IBM says its General Parallel File System (GPFS) has an important role to play in major commercial and industrial computational Grids around the world.
A large-scale Grid environment built using GPFS was named Best Commercial Application in the Fourth Annual High Performance Bandwidth Challenge at last week's Supercomputing Conference 2003, "demonstrating the important role that GPFS will play as major Grid installations are deployed around the world in industry and commercial environments," IBM says.
With GPFS, "it's now possible to rapidly transport huge amounts of data around the world," Big Blue says.
GPFS is a high-performance shared-disk file system that provides data access from all nodes in a Linux or UNIX cluster environment. Parallel and serial applications can readily access shared files using standard UNIX file system interfaces, and the same file can be accessed concurrently from multiple nodes. GPFS provides high availability through logging and replication, and can be configured for failover from both disk and server malfunctions. GPFS is currently deployed in clusters for applications like weather simulations, engineering design, seismic analysis, digital content creation and distribution, and financial modeling.
GPFS is also being implemented at the National Center for Supercomputing Applications (NCSA) and the San Diego Supercomputing Center, both part of the TeraGrid, a National Science Foundation-funded effort to build and deploy the world's largest, most comprehensive distributed infrastructure for open scientific research.
The High Performance Bandwidth Challenge asks contestants from the scientific and engineering research communities to demonstrate emerging techniques or applications that consume large amounts of network resources. IBM and the San Diego Supercomputer Center (SDSC) won using GPFS in a large-scale Grid environment spanning several sites and long distances. Using GPFS, each machine in the distributed system has the same view of the file systems and can access the same files simultaneously across the TeraGrid wide area network.
"This is first time we have used GPFS at multiple locations over the TeraGrid network," says Rob Pennington, director of NCSA's Computing and Data Management Directorate. "We have now proven that machines scattered across the country can be connected through a cyberinfrastructure like the TeraGrid and work as one machine. GPFS is an important component in creating this virtual machine."
"We were extremely pleased with the performance achieved in this distributed file system, which we believe heralds a new paradigm for Grid computing," states Phil Andrews, SDSC's director of High Performance Computing. "In this approach, data transfers across a wide area network are completely transparent to the user, avoiding any changes to their normal mode of operation."
At the conference, SDSC exhibited a cluster of 40 Intel Itanium 2-based IBM eServer xSeries systems, each connected to a Gigabit Ethernet LAN, which is connected via a Force10 switch over the SCinet network.
During the demonstration, GPFS was extended beyond the individual machine rooms at the two centers and using IBM servers on the TeraGrid and distributed among the SDSC and NCSA booths on the show floor, at SDSC (University of California) and at NCSA (University of Illinois in Urbana-Champaign). The demonstration showed how TeraGrid disk servers at both centers move data across the TeraGrid network to compute nodes in the booths, where the data could then be used by scientific applications.
Altair and HP Partner on High Performance Technical Computing
HP has agreed to resell Altair Engineering's PBS Pro distributed computing technology across its entire product line.
The companies will also work together to optimize the performance of PBS Pro on HP hardware, and deploy the integrated high-performance technical computing solution within client computing environments.
Altair's PBS Pro Grid technology lets customers create a virtual pool of computing resources and then intelligently schedule computational workload across that virtual pool.
"PBS Pro continues to gain acceptance in our core high-performance technical computing product marketplace," says Bruce Toal, HP's director of HPTC Marketing Solutions.
Singapore Computer Systems, University Of Melbourne Partner on Grid Research
Singapore Computer Systems has signed a Memorandum of Understanding with The Grid Computing and Distributed Systems (GRIDS) laboratory at The University of Melbourne, Australia.
SCS will collaborate with the GRIDS laboratory on Gridbus technology and provide resources and engineers to further develop Gridbus into production-quality software, which SCS can then deploy in its various Grid computing projects. The GRIDS laboratory will act as advisors to the project.
"We are excited about this opportunity to collaborate with Singapore Computer Systems in advancing the adoption of Grid computing," states Rajkumar Buyya, program leader and director of Grid Computing and Distributed Systems at the GRIDS laboratory. "We have collaborated informally with SCS for quite some time now, and the formal adoption of Gridbus technologies by SCS' Grid computing team will give SCS a competitive edge by being able to deploy one of world's leading Grid computing softwares."
SCS is deploying Grid technologies for the National Grid Pilot Platform Project, the first phase of a national cyberinfrastructure project that connects compute resources in Singapore via a high-speed network.
The MOU includes joint research and development on Grid computing technologies, sharing of Grid technologies via seminars and training events to be conducted in Singapore, and the co-development of Gridbus software.
AG Toolkit 2.0 and inSORS IG2.0 Found Interoperable
inSORS Integrated Communications and Argonne National Laboratory's Mathematics and Computer Science Division have confirmed the interoperability of AG (Access Grid) Toolkit 2.0 and inSORS IG2.0.
The development supports the increasing demand for collaborative application tools, supporting group-to-group collaboration across the Grid, and incorporating multimedia large-format displays, presentation and interactive environments and interfaces to Grid middleware and visualization environments, inSORS and Argonne say.
The technical milestone addresses the Grid community's interest in the open-source framework of the AG Toolkit release, combined with priority commercial features such as ease of use, application enhancements, and support packages.
Some of the specialized inSORS features include improved audio and video applications, a data-sharing application called IGPix, an integrated text chat, remote control for cameras, an integrated scheduler, an integrated multicast-unicast bridge, interoperability with H323 devices, and a record and playback application. Argonne's AG 2.0 release includes features such as data sharing and a generic framework for application integration.
Events
IE 7
Firefox 2.0
