TY - GEN
T1 - Automatically tuned collective communications
AU - Vadhiyar, Sathish S.
AU - Fagg, Graham E.
AU - Dongarra, Jack
N1 - Publisher Copyright:
© 2000 IEEE.
PY - 2000
Y1 - 2000
N2 - The performance of the MPI's collective communications is critical in most MPI-based applications. A general algorithm for a given collective communication operation may not give good performance on all systems due to the differences in architectures, network parameters and the storage capacity of the underlying MPI implementation. In this paper, we discuss an approach in which the collective communications are tuned for a given system by conducting a series of experiments on the system. We also discuss a dynamic topology method that uses the tuned static topology shape, but re-orders the logical addresses to compensate for changing run time variations. A series of experiments were conducted comparing our tuned collective communication operations to various native vendor MPI implementations. The use of the tuned collective communications resulted in about 30%-650% improvement in performance over the native MPI implelementations.
AB - The performance of the MPI's collective communications is critical in most MPI-based applications. A general algorithm for a given collective communication operation may not give good performance on all systems due to the differences in architectures, network parameters and the storage capacity of the underlying MPI implementation. In this paper, we discuss an approach in which the collective communications are tuned for a given system by conducting a series of experiments on the system. We also discuss a dynamic topology method that uses the tuned static topology shape, but re-orders the logical addresses to compensate for changing run time variations. A series of experiments were conducted comparing our tuned collective communication operations to various native vendor MPI implementations. The use of the tuned collective communications resulted in about 30%-650% improvement in performance over the native MPI implelementations.
UR - http://www.scopus.com/inward/record.url?scp=85095842731&partnerID=8YFLogxK
U2 - 10.1109/SC.2000.10024
DO - 10.1109/SC.2000.10024
M3 - Conference contribution
AN - SCOPUS:85095842731
T3 - Proceedings of the International Conference on Supercomputing
BT - SC 2000 - Proceedings of the 2000 ACM/IEEE Conference on Supercomputing
PB - Association for Computing Machinery
T2 - 2000 ACM/IEEE Conference on Supercomputing, SC 2000
Y2 - 4 November 2000 through 10 November 2000
ER -