Student Contribution

SC Conference - Activity Details

Toward Loosely-coupled Programming on Petascale Systems

Ioan Raicu  (University of Chicago)
Zhao Zhang  (University of Chicago)
Mike Wilde  (University of Chicago)
Ian Foster  (University of Chicago)
Pete Beckman  (Argonne National Laboratory)
Kamil Iskra  (Argonne National Laboratory)
Ben Clifford  (University of Chicago)
Papers Session
Runtime Systems
Tuesday,  04:00PM - 04:30PM
Room Ballroom F
We have extended the Falkon lightweight task execution framework to make loosely-coupled programming on petascale systems a practical and useful programming model. This work studies and measures the performance factors involved in applying this approach to enable the utilization of petascale systems by a broader user community, and with greater ease. Our work enables the execution of highly parallel computations composed of loosely-coupled serial jobs with no modifications to the respective applications. This approach allows new—and potentially far larger—classes of application to leverage petascale systems, such as the IBM Blue Gene/P supercomputer. We present the challenges of I/O performance encountered in making this model practical, and show results using both micro-benchmarks and real applications from two domains, economic energy modeling and molecular dynamics. Our benchmarks show that we can scale up to 160K processor cores with high efficiency, and can achieve thousands of tasks/sec sustained execution rates.
The full paper can be found in the IEEE Xplore Digital Library and ACM Digital Library
   IEEE Computer Society  /  ACM     2 0   Y E A R S   -   U N L E A S H I N G   T H E   P O W E R   O F   H P C