Abstract
Communication costs represent a significant portion of the execution time of most distributed applications. Thus, it is important to optimize the communication behavior of the algorithm to match the capabilities of the underlying communication fabric. Traditionally, optimizations to the communication behavior have been carried out statically and at the application level (optimizing partitioning, using the most appropriate communication protocols, etc). This paper introduces a new class of optimizations to communication: active run-time matching between the application communication behavior and the communication layer. We propose an active layer extension to the Message Passing Interface (MPI) that dynamically reduces the average communication overhead associated with message sends and receives. The active layer uses dynamic message aggregation to reduce the send overheads and infrequent polling to reduce the receive overhead of messages. The performance of the active layer is evaluated using a number of applications.
Support for this work was provided in part by the Advanced Research Projects Agency under contract numbers DABT63-96-C-0055 and J-FBI-93-116.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Amarasinghe, S., and Lam, M. Communication optimization and code generation for distributed memory machines. In SIGPLAN '93 Conference on Programming Language Design and Implementation (1991), pp. 126–138.
Bailey, M. L., Jr., J. V. B., and Chamberlain, R. D. Parallel logic simulation of VLSI systems. ACM Computing Surveys 26, 3 (September 1994), 255–294.
Briner, Jr., J. V.Parallel Mixed-Level Simulation of Digital Circuits using Virtual Time. PhD thesis, Duke University, Durham, North Carolina, 1990.
Center, N. A. R. NAS parallel benchmarks home page. http://science.nas.nasa.gov/Software/NPB/.
Chandrasekaran, S., and Venkatesan, S. A message optimal algorithm for distributed termination detection. Journal Of Parallel and Distributed Computing 8 (1990), 245–252.
Child, J. Bus wars the sequel: Vme shoots back. Computer Design (March 1997).
Dillon, E., Sanots, C. G. D., and Guyard, J. Homogenous and heterogenous network of workstations: Message passing overhead. In MPI Developers Conference (June 1995).
Felten, E. W. Protocol compilation: High-performance communication for parallel programs. Tech. rep., University of Washington — Dept. of Computer Science, 1993.
Fujimoto, R. Parallel discrete event simulation. Communications of the ACM 33, 10 (October 1990), 30–53.
Gropp, W., and Lusk, E. Tuning MPI programs for peak performance, http://www.mcs.anl.gov/mpi/.
Gropp, W., Lusk, E., and Skjellum, A.Using MPI: Portable Parallel Programming with the Message-Passing Interface. MIT Press, Cambridge, MA, 1994.
Karamcheti, V., and Chien, A. Software overhead in messaging layers: Where does the time go? In Proceedings of ASPLOS — VI, SanJose, California (March 1994).
Levendel, Y. H., Menon, P. R., and Patel, S. H. Special purpose computer for logic simulation using distributed processing. Bell Syst. Tech. J. 61, 10,2873–2909 (1982).
Martin, D. E., McBrayer, T. J., and Wilsey, P. A. warped: A time warp simulation kernel for analysis and application development. In 29th Hawaii International Conference on System Sciences (HICSS-29) (Jan. 1996), H. El-Rewini and B. D. Shriver, Eds., vol. Volume I, pp. 383–386.
Nevin., N. The performance of LAM 6.0 and MPICH 1.0.12 on a workstation cluster. Tech. Rep. OSC-TR-1996-4, Ohio Supercomputer Center Technical Report Columbus, OhioTech., 1996.
Nupairoj, N., and Ni., L. Performance evaluation of some MPI implementations. Tech. Rep. Tech. Rept. MSU-CPS-ACS-94, Dept. of Computer Science, Michigan State University, Sept. 1994.
Smith, S. P., Underwood, B., and Mercer, M. R. An analysis of several approaches to circuit partitioning for parallel logic simulation. In In Proceedings of the 1987 International Conference on Computer Design. (1987), IEEE, NewYork, pp. 664–667.
Sunderram, V. PVM: A framework for parallel and distributed computing. Concurrency: Practice and Experience (December 1990), 315–339.
von Eicken, T., Basu, A., Buch, V., and Vogels, W. U-net: A user-level network interface for parallel and distributes computing. In Proceedings of the 15th ACM Symposium on Operating Sysytem Principles (December 1995).
von Eicken, T., Culler, D., Goldstein, S., and Schasuer, K. Active messages: A mechanism for integrated communication and computation. In Proceedings of the 19th International Symposium on Computer Architecture (May 1992).
Xu, Zhiwei. Hwang, K. Modeling communication overhead: MPI and MPL performance on the IBM SP. IEEE Parallel & Distributed Technology. 4, 1 (Spring 1996), 9–23.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chetlur, M., Sharma, G.D., Abu-Ghazaleh, N., Rajasekaran, U.K.V., Wilsey, P.A. (1998). An active layer extension to MPI. In: Alexandrov, V., Dongarra, J. (eds) Recent Advances in Parallel Virtual Machine and Message Passing Interface. EuroPVM/MPI 1998. Lecture Notes in Computer Science, vol 1497. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0056564
Download citation
DOI: https://doi.org/10.1007/BFb0056564
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65041-6
Online ISBN: 978-3-540-49705-9
eBook Packages: Springer Book Archive