Abstract
This paper introduces Madeleine II, a new adaptive and portable multi-protocol implementation of the Madeleine communication library. Madeleine II has the ability to control multiple network interfaces (BIP, SISCI, VIA) and multiple network adapters (Ethernet, Myrinet, SCI) within the same application session. We report on performance measurements obtained using BIP/Myrinet and SISCI/SCI and we present preliminary results about our MPICH/Madeleine II and Nexus/Madeleine II ports. We also discuss an extension of Madeleine II for clusters of clusters which is able to handle heterogeneous networks. In particular, we present the fast internal data-forwarding mechanism that is used on gateway nodes to speed up inter-cluster transmissions. Preliminary experiments show that the resulting inter-cluster bandwidth is close to the one delivered by the hardware.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
I. Foster, C. Kesselman and S. Tuecke, The Nexus approach to integrating multithreading and communications, Journal on Parallel and Distributed Computing 37(1) (1996) 70–82.
R. Namyst and J.-F. Méhaut, PM2: Parallel Multithreaded Machine; a computing environment for distributed architectures, in: Parallel Computing (ParCo'95) (Elsevier, Amsterdam, 1995) pp. 279–285.
L. Bougé, J.-F. Méhaut and R.-Namyst, Efficient communications in multithreaded runtime systems, in: Parallel and Distributed Processing. Proc. 3rd Workshop on Runtime Systems for Parallel Programming (RTSPP'99), Lecture Notes in Computer Science, Vol. 1586 (Springer, San Juan, Puerto Rico, 1999) pp. 468–482.
J. Worringen and T. Bemmerl, MPICH for SCI-connected clusters, in: SCI Europe'99 (Bordeaux, France, 1999) pp. 3–11.
D. Dunning, G. Regnier, G. McAlpine, D. Cameron, B. Shubert, F. Berry, A.-M. Merritt, E. Gronke and C. Dodd, The virtual interface architecture, IEEE Micro (March 1998) 66–75.
E. Lusk and W. Gropp. MPICH working note: The second-generation ADI for the MPICH implementation ofMPI, Technical report, Argonne National Laboratory (1996).
S. Pakin, V. Karamcheti and A. Chien, Fast Messages: Efficient, portable communication for workstation clusters and MPPs, IEEE Concurrency 5(2) (1997) 60–73.
H. Hellwagner and A. Reinefeld, eds., SCI: Scalable Coherent Interface, Architecture and Software for High-Performance Compute Clusters, Lecture Notes in Computer Science, State-of-the-Art Surveys, Vol. 1734 (Springer, Amsterdam, 1999).
L. Prylli and B. Tourancheau, BIP: a new protocol designed for high performance networking on Myrinet, in: 1st Workshop on Personal Computer Based Networks Of Workstations (PC-NOW'98), Lecture Notes in Computer Science, Vol. 1388 (Springer, Amsterdam, 1998) pp. 472–485.
G. Mercier, Support efficace de l'hétérogénéité des réseaux dans MPI, Master's thesis report, DEA d'Informatique Fondamentale, ENS Lyon, France (June 2000).
A. Denis, Adaptation de l'environnemnet générique de metacomputing Globus à des réseaux haut débit, Master's thesis report, DEA d'Informatique Fondamentale, ENS Lyon, France (June 2000).
Sca-MPI, http://www.scali.com/.
I. Foster, J. Geisler, C. Kesselman and S. Tuecke, Managing multiple communication methods in high-performance networked computing systems, Journal on Parallel and Distributed Computing 40 (1997) 35–48.
PACX-MPI, http://www.hlrs.de/structure/ organisation/par/projects/pacx-mpi/.
R. Russell and P. Hatcher, Efficient kernel support for reliable communication, in: 13th ACM Symposium on Applied Computing (Atlanta, GA, 1998) pp. 541–550.
V. Danjean, R. Namyst and R. Russell, Integrating kernel activations in a multithreaded runtime system on Linux, in: Proc. 4th Workshop on Runtime Systems for Parallel Processing (RTSPP'00), Lecture Notes in Computer Science, Vol. 1800 (Springer, Cancun, Mexico, 2000) pp. 1160–1167.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Aumage, O., Bougé, L., Eyraud, L. et al. High Performance Computing on Heterogeneous Clusters with the Madeleine II Communication Library. Cluster Computing 5, 43–54 (2002). https://doi.org/10.1023/A:1012792605138
Issue Date:
DOI: https://doi.org/10.1023/A:1012792605138