News
NPACI ROCKS Open-Source Toolkit Improves Speed and Ease of Use in Cluster Configuration
Published March 18, 2001
NPACI Rocks Cluster Toolkit (http://rocks.npaci.edu/), a set of open-source enhancements for managing Linux-based clusters, has been improved and upgraded to version 2.0, which involves structural changes and greatly decreases the time required to put together a cluster. A full Linux distribution based on RedHat 7.0, NPACI Rocks simplifies the installation and maintenance of commodity clusters and has already proved itself on nearly 10 cluster systems.
NPACI Rocks has been used to build and install the new Meteor cluster at the San Diego Supercomputer Center (SDSC) as well as several other clusters at the University of California, San Diego (UCSD), forming the start of a campus cluster grid environment. NPACI Rocks has also been used to establish clusters at the Pacific Northwest National Laboratory, Nothwestern University, the University of Texas, and Caltech, and is being used by the Grid Physics Network (GriPhyN) project.
"We are prototyping so-called Tier2 computing facilities for the GriPhyN project," said Paul Avery, lead scientist for the NSF-funded GriPhyN project and University of Florida professor of physics. "As a computing grid, these sites will provide roughly one-third of the cycles needed by high-energy physicists to analyze data coming from experiments at CERN's Large Hadron Collider. We're using the NPACI Rocks Toolkit because it gives us the ability to manage cluster configurations in a straightforward and flexible manner. It's current level of automation put us well on the path of being able to quickly adapt computing configurations to the changing and emerging needs of our experiments."
Scientists have been using nodes on SDSC's Meteor cluster with great success almost non-stop since its inception. Kim Baldridge, director of the National Biomedical Computational Resource (NBCR) at SDSC and adjunct professor of chemistry at UCSD and UCSD graduate student Laura Gregerson have been running calculations on a molecule called hydroxymethyl acylfulvene (HMAF).
"HMAF is an anti-tumor drug in phase-three clinical trials, and we're trying to determine its mechanism of action in the cell," Baldridge said. "Rocks has allowed us to do the computations on the Meteor cluster and get the results much more quickly. Typically, we'd be working on traditional architectures that are smaller and jobs would take much longer or we wouldn't be able to do them at all because of memory and disk limitations."
NPACI Rocks 2.0 is targeted to the non-cluster expert, according to Greg Bruno, a researcher in the Distributed Computing group at SDSC. It is heavily automated - only a handful of simple steps are required to bring up a full-featured cluster. A Web page greets users with questions that capture all the customization needed for the front-end machine. Everything is captured in one file, fed to the installation, and configured automatically. In addition to essential cluster tools, NPACI Rocks installs the Maui Scheduler, Portable Batch System, and other tools in a complete production environment.
"We've built about eight different clusters with the Rocks 1.0 toolkit and have gotten great feedback from users," said Phil Papadopoulos, group leader for Distributed Computing at SDSC. "We've made some important changes to greatly simplify initial installation, have upgraded to RedHat 7.0, and have integrated some important new tools, such as automatic integration of compute nodes, simplified management of Myrinet networks, and better integration of UC Berkeley's SSL-based remote execution tool."
The SDSC Distributed Computing group has been collaborating with the UC Berkeley Millennium Project, led by computer science professor David Culler, to develop NPACI Rocks 2.0. The NPACI Rocks Cluster Toolkit was built with an SDSC-developed tool, which merges the RedHat 7.0 base distribution with all of RedHat's security advisory packages plus their bug fixes. The toolkit also integrates cluster-enabling packages developed by SDSC and UC Berkeley, as well as other standard open-source clustering packages. Compaq Computer has supported the Meteor and Rocks efforts through its loan program.
More information on NPACI Rocks, including software downloads and instructions for requesting a CD-ROM or downloading a bootable CD image, is available on the Web at http://rocks.npaci.edu/.
SDSC (http://www.sdsc.edu) is an organized research unit of the University of California, San Diego, and the leading-edge site of the National Partnership for Advanced Infrastructure (NPACI) ( http://www.npaci.edu/). SDSC is funded by the National Science Foundation through NPACI and other federal agencies, the State and University of California, and private organizations. For additional information about SDSC and NPACI, contact David Hart, dhart@sdsc.edu, 858-534-8314.
The National Partnership for Advanced Computational Infrastructure (NPACI) unites 46 universities and research institutions to build the computational environment for tomorrow's scientific discovery. Led by UC San Diego and the San Diego Supercomputer Center (SDSC), NPACI is funded by the National Science Foundation's Partnerships for Advanced Computational Infrastructure (PACI) program and receives additional support from the State and University of California, other government agencies, and partner institutions. The NSF PACI program also supports the National Computational Science Alliance. For additional information about SDSC and NPACI, see http://www.npaci.edu/ or contact David Hart, dhart@sdsc.edu, 858-534-8314.
Contact: David Hart, SDSC, dhart@sdsc.edu, 858-534-8314