TY - GEN
T1 - The importance of dynamic load balancing among openmp thread teams for irregular workloads
AU - Xiao, Xiong
AU - Hirasawa, Shoichi
AU - Takizawa, Hiroyuki
AU - Kobayashi, Hiroaki
PY - 2017/1/13
Y1 - 2017/1/13
N2 - Recently, massively-parallel many-core processors such as Intel Xeon Phi coprocessors have attracted researchers' attentions because various applications are significantly accelerated with those processors. In the field of high-performance computing, OpenMP is a standard programming model commonly used to parallelize a kernel loop for many-core processors. For hierarchical parallel processing, OpenMP version 4.0 or later allows programmers to group threads into multiple thread teams. In this paper, we first show the performance gain of using multiple thread teams even for one many-core processor. Then, we demonstrate that dynamic load balancing among those thread teams has a potential of significantly improving the performance of irregular workloads on a many-core processor. Although the current OpenMP specification does not offer such a dynamic load balancing mechanism, we discuss possible benefits of dynamic load balancing among thread teams through experiments using the Intel Xeon Phi coprocessor.
AB - Recently, massively-parallel many-core processors such as Intel Xeon Phi coprocessors have attracted researchers' attentions because various applications are significantly accelerated with those processors. In the field of high-performance computing, OpenMP is a standard programming model commonly used to parallelize a kernel loop for many-core processors. For hierarchical parallel processing, OpenMP version 4.0 or later allows programmers to group threads into multiple thread teams. In this paper, we first show the performance gain of using multiple thread teams even for one many-core processor. Then, we demonstrate that dynamic load balancing among those thread teams has a potential of significantly improving the performance of irregular workloads on a many-core processor. Although the current OpenMP specification does not offer such a dynamic load balancing mechanism, we discuss possible benefits of dynamic load balancing among thread teams through experiments using the Intel Xeon Phi coprocessor.
UR - http://www.scopus.com/inward/record.url?scp=85015258445&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85015258445&partnerID=8YFLogxK
U2 - 10.1109/CANDAR.2016.48
DO - 10.1109/CANDAR.2016.48
M3 - Conference contribution
AN - SCOPUS:85015258445
T3 - Proceedings - 2016 4th International Symposium on Computing and Networking, CANDAR 2016
SP - 529
EP - 535
BT - Proceedings - 2016 4th International Symposium on Computing and Networking, CANDAR 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 4th International Symposium on Computing and Networking, CANDAR 2016
Y2 - 22 November 2016 through 25 November 2016
ER -