|A grant from the National Science Foundation will allow the College of Science to build a state-of-the-art parallel computing lab that could significantly speed up the time it takes to complete projects and research.|
The National Science Foundation has awarded College of Science faculty members a two-year, $160,000 grant that will pay for a computer workstation network designed to drive research in parallel and distributed computing.
While each successive generation of computers push speed and increase raw processing power, there are still limits on what one computer can do when it comes to sorting through mountains of raw data. Parallel computing takes a teamwork approach to such undertakings by dividing one project into several tasks among a network of computers.
?Ten people digging a hole is much faster than one person doing it sequentially,? says Amar Raheja, an assistant professor in Computer Science and a co-investigator on the project.
Working as a team with individual tasks, a network of computers significantly speed up the time it takes to complete a project. By the end of this academic year, Cal Poly Pomona will have a state-of-the-art parallel computing lab thanks to the NSF grant.
?Any computation-intensive application could potentially benefit from parallel computing,? says Hairong Kuang, an assistant professor of Computer Science and the principal investigator on the grant. ?Sorting data is just one type of application.?
While a graduate student at UC Irvine, Kuang used parallel computing to help molecular biologists cut the time needed for a mammoth protein identity project from 30 days down to three days.
?They don't need to learn how to write parallel programs, but with the tool I developed, they get the speedup,? Kuang says.
In the new lab, Kuang plans to continue her work, which reshapes existing computer programs to take advantage of the parallel computing model. She also plans to introduce students to the concept in the forthcoming lab and make it available for senior projects. She hopes the prospect of getting to know this type of computing will attract students to her lab.
Raheja plans to use the computers to work on quickening medical imaging technology from minutes to close to real time.
The new computer workstations will also have practical implications when installed.
Dennis Livesay, an assistant professor of Chemistry and one of the grant?s co-investigators, plans to use the new lab to analyze the proteins encoded within multiple microbial genomes.
Solving a protein?s structure and subsequently which regions are responsible for its function (catalysis) has traditionally been a long and labor-intensive process. Livesay?s work attempts to reduce that time investment by skipping structure and predict functions from the sequence of amino acids that make a particular protein.
Functional regions are predicted from locally conserved strings of amino acids. In the hands of a biochemist, knowing the functional regions can help design further experiments and aid the overall understanding of the protein.
?Recent genomic sequencing efforts, like the Human Genome Project, have created a biological data avalanche,? Livesay says. ?This glut of data is generating nearly as many problems as it provides answers. Work like ours attempts to create efficient and accurate computational tools for managing and understanding this huge amount of data. And then, only after these tools have been developed, will the promised biomedical advances begin to appear.?
The parallel computing lab includes an Mbps network with eight Sun workstations. The grant will pay for a Gbps Myrinet and 8 PCs in the 2004/05 academic year.
Cal Poly Pomona faculty members who will benefit from the grant include: Hairong Kuang, Amar Raheja, Gilbert Young, Keyu Jiang and Sang-Eon Park from the Computer Science department and Dennis Livesay from the Chemistry department.