Design, Implementation, and Performance of MPI on Portals 3.0

Author:

Brightwell Ron,Riesen Rolf1,Maccabe Arthur B.2

Affiliation:

1. CENTER FOR COMPUTATION, COMPUTERS, INFORMATION, AND MATHEMATICS, SANDIA NATIONAL LABORATORIES, U.S.A.

2. DEPARTMENT OF COMPUTER SCIENCE, UNIVERSITY OF NEW MEXICO, U.S.A.

Abstract

This paper describes an implementation of the Message Passing Interface (MPI) on the Portals 3.0 data movement layer. Portals 3.0 provides low-level building blocks that are flexible enough to support higher-level message passing layers, such as MPI, very efficiently. Portals 3.0 is also designed to allow for programmable network interface cards to offload message processing from the host processor, allowing for the ability to overlap computation and MPI communication. We describe the basic building blocks in Portals 3.0, show how they can be put together to implement MPI, and describe the protocols of our MPI implementation. We look at several key operations within the implementation and describe the effects that a Portals 3.0 implementation has on scalability and performance. We also present preliminary performance results from our implementation for Myrinet.

Publisher

SAGE Publications

Subject

Hardware and Architecture,Theoretical Computer Science,Software

Reference8 articles.

1. Myrinet: a gigabit-per-second local area network

2. Massively parallel computing using commodity components

3. A high-performance, portable implementation of the MPI message passing interface standard

4. MPI-FM: High Performance MPI on Workstation Clusters

5. Message Passing Interface Forum. 1994. MPI: A message-passing interface standard . The International Journal of Supercomputer Applications and High Performance Computing, 8: (3-4): 159-416

Cited by 11 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Overview: The Birth of Lightweight Kernels;Operating Systems for Supercomputers and High Performance Computing;2019

2. Mitigating MPI Message Matching Misery;Lecture Notes in Computer Science;2016

3. Toward Implementing Robust Support for Portals 4 Networks in MPICH;2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing;2015-05

4. HiCOO: Hierarchical cooperation for scalable communication in Global Address Space programming models on Cray XT systems;Journal of Parallel and Distributed Computing;2012-11

5. A Low Impact Flow Control Implementation for Offload Communication Interfaces;Recent Advances in the Message Passing Interface;2012

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3