Summary
A flexible intermediate library named Stampi realizes seamless MPI operations on interconnected parallel computers. In message transfer of Stampi, a vendor-supplied MPI library and TCP sockets are used selectively among MPI processes. Its router process mechanism hides a complex network configuration in message transfer among computers. Besides, dynamic process creation and MPI-I/O are also available. With the help of its flexible communication mechanism, users can execute MPI functions without awareness of underlying communication and I/O mechanisms. To realize distributed I/O operations with high performance, a Parallel Virtual File System (PVFS) has been introduced in the MPI-I/O mechanism of Stampi. Collective MPI-I/O functions have been evaluated and sufficient performance has been achieved.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Message Passing Interface Forum: MPI: A message-passing interface standard (1995)
Message Passing Interface Forum: MPI-2: Extensions to the message-passing interface standard (1997)
Imamura, T., Tsujita, Y., Koide, H., and Takemiya, H.: An architecture of Stampi: MPI library on a cluster of parallel computers. In Recent Advances in Parallel Virtual Machine and Message Passing Interface, vol. 1908 of Lecture Notes in Computer Science, Springer, 200–207 (2000)
Tsujita, Y., Imamura, T., Takemiya, H., and Yamagishi, N.: Stampi-I/O: A flexible parallel-I/O library for heterogeneous computing environment. In Recent Advances in Parallel Virtual Machine and Message Passing Interface, vol. 2474 of Lecture Notes in Computer Science, Springer, 288–295 (2002)
Tsujita, Y.: Flexible Intermediate Library for MPI-2 Support on an SCore Cluster System. In Grid and Cooperative Computing, vol. 3033 of Lecture Notes in Computer Science, Springer, 129–136 (2004)
Carns, P., Ligion III, W., Ross, R., and Thakur, R.: PVFS: A parallel file system for Linux clusters. In Proceedings of the 4th Annual Linux Showcase and Conference, USENIX Association, 317–327 (2000)
Thakur, R., Gropp, W., and Lusk, E.: On implementing MPI-IO portably and with high performance. In Proceedings of the Sixth Workshop on Input/Output in Parallel and Distributed Systems, 23–32 (1999)
Gropp, W., Lusk, E., Doss, N., and Skjellum, A.: A high-performance, portable implementation of the MPI message-passing interface standard. Parallel Computing, 22(6), 789–828 (1996)
PC Cluster Consortium: http://www.pccluster.org/
Matsuda, M., Kudoh, T., and Ishikawa, Y.: Evaluation of MPI implementations on grid-connected clusters using an emulated WAN environment. In Proceedings of the 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGrid 2003), IEEE Computer Society, 10–17 (2003)
Thakur, R., Gropp, W., and Lusk, E.: An abstract-device interface for implementing portable parallel-I/O interfaces. In Proceedings of the Sixth Symposium on the Frontiers of Massively Parallel Computation, 180–187 (1996)
Fagg, G. E., Gabriel, E., Resch, M., and Dongarra, J. J.: Parallel IO support for meta-computing applications: MPI_Connect IO applied to PACX-MPI. In Recent Advances in Parallel Virtual Machine and Message Passing Interface, vol. 2131 of Lecture Notes in Computer Science, Springer, 135–147 (2001)
Gabiriel, E., Resch, M., Beisel, T., and Keller, R.: Distributed computing in a heterogeneous computing environment. In Recent Advances in Parallel Virtual Machine and Message Passing Interface, vol. 1497 of Lecture Notes in Computer Science, Springer, 180–188 (1998)
Dickens, P. and Thakur, R.: Improving collective I/O performance using threads. In Proceedings of the Joint International Parallel Processing Symposium and IEEE Symposium on Parallel and Distributed Processing, 38–45 (1999)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tsujita, Y. (2005). Efficient MPI-I/O Support in Data-Intensive Remote I/O Operations Using a Parallel Virtual File System. In: Zhang, W., Tong, W., Chen, Z., Glowinski, R. (eds) Current Trends in High Performance Computing and Its Applications. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-27912-1_13
Download citation
DOI: https://doi.org/10.1007/3-540-27912-1_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-25785-1
Online ISBN: 978-3-540-27912-9
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)