Supporting the Global Arrays PGAS Model Using MPI One-Sided Communication

被引：25

作者：

Dinan, James ^{[1
]}

Balaji, Pavan ^{[1
]}

Hammond, Jeff R. ^{[1
]}

Krishnamoorthy, Sriram ^{[2
]}

Tipparaju, Vinod

机构：

[1] Argonne Natl Lab, Argonne, IL 60439 USA

[2] Pacific Northwest Natl Lab, Richland, WA 99354 USA

来源：

2012 IEEE 26TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM (IPDPS) | 2012年

关键词：

One-sided communication; Global address space; MPI; Global Arrays; ARMCI; NWChem;

D O I：

10.1109/IPDPS.2012.72

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

The industry-standard Message Passing Interface (MPI) provides one-sided communication functionality and is available on virtually every parallel computing system. However, it is believed that MPI's one-sided model is not rich enough to support higher-level global address space parallel programming models. We present the first successful application of MPI one-sided communication as a runtime system for a PGAS model, Global Arrays (GA). This work has an immediate impact on users of GA applications, such as NWChem, who often must wait several months to a year or more before GA becomes available on a new architecture. We explore challenges present in the application of MPI-2 to PGAS models and motivate new features in the upcoming MPI-3 standard. The performance of our system is evaluated on several popular high-performance computing architectures through communication benchmarking and application benchmarking using the NWChem computational chemistry suite.

引用

页码：739 / 750

页数：12

共 21 条

[1] ADELSONVELSKII GM, 1962, DOKL AKAD NAUK SSSR+, V146, P263
[2] [Anonymous], 1996, MPI2 U TENN
[3] Apra Edoardo., 2009, Proceedings of the conference on high performance computing networking, storage and analysis, P66, DOI [10.1145/1654059.1654127, DOI 10.1145/1654059.1654127]
[4] Berkeley UPC, 2009, BERK UPC US GUID VER
[5] Bonachea D., 2004, International Journal of High Performance Computing and Networking, V1, P91, DOI 10.1504/IJHPCN.2004.007569
[6] Bonachea D., AMMPI: Active Messages over MPI - Quick Overview
[7] Bonachea Dan, 2002, UCBCSD021207
[8] Cormen TH., 2009, Introduction to Algorithms, V3
[9] Cui Y., 2010, P IEEE ACM INT C HIG
[10] Dinan James, 2011, RECENT ADV MESSAGE P

← 1 2 3 →