TY - GEN
T1 - MPI-Reduce algorithm for OpenFlow-enabled network
AU - Makpaisit, Pisit
AU - Ichikawa, Kohei
AU - Uthayopas, Putchong
AU - Date, Susumu
AU - Takahashi, Keichi
AU - Khureltulga, Dashdavaa
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2016/4/22
Y1 - 2016/4/22
N2 - The MPI reduction operation such as MPI-Reduce and MPI-Allreduce are frequently used and time-consuming operations. The performance enhancement of these operations can substantially speed up large-scale parallel applications. In this paper, a greedy based MPI-Reduce algorithm called Greedy Shortest Binomial Tree (GSBT) is proposed. This proposed algorithm leverages SDN technology and OpenFlow network to speed up MPI reduction operations. This is accomplished using network topology information from the OpenFlow controller to reduce overall hops in message transmission. The implementation of the proposed algorithm by modifying MPI library and OpenFlow controller is presented. The proposed GSBT algorithm has been evaluated in a real test-bed to compare with the traditional approaches used in both MPICH and Open MPI. The result shows that GSBT algorithm is faster than standard algorithms 30.48-66.35% for Open MPI and faster 50.77-82.89% for MPICH when message size between 2 KB - 24 KB.
AB - The MPI reduction operation such as MPI-Reduce and MPI-Allreduce are frequently used and time-consuming operations. The performance enhancement of these operations can substantially speed up large-scale parallel applications. In this paper, a greedy based MPI-Reduce algorithm called Greedy Shortest Binomial Tree (GSBT) is proposed. This proposed algorithm leverages SDN technology and OpenFlow network to speed up MPI reduction operations. This is accomplished using network topology information from the OpenFlow controller to reduce overall hops in message transmission. The implementation of the proposed algorithm by modifying MPI library and OpenFlow controller is presented. The proposed GSBT algorithm has been evaluated in a real test-bed to compare with the traditional approaches used in both MPICH and Open MPI. The result shows that GSBT algorithm is faster than standard algorithms 30.48-66.35% for Open MPI and faster 50.77-82.89% for MPICH when message size between 2 KB - 24 KB.
KW - MPI
KW - reduction operation
KW - SDN
UR - http://www.scopus.com/inward/record.url?scp=84979018841&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84979018841&partnerID=8YFLogxK
U2 - 10.1109/ISCIT.2015.7458357
DO - 10.1109/ISCIT.2015.7458357
M3 - Conference contribution
AN - SCOPUS:84979018841
T3 - 2015 15th International Symposium on Communications and Information Technologies, ISCIT 2015
SP - 261
EP - 264
BT - 2015 15th International Symposium on Communications and Information Technologies, ISCIT 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th International Symposium on Communications and Information Technologies, ISCIT 2015
Y2 - 7 October 2015 through 9 October 2015
ER -