A scalable architecture for distributed shared memory multiprocessors using optical interconnects

Avinash Karanth Kodi, Ahmed Louri

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

In this paper, we describe the design and analysis of a scalable architecture suitable for large-scale DSMs (Distributed Shared Memory) systems. The approach is based on an interconnect technology which combines optical components and a novel architecture design. In DSM systems, as the network size increases, network contention results in increasing the critical remote memory access latency, which significantly penalizes the performance of DSM systems. In our proposed architecture called RAPID (Reconfigurable and scalable All-Photonic Interconnect for Distributed-shared memory), we provide high connectivity by maximizing the channel availability for remote communication to reduce the remote memory access latency. RAPID also provides fast and efficient unicast, multicast and broadcast capabilities using a combination of aggressively designed wavelength, time and space-division multiplexing techniques. We evaluated RAPID based on network characteristics, power budget criteria and simulation using synthetic traffic workloads and compared it against other scalable electrical networks. We found that RAPID, not only outperforms other networks, but also, satisfies most of the requirements of shared memory multiprocessor design such as low latency, high bandwidth, high connectivity, and easy scalability.

Original languageEnglish (US)
Title of host publicationProceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM)
Pages143-152
Number of pages10
Volume18
StatePublished - 2004
EventProceedings - 18th International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM) - Santa Fe, NM, United States
Duration: Apr 26 2004Apr 30 2004

Other

OtherProceedings - 18th International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM)
CountryUnited States
CitySanta Fe, NM
Period4/26/044/30/04

Fingerprint

Optical interconnects
Data storage equipment
Multiplexing
Photonics
Scalability
Availability
Bandwidth
Wavelength
Communication

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Kodi, A. K., & Louri, A. (2004). A scalable architecture for distributed shared memory multiprocessors using optical interconnects. In Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM) (Vol. 18, pp. 143-152)

A scalable architecture for distributed shared memory multiprocessors using optical interconnects. / Kodi, Avinash Karanth; Louri, Ahmed.

Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM). Vol. 18 2004. p. 143-152.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kodi, AK & Louri, A 2004, A scalable architecture for distributed shared memory multiprocessors using optical interconnects. in Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM). vol. 18, pp. 143-152, Proceedings - 18th International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM), Santa Fe, NM, United States, 4/26/04.
Kodi AK, Louri A. A scalable architecture for distributed shared memory multiprocessors using optical interconnects. In Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM). Vol. 18. 2004. p. 143-152
Kodi, Avinash Karanth ; Louri, Ahmed. / A scalable architecture for distributed shared memory multiprocessors using optical interconnects. Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM). Vol. 18 2004. pp. 143-152
@inproceedings{29c0465640e24d348d5794a61ef1170a,
title = "A scalable architecture for distributed shared memory multiprocessors using optical interconnects",
abstract = "In this paper, we describe the design and analysis of a scalable architecture suitable for large-scale DSMs (Distributed Shared Memory) systems. The approach is based on an interconnect technology which combines optical components and a novel architecture design. In DSM systems, as the network size increases, network contention results in increasing the critical remote memory access latency, which significantly penalizes the performance of DSM systems. In our proposed architecture called RAPID (Reconfigurable and scalable All-Photonic Interconnect for Distributed-shared memory), we provide high connectivity by maximizing the channel availability for remote communication to reduce the remote memory access latency. RAPID also provides fast and efficient unicast, multicast and broadcast capabilities using a combination of aggressively designed wavelength, time and space-division multiplexing techniques. We evaluated RAPID based on network characteristics, power budget criteria and simulation using synthetic traffic workloads and compared it against other scalable electrical networks. We found that RAPID, not only outperforms other networks, but also, satisfies most of the requirements of shared memory multiprocessor design such as low latency, high bandwidth, high connectivity, and easy scalability.",
author = "Kodi, {Avinash Karanth} and Ahmed Louri",
year = "2004",
language = "English (US)",
isbn = "0769521320",
volume = "18",
pages = "143--152",
booktitle = "Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM)",

}

TY - GEN

T1 - A scalable architecture for distributed shared memory multiprocessors using optical interconnects

AU - Kodi, Avinash Karanth

AU - Louri, Ahmed

PY - 2004

Y1 - 2004

N2 - In this paper, we describe the design and analysis of a scalable architecture suitable for large-scale DSMs (Distributed Shared Memory) systems. The approach is based on an interconnect technology which combines optical components and a novel architecture design. In DSM systems, as the network size increases, network contention results in increasing the critical remote memory access latency, which significantly penalizes the performance of DSM systems. In our proposed architecture called RAPID (Reconfigurable and scalable All-Photonic Interconnect for Distributed-shared memory), we provide high connectivity by maximizing the channel availability for remote communication to reduce the remote memory access latency. RAPID also provides fast and efficient unicast, multicast and broadcast capabilities using a combination of aggressively designed wavelength, time and space-division multiplexing techniques. We evaluated RAPID based on network characteristics, power budget criteria and simulation using synthetic traffic workloads and compared it against other scalable electrical networks. We found that RAPID, not only outperforms other networks, but also, satisfies most of the requirements of shared memory multiprocessor design such as low latency, high bandwidth, high connectivity, and easy scalability.

AB - In this paper, we describe the design and analysis of a scalable architecture suitable for large-scale DSMs (Distributed Shared Memory) systems. The approach is based on an interconnect technology which combines optical components and a novel architecture design. In DSM systems, as the network size increases, network contention results in increasing the critical remote memory access latency, which significantly penalizes the performance of DSM systems. In our proposed architecture called RAPID (Reconfigurable and scalable All-Photonic Interconnect for Distributed-shared memory), we provide high connectivity by maximizing the channel availability for remote communication to reduce the remote memory access latency. RAPID also provides fast and efficient unicast, multicast and broadcast capabilities using a combination of aggressively designed wavelength, time and space-division multiplexing techniques. We evaluated RAPID based on network characteristics, power budget criteria and simulation using synthetic traffic workloads and compared it against other scalable electrical networks. We found that RAPID, not only outperforms other networks, but also, satisfies most of the requirements of shared memory multiprocessor design such as low latency, high bandwidth, high connectivity, and easy scalability.

UR - http://www.scopus.com/inward/record.url?scp=12444271118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=12444271118&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:12444271118

SN - 0769521320

SN - 9780769521329

VL - 18

SP - 143

EP - 152

BT - Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM)

ER -