PIO: Parallel I/O System for Massively Parallel Processors
In this paper, we propose a parallel I/O system utilizing parallel commodity network attached to multiple I/O processors of parallel processing systems. I/O requests from user application are automatically distributed to parallel I/O channels to achieve the best utilization of parallel network, and the system provides highly scalable performance according to the number of channels. We also provides an API for easy user-level programming and a set of utilities for high-speed data transfer and real-time parallelized visualization based on it.
KeywordsData Transfer Server Process Parallel Channel User Process Parallel Processor
Unable to display preview. Download preview PDF.
- 1.Watson, R.W., Coyne, R.A.: The Parallel I/O Architecture of the High-Performance Storage System (HPSS), Proc. of the 1995 IEEE MSS Symposium, 1995.Google Scholar
- 2.Sang, J.: Multithreaded Data Transfer over Parallel Links, Proc. of PDPTA’99, 1999.Google Scholar
- 3.Sterling, T., Becker, D., Savarese, D., et.al.: BEOWULF: A Parallel Workstation for Scientific Computation, Proc. of Int. Conf. on Parallel Processing 1995, 1995.Google Scholar
- 4.MPI-2: Extensions to the Message-Passing Interface, http://www.mpi-forum.org/
- 5.Matsubara, M., Numa, H., Boku, T.: Commodity Network based Parallel I/O System for Massively Parallel Processors, Proc. of PDPTA’99, 1999.Google Scholar
- 6.Boku, T., Itakura, K., Nakamura, H., Nakazawa, K.: CP-PACS: A massively parallel processor for large scale scientific calculations, Proc. of ICS’97, 1997.Google Scholar
- 7.Advanced Visual Systems Inc.: AVS/Express, http://www.avs.com/software/soft_t/avsxps.html