Abstract
The Clint network provides an FPGA-based segregated architecture with a bulk channel controlled by a quick channel. We report in this paper how, in order to implement efficiently the MPI APIs on top of this network, we “codesigned” the interface between the SunTM MPI communication stack and the network FPGAs.
The SunTM MPI “Protocol Module” we developed implements functions to enable a full support of Sun MPI and gave us an insightful view of the design problems and performance bottlenecks. Hence, we were able to provide pertinent feedback to the hardware designers who then, thanks to the use of rapid FPGA-prototyping, implemented the corresponding hardware enhancements. As a result, our software architecture fits as much as possible with the hardware capabilities and the resulting prototype exploits the best of the overall architecture.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Bibeling, G., Hoppe, H.C., Supalov, A., Lagier, P., Latour, J.: Fujitsu mpi-2: fast locally, reaching globally. In: Kranzlmüller, D., Kacsuk, P., Dongarra, J., Volkert, J. (eds.) PVM/MPI 2002. LNCS, vol. 2474, pp. 401–409. Springer, Heidelberg (2002)
Brightwell, R., Maccabe, A., Riesen, R.: Design and implementation of mpi on portals 3.0. In: Kranzlmüller, D., Kacsuk, P., Dongarra, J., Volkert, J. (eds.) PVM/MPI 2002. LNCS, vol. 2474, pp. 331–340. Springer, Heidelberg (2002)
Chiola, G., Ciaccio, G.: Gamma: a low-cost network of workstations based on active messages. In: PDP 1997, 5th EUROMICRO workshop on Parallel and Distributed Processing, London, UK (January 1997)
Chiola, G., Ciaccio, G.: Porting MPICH ADI on GAMMA with flow control. In: MWPP 1999, 1999 Midwest Workshop on Parallel Processing, Kent, Ohio (August 1999)
Duato, J., López, P., Silla, F.: A High Performance Router Architecture for Interconnection Networks. In: Proc. Int. Conf. On Parallel Processing (1996)
Eberle, H., Gura, N.: Separated high-bandwidth and low-latency communication in the cluster interconnect clint. IEEE, Los Alamitos (2002)
Geoffray, P., Prylli, L., Tourancheau, B.: BIP-SMP: High performance message passing over a cluster of commodity SMPs. In: Supercomputing (SC 1999), Portland, OR (November 1999)
Gropp, W., Lusk, E., Doss, N., Skjellum, A.: A high-performance, portable implementation of the MPI message passing interface standard. Parallel Computing 22(6), 789–828 (1996)
Gura, N., Eberle, H.: The least choice first scheduling method for high-speed network switches. In: Proceedings of the International Parallel and Distributed Processing Symposium, IEEE, Los Alamitos (2002)
Sun Microsystems Inc. Sun hpc clustertools. Technical report (2003), http://www.sun.com/servers/hpc/software/
Sun Microsystems Inc. Sun mpi. Technical report (2003), http://www.sun.com/servers/hpc/software/specifications.html#sunmpi
Karol, M., Hluchyi, M., Morgan, S.: Input versus Output Queuing on a Space- Division Packet Switch. IEEE Transactions on Communications C-35(12), 1347–1356 (1987)
Kim, J., Lilja, D.: Utilizing Heterogeneous Networks in Distributed Parallel Computing Systems. In: Proc. of the 6th Int. Symposium on High Performance Computing (1997)
Kim, J., Lilja, D.: Performance-Based Path Determination for Interprocessor Communication in Distributed Computing Systems. IEEE Trans. on Parallel and Distributed Systems 10(3), 316–327 (1999)
MPI-comittee. Message passing interface forum. Technical report, http://www-unix.mcs.anl.gov/mpi/
Pakin, S., Karamcheti, V., Chien, A.: Fast Messages (FM): Efficient, Portable Communication for Workstation Clusters and Massively-Parallel Processor. IEEE Concurrency 5(2), 60–73 (1997)
Peh, L., Dally, W.: Flit-Reservation Flow Control. In: Proc. 6th Int. Symposium on High-Performance Computer Architecture, Toulouse, France, January 2000, pp. 73–84 (2000)
Prylli, L., Tourancheau, B., Westrelin, R.: The design for a high performance MPI implementation on the Myrinet network. In: Margalef, T., Dongarra, J., Luque, E. (eds.) PVM/MPI 1999. LNCS, vol. 1697, pp. 223–230. Springer, Heidelberg (1999)
Prylli, L., Tourancheau, B.: BIP: a new protocol designed for high performance networking on Myrinet. In: Rolim, J.D.P. (ed.) IPPS-WS 1998 and SPDP-WS 1998. LNCS, vol. 1388, pp. 472–485. Springer, Heidelberg (1998)
Task-Group-btgt11.1. Scheduled Transfer Protocol (ST). Technical Report 3.6, National Committee for Information Technology Standardization (January 2000), http://www.hippi.org
Tourancheau, B., Westrelin, R.: Support for MPI at the network interface level. In: Cotronis, Y., Dongarra, J. (eds.) PVM/MPI 2001. LNCS, vol. 2131, p. 52. Springer, Heidelberg (2001)
Wolf, W.: A decade of hardware/software codesign. IEEE Computer 36(4), 38–44 (2003)
Yuan, X., Melhelm, R., Gupta, R.: Distributed Path Reservation Algorithms for Multiplexed All-optical Interconnection Networks. In: 3rd IEEE Symposium on High-Performance Computer Architecture, San Antonio, Texas (February 1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Fugier, N., Herbert, M., Lemoine, E., Tourancheau, B. (2003). MPI for the Clint Gb/s Interconnect. In: Dongarra, J., Laforenza, D., Orlando, S. (eds) Recent Advances in Parallel Virtual Machine and Message Passing Interface. EuroPVM/MPI 2003. Lecture Notes in Computer Science, vol 2840. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-39924-7_54
Download citation
DOI: https://doi.org/10.1007/978-3-540-39924-7_54
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20149-6
Online ISBN: 978-3-540-39924-7
eBook Packages: Springer Book Archive