Features of optical interconnects in distributed-shared memory organized MIMD architectures: the ultimate goal

被引:0
|
作者
Frietman, EEE [1 ]
Ernst, RJ [1 ]
Crosbie, P [1 ]
Shimoji, M [1 ]
机构
[1] Delft Univ Technol, Fac Appl Phys, NL-2628 CJ Delft, Netherlands
关键词
D O I
10.1109/ISPAN.1997.645050
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The antipode of a sequential computer using a single CPU to execute the tasks, is a massively parallel processor containing large numbers of computing nodes. Supplemental to the shared and distributed memory organizations are the distributed-shared systems, allowing nodes to explore a single globally shared address space. A different vision is introduced concerning the prospects of an optically coupled version of such a system. In each cluster, an electrical crossbar looks after the interconnections between the nodes, the various memory modules and external I/O channels. The clusters itself are optically coupled through a free space oriented data distributing system. Analogies found in the design of the Convex SPP1000 validate the closeness to reality of such an architecture. Following the preceding clarification also all idealized picture of the fundamental properties of an optically based, fully connected distributed - (virtual) shared memory architecture will be outlined.
引用
收藏
页码:37 / 43
页数:7
相关论文
共 48 条
  • [41] In-NoC circuits for low-latency cache coherence in distributed shared-memory architectures
    Masing, Leonard
    Kress, Fabian
    Anantharajaiah, Nidhi
    Becker, Jurgen
    Srivatsa, Akshay
    Herkersdorf, Andreas
    2018 IEEE 12TH INTERNATIONAL SYMPOSIUM ON EMBEDDED MULTICORE/MANY-CORE SYSTEMS-ON-CHIP (MCSOC 2018), 2018, : 138 - 145
  • [42] Parallel numerical simulation of two-phase flow model in porous media using distributed and shared memory architectures
    Miguel de la Cruz, Luis
    Monsivais, Daniel
    GEOFISICA INTERNACIONAL, 2014, 53 (01): : 59 - 75
  • [43] Multi-Chiplet System Architecture With Shared Uniform Access Memory Based on Board-Level Optical Interconnects
    Sharma, Arastu
    Bamiedakis, Nikolaos
    Karinou, Fotini
    Penty, Richard
    2021 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXPOSITION (OFC), 2021,
  • [44] Predicting the performance measures of an optical distributed shared memory multiprocessor by using support vector regression
    Akay, M. Fatih
    Abasikeles, Ipek
    EXPERT SYSTEMS WITH APPLICATIONS, 2010, 37 (09) : 6293 - 6301
  • [45] Performance evaluation of directory protocols on an optical broadcast-based distributed shared memory multiprocessor
    Abasikeles, Ipek
    Akay, M. Fatih
    COMPUTERS & ELECTRICAL ENGINEERING, 2010, 36 (01) : 114 - 131
  • [46] Scalable 3D hybrid parallel Delaunay image-to-mesh conversion algorithm for distributed shared memory architectures
    Feng, Daming
    Tsolakis, Christos
    Chernikov, Andrey N.
    Chrisochoides, Nikos P.
    COMPUTER-AIDED DESIGN, 2017, 85 : 10 - 19
  • [47] Scalable 3D Hybrid Parallel Delaunay Image-to-Mesh Conversion Algorithm for Distributed Shared Memory Architectures
    Feng, Daming
    Tsolakis, Christos
    Chernikov, Andrey N.
    Chrisochoides, Nikos P.
    24TH INTERNATIONAL MESHING ROUNDTABLE, 2015, 124 : 18 - 30
  • [48] High-Speed Optical Cache Memory as Single-Level Shared Cache in Chip-Multiprocessor architectures
    Maniotis, P.
    Gitzenis, S.
    Tassiulas, L.
    Pleros, N.
    2015 WORKSHOP ON EXPLOITING SILICON PHOTONICS FOR ENERGY-EFFICIENT HIGH PERFORMANCE COMPUTING (SIPHOTONICS), 2014, : 1 - 8