An implementation and evaluation of the MPI 3.0 one‐sided communication interface. (9th January 2016)

• Record Type: Journal Article
• Authors: Dinan, James; Balaji, Pavan; Buntinas, Darius; Goodell, David; Gropp, William; Thakur, Rajeev
• Is Part Of: Concurrency and computation. Volume 28:Number 17(2016)
• Page Start: 4385
• Page End: 4404
• Publication Date: 2016-01-09
• Subjects: Message Passing Interface (MPI) -- one‐sided communication -- remote memory access (RMA) -- MPICH; Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.3758
• Languages: English
• ISSNs: 1532-0626

Abstract

Summary: The Message Passing Interface (MPI) 3.0 standard includes a significant revision to MPI's remote memory access (RMA) interface, which provides support for one‐sided communication. MPI‐3 RMA is expected to greatly enhance the usability and performance of MPI RMA. We present the first complete implementation of MPI‐3 RMA and document implementation techniques and performance optimization opportunities enabled by the new interface. Our implementation targets messaging‐based networks and is publicly available in the latest release of the MPICH MPI implementation. Using this implementation, we explore the performance impact of new MPI‐3 functionality and semantics. Results indicate that the MPI‐3 RMA interface provides significant advantages over the MPI‐2 interface by enabling increased communication concurrency through relaxed semantics in the interface and additional routines that provide new window types, synchronization modes, and atomic operations. Copyright © 2016 John Wiley & Sons, Ltd.