20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP 2015), February 7–11, 2015, San Francisco, CA, USA

Desktop Layout

Locking and Locality
Session
MPI+Threads: Runtime Contention and Remedies
Abdelhalim Amer, Huiwei Lu, Yanjie Wei, Pavan Balaji, and Satoshi Matsuoka
(Tokyo Institute of Technology, Japan; Argonne National Laboratory, USA; Shenzhen Institute of Advanced Technologies at Chinese Academy of Sciences, China)
Publisher's Version
Abstract: Hybrid MPI+Threads programming has emerged as an alternative model to the “MPI everywhere” model to better handle the increasing core density in cluster nodes. While the MPI standard allows multithreaded concurrent communication, such flexibility comes with the cost of maintaining thread safety within the MPI implementation, typically implemented using critical sections. In contrast to previous works that studied the importance of critical-section granularity in MPI implementations, in this paper we investigate the implication of critical-section arbitration on communication performance. We first analyze the MPI runtime when multithreaded concurrent communication takes place on hierarchical memory systems. Our results indicate that the mutex-based approach that most MPI implementations use today can incur performance penalties due to unfair arbitration. We then present methods to mitigate these penalties with a first-come, first-served arbitration and a priority locking scheme that favors threads doing useful work. Through evaluations using several benchmarks and applications, we demonstrate up to 5-fold improvement in performance.

Authors:


Time stamp: 2019-07-23T01:11:38+02:00