COMPASSION: A parallel I/O runtime system including chunking and compression for irregular applications
In this paper we present an experimental evaluation of COMPASSION, a runtime system for irregular applications based on collective I/O techniques. It provides a “Collective I/O” model, enhanced with “Pipelined” operations and compression. All processors participate in the I/O simultaneously, alone or grouped, making scheduling of I/O requests simpler and providing support for contention management. In-memory compression mechanisms reduce the total execution time by diminishing the amount of I/O requested and the I/O contention. Our experiments, executed on an Intel Paragon and on the ASCI/Red teraflops machine, demonstrate that COMPASSION can obtain significantly high-performance for I/O above what has been possible so far.
Keywordsirregular applications runtime systems parallel I/O compression
Unable to display preview. Download preview PDF.
- 1.J. Carretero, J. No, S. Park, A. Choudhary, and P. Chen. Compassion: a parallel i/o runtime system including chunking and compression for irregular applications. In Proceedings of the International Conference on High-Performance Computing and Networking 1998, Amsterdam, Holland, April 1998.Google Scholar
- 2.Alok Choudhary, Rajesh Bordawekar, Michael Harry, Rakesh Krishnaiyer, Ravi Ponnusamy, Tarvinder Singh, and Rajeev Thakur. PASSION: parallel and scalable software for input-output. Technical Report SOCS-636, ECE Dept., NPAC and CASE Center, Syracuse University, September 1994.Google Scholar
- 3.Juan Miguel del Rosario and Alok Choudhary. High performance I/O for parallel computers: Problems and prospects. IEEE Computer, 27(3):59–68, March 1994.Google Scholar
- 4.T. Mattson and G. Henry. The asci option red supercomputer. In Intel Supercomputer Users Group. Thirteenth Annual Conference, Albuquerque, USA, June 1997.Google Scholar
- 5.J. No, S. Park, J. Carretero, A. Choudhary, and P. Chen. Design and implementation of a parallel i/o runtime system for irregular applications. In Proceedings of the 12th International Parallel Processing Symposium, Orlando, USA, March 1998.Google Scholar
- 6.K. É. Seamons and M. Winslett. A data management approach for handling large compressed arrays in high performance computing. In Proceedings of the Fifth Symposium on the Frontiers of Massively Parallel Computation, pages 119–128, February 1995.Google Scholar