Processor allocation scheme for hypercube computers

Jong Kim, Chitaranjan Das, Woei Lin

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

A processor allocation policy for hypercube computers is presented. The allocation policy is called free list since it maintains a list of free subcubes available in the system. An incoming request of dimension k (2k nodes) is allocated by finding a free subcube of dimension k. The free list policy is different from other hypercube allocation schemes in that it uses a top-down allocation rule in contrast to the bottom-up approach used by all other allocations. This new policy is compared with the buddy, the gray code, and the modified buddy allocation policies reported for the hypercubes. It is shown that the free list policy is statically optimal, as are the other policies, but it gives better subcube recognition ability compared to the previous schemes in a dynamic environment. The performance of this policy, in terms of parameters such as the time complexity, average delay, and system utilization, is compared with those of the other schemes, demonstrating its effectiveness.

Original languageEnglish (US)
Pages (from-to)231-238
Number of pages8
JournalProceedings of the International Conference on Parallel Processing
Volume2
StatePublished - 1989

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

@article{3035200511504cf381b576e5f2920c15,
title = "Processor allocation scheme for hypercube computers",
abstract = "A processor allocation policy for hypercube computers is presented. The allocation policy is called free list since it maintains a list of free subcubes available in the system. An incoming request of dimension k (2k nodes) is allocated by finding a free subcube of dimension k. The free list policy is different from other hypercube allocation schemes in that it uses a top-down allocation rule in contrast to the bottom-up approach used by all other allocations. This new policy is compared with the buddy, the gray code, and the modified buddy allocation policies reported for the hypercubes. It is shown that the free list policy is statically optimal, as are the other policies, but it gives better subcube recognition ability compared to the previous schemes in a dynamic environment. The performance of this policy, in terms of parameters such as the time complexity, average delay, and system utilization, is compared with those of the other schemes, demonstrating its effectiveness.",
author = "Jong Kim and Chitaranjan Das and Woei Lin",
year = "1989",
language = "English (US)",
volume = "2",
pages = "231--238",
journal = "Proceedings of the International Conference on Parallel Processing",
issn = "0190-3918",

}

Processor allocation scheme for hypercube computers. / Kim, Jong; Das, Chitaranjan; Lin, Woei.

In: Proceedings of the International Conference on Parallel Processing, Vol. 2, 1989, p. 231-238.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Processor allocation scheme for hypercube computers

AU - Kim, Jong

AU - Das, Chitaranjan

AU - Lin, Woei

PY - 1989

Y1 - 1989

N2 - A processor allocation policy for hypercube computers is presented. The allocation policy is called free list since it maintains a list of free subcubes available in the system. An incoming request of dimension k (2k nodes) is allocated by finding a free subcube of dimension k. The free list policy is different from other hypercube allocation schemes in that it uses a top-down allocation rule in contrast to the bottom-up approach used by all other allocations. This new policy is compared with the buddy, the gray code, and the modified buddy allocation policies reported for the hypercubes. It is shown that the free list policy is statically optimal, as are the other policies, but it gives better subcube recognition ability compared to the previous schemes in a dynamic environment. The performance of this policy, in terms of parameters such as the time complexity, average delay, and system utilization, is compared with those of the other schemes, demonstrating its effectiveness.

AB - A processor allocation policy for hypercube computers is presented. The allocation policy is called free list since it maintains a list of free subcubes available in the system. An incoming request of dimension k (2k nodes) is allocated by finding a free subcube of dimension k. The free list policy is different from other hypercube allocation schemes in that it uses a top-down allocation rule in contrast to the bottom-up approach used by all other allocations. This new policy is compared with the buddy, the gray code, and the modified buddy allocation policies reported for the hypercubes. It is shown that the free list policy is statically optimal, as are the other policies, but it gives better subcube recognition ability compared to the previous schemes in a dynamic environment. The performance of this policy, in terms of parameters such as the time complexity, average delay, and system utilization, is compared with those of the other schemes, demonstrating its effectiveness.

UR - http://www.scopus.com/inward/record.url?scp=0024886934&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024886934&partnerID=8YFLogxK

M3 - Article

VL - 2

SP - 231

EP - 238

JO - Proceedings of the International Conference on Parallel Processing

JF - Proceedings of the International Conference on Parallel Processing

SN - 0190-3918

ER -