center in
Hampton, VA. LSF MultiCluster extends the reach of Platform’s market leading
Load Sharing Facility (LSF) distributed resource management (DRM) capabilities
by coordinating the scheduling of the computing resources between the NASA
centers in a fail safe, lights-out, peer-to-peer operational relationship,
without requiring the addition of new hardware and without having to learn or
implement any new job submission language or queuing system syntax.
Prior to implementing LSF MultiCluster, users at the centers were
constrained by the capacity of their local compute farms. Although local
resources were optimized through the use of the LSF Standard software, users
often needed more processors, or processors with different performance
characteristics, for jobs that exceeded the capabilities of their home SGI and
Linux clusters. LSF MultiCluster expands the available capacity by enabling
transparent user access to resources at both centers.
“Our CFD and structural analysts applications are used in very complex
aerodynamics and propulsion projects that require prompt access to significant
computing resources,” explains Isaac Lopez, Parallel Computing Testbed
Development and Application Manager at the Glenn Center. “Platform’s LSF
MultiCluster increases our available computing capacity by harnessing the
computing resources from the Glenn and Langley centers. We were pleasantly
surprised to find that installation took less than one day for both sites, and
we are now exploring opportunities for inter-agency meta computing as well.”
“LSF clusters computers together and enables users to utilize their
combined computing resources so researchers can explore significantly more
sophisticated computing scenarios,” explains Phil Weaver, President and Chief
Operating Officer, Platform Computing. “This is an example of how powerful LSF
can be even in complex distributed computing environments like NASA’s.”
About LSF MultiCluster
The LSF MultiCluster design ensures that all individual computer clusters
retain their independence in terms of management and job prioritization. This
means that when more computing resources are needed to meet job requirements,
LSF MultiCluster requests permission from the distributed clusters without
imposing arbitrary demands. What makes this possible is LSF’s focus on
providing scalable, policy-based prioritization that dynamically matches
computing resources with processing requirements.
LSF and LSF MultiCluster are also designed so that end users are not
required to implement any code or programming to initiate the resource sharing
capabilities. Clusters are managed and organized so that the users do not need
to have any specific knowledge about the IT-infrastructure.
About NASA Glenn and NASA Langley
Since 1941, the men and women of the NASA John H. Glenn Research Center at
Lewis field and the NASA Langley Research Center in Hampton, VA have been the
pioneers and innovators whose work has expanded horizons and opened frontiers
for our explorers in air and space. These Centers define and develop
propulsion, space electrical power, structural integrity and communications
technologies for NASA’s aeronautics and space missions. Visit our web sites
at: http://www.grc.nasa.gov and http://www.larc.nasa.gov.
About Platform Computing Inc.
Established in 1992, Platform Computing Inc. is a strategic partner for
Sun, IBM, Cisco, HP, SGI and Compaq to enable some of the world’s most
creative companies to achieve better results by allowing them to harness the
untapped potential of their people, products and processes. Platform’s
software solutions, LSF Suite and SiteAssure Suite give both engineering and
Internet environments unprecedented control over their mission-critical
hardware, software and network resources. With Platform solutions, customers
are able to bridge the worlds of networks, systems and applications, and
realize higher availability and performance to achieve their business
objectives. For more information, please visit http://www.platform.com.
