Scheduled System Maintenance:
Some services will be unavailable Sunday, March 29th through Monday, March 30th. We apologize for the inconvenience.
By Topic

Adaptive connection management for scalable MPI over InfiniBand

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

The purchase and pricing options are temporarily unavailable. Please try again later.
3 Author(s)
Weikuan Yu ; Dept. of Comput. Sci. & Eng., Ohio State Univ., USA ; Qi Gao ; Panda, D.K.

Supporting scalable and efficient parallel programs is a major challenge in parallel computing with the widespread adoption of large-scale computer clusters and supercomputers. One of the pronounced scalability challenges is the management of connections between parallel processes, especially over connection-oriented interconnects such as VIA and InfiniBand. In this paper, we take on the challenge of designing efficient connection management for parallel programs over InfiniBand clusters. We propose adaptive connection management (ACM) to dynamically control the establishment of InfiniBand reliable connections (RC) based on the communication frequency between MPI processes. We have investigated two different ACM algorithms: an on-demand algorithm that starts with no InfiniBand RC connections; and a partial static algorithm with only 2 * logN number of InfiniBand RC connections initially. We have designed and implemented both ACM algorithms in MVAPICH to study their benefits. Two mechanisms have been exploited for the establishment of new RC connections: one using InfiniBand unreliable datagram and the other using InfiniBand connection management. For both mechanisms, MPI communication issues, such as progress rules, reliability and race conditions are handled to ensure efficient and lightweight connection management. Our experimental results indicate that ACM algorithms can benefit parallel programs in terms of the process initiation time, the number of active connections, and the resource usage. For parallel programs on a 16-node cluster, they can reduce the process initiation time by 15% and the initial memory usage by 18%.

Published in:

Parallel and Distributed Processing Symposium, 2006. IPDPS 2006. 20th International

Date of Conference:

25-29 April 2006