More on the team

For the past 20 years, Daniel Dias has worked at the edge of parallel computer processing, pushing the limits of speed. With the introduction of Grid computing, increasing processing capacity is no longer a matter of getting more machines, but making better use of what you already have. Some current demonstrations of Grid computing show a 100-fold decrease in processing time for complex applications using available capacity on existing computers.

Daniel's work has shaped the development of clustered and parallel computing systems. He first worked on the architecture and performance of clustered mainframe and supercomputers; then the development of a scalable Web server; clustered Web caching, and now Grid computing. Along the way, he has been awarded several U.S. patents for his work.

Parallel computing is about splitting up the processing tasks so that they can be done simultaneously. Done correctly, processing in parallel greatly decreases the time it takes to get results and increases the number of tasks that can be performed at a given moment. The kind of applications that can take advantage of parallel processing has changed over the years, as has the scale of the problems that can be solved. The key thing driving it all is the amount of processing power that can be harnessed.

First, processing was shared across processors in the same computer; then among computers in the same room, site, organization; and now multiple sites around the world. With each incremental gain in distribution comes a leap in the amount of processing available without adding capacity.

Today, Daniel Dias spends his time working on how to harness all that power for new classes of applications to do more with less. His craze for parallel processing extends to his weekends -- he likes cooking four dishes "in parallel" and playing different games simultaneously with his children.