Study of OpenMP applications on the InfiniBand-based software distributed shared-memory system

Inho Park, Seon Wook Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

For the past decades computer engineers have focused on building high-performance and large-scale computer systems with low-cost. One of the examples is a distributed-memory computer system like a cluster, where fast processing nodes to use commodity processors are connected through a high speed network. But it is not easy to develop applications on this system, because a programmer must consider all data and control dependences between processes and program them explicitly. For alleviating this problem the distributed virtual shared-memory (DVSM) system has been proposed. It is well known that the performance of the DVSM system highly depends on the network's performance and programming semantics, and currently its performance is very limited on a conventional network. Recently many advanced hardware-based interconnection technologies have been introduced, and one of them is the InfiniBand Architecture (IBA) which supports shared-memory programming semantics by means of remote direct-memory access (RDMA) and atomic operations. In this paper, we present the implementation of our InfiniBand-based DVSM system and analyze the performance of SPEC OMP benchmarks in detail by comparing with the DVSM based on the traditional network architecture and the hardware shared-memory multiprocessor (SMP) system. As experiment result, we show that our DVSM system to use full features of the IBA can improve the performance significantly over the IPoIB-based traditional system on the IBA, and furthermore the performance of one application on the IBA-based DVSM system is better than on the hardware SMP.

Original languageEnglish
Pages (from-to)1099-1113
Number of pages15
JournalParallel Computing
Volume31
Issue number10-12
DOIs
Publication statusPublished - 2005 Oct 1

Fingerprint

InfiniBand
Distributed Shared Memory
OpenMP
Virtual Memory
Computer systems
Shared Memory
Data storage equipment
Software
Shared-memory multiprocessors
Hardware
Computer programming
Computer hardware
Network Programming
Semantics
High-speed Networks
Multiprocessor Systems
Distributed Memory
Network Performance
Network Architecture
HIgh speed networks

Keywords

  • Distributed virtual shared-memory
  • InfiniBand
  • OpenMP
  • Performance

ASJC Scopus subject areas

  • Computer Science Applications
  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

Study of OpenMP applications on the InfiniBand-based software distributed shared-memory system. / Park, Inho; Kim, Seon Wook.

In: Parallel Computing, Vol. 31, No. 10-12, 01.10.2005, p. 1099-1113.

Research output: Contribution to journalArticle

@article{d91b4e7445df4edf929005aab118fc06,
title = "Study of OpenMP applications on the InfiniBand-based software distributed shared-memory system",
abstract = "For the past decades computer engineers have focused on building high-performance and large-scale computer systems with low-cost. One of the examples is a distributed-memory computer system like a cluster, where fast processing nodes to use commodity processors are connected through a high speed network. But it is not easy to develop applications on this system, because a programmer must consider all data and control dependences between processes and program them explicitly. For alleviating this problem the distributed virtual shared-memory (DVSM) system has been proposed. It is well known that the performance of the DVSM system highly depends on the network's performance and programming semantics, and currently its performance is very limited on a conventional network. Recently many advanced hardware-based interconnection technologies have been introduced, and one of them is the InfiniBand Architecture (IBA) which supports shared-memory programming semantics by means of remote direct-memory access (RDMA) and atomic operations. In this paper, we present the implementation of our InfiniBand-based DVSM system and analyze the performance of SPEC OMP benchmarks in detail by comparing with the DVSM based on the traditional network architecture and the hardware shared-memory multiprocessor (SMP) system. As experiment result, we show that our DVSM system to use full features of the IBA can improve the performance significantly over the IPoIB-based traditional system on the IBA, and furthermore the performance of one application on the IBA-based DVSM system is better than on the hardware SMP.",
keywords = "Distributed virtual shared-memory, InfiniBand, OpenMP, Performance",
author = "Inho Park and Kim, {Seon Wook}",
year = "2005",
month = "10",
day = "1",
doi = "10.1016/j.parco.2005.03.014",
language = "English",
volume = "31",
pages = "1099--1113",
journal = "Parallel Computing",
issn = "0167-8191",
publisher = "Elsevier",
number = "10-12",

}

TY - JOUR

T1 - Study of OpenMP applications on the InfiniBand-based software distributed shared-memory system

AU - Park, Inho

AU - Kim, Seon Wook

PY - 2005/10/1

Y1 - 2005/10/1

N2 - For the past decades computer engineers have focused on building high-performance and large-scale computer systems with low-cost. One of the examples is a distributed-memory computer system like a cluster, where fast processing nodes to use commodity processors are connected through a high speed network. But it is not easy to develop applications on this system, because a programmer must consider all data and control dependences between processes and program them explicitly. For alleviating this problem the distributed virtual shared-memory (DVSM) system has been proposed. It is well known that the performance of the DVSM system highly depends on the network's performance and programming semantics, and currently its performance is very limited on a conventional network. Recently many advanced hardware-based interconnection technologies have been introduced, and one of them is the InfiniBand Architecture (IBA) which supports shared-memory programming semantics by means of remote direct-memory access (RDMA) and atomic operations. In this paper, we present the implementation of our InfiniBand-based DVSM system and analyze the performance of SPEC OMP benchmarks in detail by comparing with the DVSM based on the traditional network architecture and the hardware shared-memory multiprocessor (SMP) system. As experiment result, we show that our DVSM system to use full features of the IBA can improve the performance significantly over the IPoIB-based traditional system on the IBA, and furthermore the performance of one application on the IBA-based DVSM system is better than on the hardware SMP.

AB - For the past decades computer engineers have focused on building high-performance and large-scale computer systems with low-cost. One of the examples is a distributed-memory computer system like a cluster, where fast processing nodes to use commodity processors are connected through a high speed network. But it is not easy to develop applications on this system, because a programmer must consider all data and control dependences between processes and program them explicitly. For alleviating this problem the distributed virtual shared-memory (DVSM) system has been proposed. It is well known that the performance of the DVSM system highly depends on the network's performance and programming semantics, and currently its performance is very limited on a conventional network. Recently many advanced hardware-based interconnection technologies have been introduced, and one of them is the InfiniBand Architecture (IBA) which supports shared-memory programming semantics by means of remote direct-memory access (RDMA) and atomic operations. In this paper, we present the implementation of our InfiniBand-based DVSM system and analyze the performance of SPEC OMP benchmarks in detail by comparing with the DVSM based on the traditional network architecture and the hardware shared-memory multiprocessor (SMP) system. As experiment result, we show that our DVSM system to use full features of the IBA can improve the performance significantly over the IPoIB-based traditional system on the IBA, and furthermore the performance of one application on the IBA-based DVSM system is better than on the hardware SMP.

KW - Distributed virtual shared-memory

KW - InfiniBand

KW - OpenMP

KW - Performance

UR - http://www.scopus.com/inward/record.url?scp=27844537439&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27844537439&partnerID=8YFLogxK

U2 - 10.1016/j.parco.2005.03.014

DO - 10.1016/j.parco.2005.03.014

M3 - Article

VL - 31

SP - 1099

EP - 1113

JO - Parallel Computing

JF - Parallel Computing

SN - 0167-8191

IS - 10-12

ER -