|
High Performance Computing
High Performance Systems Lab has been actively developing algorithms and systems
for a variety of challenging problems in parallel and grid computing. Recent focus
has been on performance prediction in unpredictable environments. The approach taken
is based on Performance Skeletons that provide
nearly instant execution time estimates with direct monitored execution, Another
emerging direction is high performance truly parallel computing on idles desktops
under Volpex (Parallel Execution on Volatile Nodes)
project. Other recent projects have spanned, network bandwidth prediction, network
based computing, and web caching.
Please contact us if you are intersted in finding out more about our research, collaborating
with us, or if you are a student interested in this direction of research for your
dissertation. Occasionally paid Research Assistant positions are available for motivated
Ph.D. and M.S. students.
|
|
|
