Nonuniform traffic spots (NUTS) in multistage interconnection networks

Lang, T; Kurisaki, L

doi:10.1016/0743-7315(90)90006-B

Title: Nonuniform traffic spots (NUTS) in multistage interconnection networks

Full Record
Other Related Research

Abstract

The performance of multistage interconnection networks for multiprocessors is degraded when the traffic pattern produces nonuniform congestion in the blocking switches, that is, when there exist nonuniform traffic spots. For some specific patterns the authors evaluate this degradation in performance and propose modifications to the network organization and operation to reduce the degradation. Successful modifications are the use of diverting switches and the extension of the network with additional links. The use of these modifications makes the network more effective for a larger variety of traffic patterns. The authors also consider the case in which the network carries the superposition of two types of traffic. One type is the high throughput data and instruction traffic, while the other consists of control and I/O packets which are of low throughput but have severe real-time constraints. The authors conclude that diverting switches and networks with additional links are also suitable for assuring low latency for the real-time traffic, especially when using the displacing mode.

Authors:

Lang, T; Kurisaki, L ^[1]

California Univ., Los Angeles, CA (USA). Dept. of Computer Science

Publication Date:: Sat Sep 01 00:00:00 EDT 1990

OSTI Identifier:: 6467833

Resource Type:: Journal Article

Journal Name:: Journal of Parallel and Distributed Computing; (USA)

Additional Journal Information:: Journal Volume: 10:1; Journal ID: ISSN 0743-7315

Country of Publication:: United States

Language:: English

Subject:: 99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; ARRAY PROCESSORS; COMPUTER NETWORKS; PARALLEL PROCESSING; PERFORMANCE; DESIGN; MODIFICATIONS; PROGRAMMING; 990200* - Mathematics & Computers

Citation Formats


                    Lang, T, and Kurisaki, L. Nonuniform traffic spots (NUTS) in multistage interconnection networks.  United States: N. p., 1990. 
        Web.  doi:10.1016/0743-7315(90)90006-B.

Copy to clipboard


                    Lang, T, & Kurisaki, L. Nonuniform traffic spots (NUTS) in multistage interconnection networks.  United States.  https://doi.org/10.1016/0743-7315(90)90006-B

Copy to clipboard


                    Lang, T, and Kurisaki, L. 1990.  
        "Nonuniform traffic spots (NUTS) in multistage interconnection networks".  United States.  https://doi.org/10.1016/0743-7315(90)90006-B.

Copy to clipboard


                    
@article{osti_6467833,

  title        = {Nonuniform traffic spots (NUTS) in multistage interconnection networks},

  author       = {Lang, T and Kurisaki, L},

  abstractNote = {The performance of multistage interconnection networks for multiprocessors is degraded when the traffic pattern produces nonuniform congestion in the blocking switches, that is, when there exist nonuniform traffic spots. For some specific patterns the authors evaluate this degradation in performance and propose modifications to the network organization and operation to reduce the degradation. Successful modifications are the use of diverting switches and the extension of the network with additional links. The use of these modifications makes the network more effective for a larger variety of traffic patterns. The authors also consider the case in which the network carries the superposition of two types of traffic. One type is the high throughput data and instruction traffic, while the other consists of control and I/O packets which are of low throughput but have severe real-time constraints. The authors conclude that diverting switches and networks with additional links are also suitable for assuring low latency for the real-time traffic, especially when using the displacing mode.},

  doi          = {10.1016/0743-7315(90)90006-B},

  url          = {https://www.osti.gov/biblio/6467833},
  journal      = {Journal of Parallel and Distributed Computing; (USA)},

  issn         = {0743-7315},

  number       = ,

  volume       = 10:1,

  place        = {United States},

  year         = {Sat Sep 01 00:00:00 EDT 1990},

  month        = {Sat Sep 01 00:00:00 EDT 1990}

}

Copy to clipboard

Journal Article:

https://doi.org/10.1016/0743-7315(90)90006-B

Other availability

Find in Google Scholar

Search WorldCat to find libraries that may hold this journal

Save / Share:

