Home » Technology » Scientific Computing » Web-Based High Performance Computing

Web-Based High Performance Computing

© Adam Hughes

As the Christmas shopping season quickly descends upon us, the American consumer landscape looks quite different from what we've seen in the past. Oh sure, the malls are still packed, and people are still lovingly selecting their yuletide evergreens, but there is something more. In unprecedented numbers, people are flocking to the internet to find their holiday treasures, and this activity represents something a little more profound than the expansion of our marketplace. The real significance is that Americans are starting to tap into the vast interactive potential of ther World Wide Web that has the capability of affecting lives far beyond the simple exchange of information. It is this potential that spurs many of the advances taking place in internet technology, and one of the disciplines pushing this frontiers is scientific computing.

In general, computational scientists, like the rest of society, like to have tools at their disposal which allow them make significant progress in their work without a large associated overhead. Unfortunately, for many of these researchers, the hardware and/or software needed to study a particular problem can often seem quite daunting and complicated. One tool that is particularly comfortable for many scientists these days is the ubiquitous internet browser (pick your favorite flavor). As with most segments of society, the internet has been heartily embraces by science in general, and most researchers use it regularly in some capacity or another. There is now a push to parlay this familiarity with basic web tools into into methods for utilizing more complicated and powerful computer resources.

One area in which this idea is making an impact is the use of supercomputer resources. While many researcher currently use these types of systems, the fact remains that the surface has just been scratched on how profound an impact high- performance computing (HPC) might have on scientific discovery. Making the shift from a Windows environment to a sometimes cold, impersonal Unix world can be quite daunting. But much of this concern could be alleviated if the researcher could interact with the HPC resources via the familiar web browser. This is the end goal of several projects currently underway across the nation.

In order to accomplish a viable web product for accessing HPC resources, several components must be in place. First, of course, there has to be a stable of computing resources available, including hardware and software, that will do the actual grunt work in the simulation process. Secondly, there needs to be a web interface that is friendly enough that the user will want take advantage of its presence. Finally, there needs to be some way for these two ends to

The copyright of the article Web-Based High Performance Computing in Scientific Computing is owned by . Permission to republish Web-Based High Performance Computing in print or online must be granted by the author in writing.