Export Metadata

Save to My Library

Similar records in OSTI.GOV collections:

Multistage interconnection networks with multiple outlets

Conference Hanawa, Toshihiro; Amano, Hideharu; Fujikawa, Yoshifumi

Multistage Interconnection Networks (MINs) with multiple outlets are networks which can support higher bandwidth than that of nonblocking networks by passing multiple packets to the same destination. A novel MIN topology with multiple outlets called Piled Banyan Switching Fabrics (PBSF) is proposed for the Simple Serial Synchronized (SSS)-MIN used in multiprocessors, and analyzed with other two types of MIN with multiple outlets called Multi-Banyan Switching Fabrics (MBSF) and Tandem Banyan Switching Fabrics (TBSF). The throughput of these MINs is evaluated and compared with both the theoretical model and simulation. The PBSF supports the best throughput and latency used for themore »« less
Dynamics of TCP traffic over ATM networks

Conference Floyd, S; Romanow, A

The authors investigate the performance of TCP (Transport Control Protocol) connections over ATM (Asynchronous Transfer Mode) networks without ATM-level congestion control, and compare it to the performance of TCP over packet-based networks. For simulations of congested networks, the effective throughput of TCP over ATM can be quite low when cells are dropped at the congested ATM switch. The low throughput is due to wasted bandwidth as the congested link transmits cells from ``corrupted`` packets, i.e., packets in which at least one cell is dropped by the switch. This fragmentation effect can be corrected and high throughput can be achieved ifmore »« less
Full Text Available
The performance of the Cedar multistage switching network

Book Torrellas, J; Zhang, Z

While multistage switching networks for vector multiprocessors have been studied extensively, detailed evaluations of their performance are rare. Indeed, analytical models, simulations with pseudo-synthetic loads, studies focused on average-value parameters, and measurements of networks disconnected from the machine, all provide limited information. In this paper, instead, the authors present an in-depth empirical analysis of a multistage switching network in a realistic setting: they use hardware probes to examine the performance of the omega network of the Cedar shared memory machine executing real applications. The machine is configured with 16 vector processors. The analysis suggests that the performance of multistage switchingmore »« less
SNAIL: A multiprocessor based on the simple serial synchronized multistage interconnection network architecture

Conference Sasahara, M; Terada, J; Zhou, L; ...

Simple Serial Synchronized (SSS) Multistage Interconnection Network (MIN) is a novel MIN architecture for connecting processors and memory modules in multiprocessors. Synchronized bit-serial communication simplifies the structure/control, and also solves the pin-limitation problem. Here, design, implementation, and evaluation of a multiprocessor prototype called SNAIL with the SSS-MIN are presented. The heart of SNAIL is the prototype 1 {mu} CMOS SSS-MIN gate array chip which exchanges packets from 16 inputs with 50MHz clock. The message combining is implemented only with 20% increases of the hardware. From the empirical evaluation with some application programs, it appears that the latency and synchronization overheadmore »« less
Traffic-specific interconnection networks for multicomputers

Journal Article Nichols, K; Messerschmitt, D - IEEE Trans. Comput.; (United States)

Special-purpose networks of processors and intelligent switching devices can be programmed to solve problems with inherent parallelism. The target algorithm is coded into a number of asynchronously executing parallel tasks, and assigned to specific processors. For well-known algorithms, information about the data communicated between these tasks (traffic) can be characterized and used by a heuristic algorithm to produce a specialized interconnection network of switching devices at a lower cost and increased performance over the use of a general purpose structure. A program was developed which uses intertask communications to automatically generate interconnection structures of switch clips (SC's). Simulation results weremore »« less
https://doi.org/10.1109/TC.1987.1676859

Similar Records