Limit allocation: An efficient processor management scheme for hypercubes

Chansu Yu, Chitaranjan Das

Research output: Contribution to journalConference article

9 Citations (Scopus)

Abstract

Efficient task management in a hypercube multi-processor becomes difficult due to system overflow, where an incoming job cannot be allocated in spite of a sufficient number of free processors. Overflow occurs either due to the inability of recognizing a free subcube or due to external fragmentation. In this paper, we propose an allocation strategy that tries to scale down an incoming job size if it cannot fit into a fragmented hypercube. We call it limit allocation. We discuss three simple schemes, Limit-k, Greedy and Average. We conduct both analysis and simulation to characterize and compare various allocation policies. An M/M/m queueing model is developed to predict the behavior of buddy, free list and limit-k policies. The simulation study shows that the two adaptive schemes, greedy and average, outperform all other schemes reported so far in the literature.

Original languageEnglish (US)
Article number5727776
JournalProceedings of the International Conference on Parallel Processing
Volume2
DOIs
StatePublished - Jan 1 1994
Event23rd International Conference on Parallel Processing, ICPP 1994 - Raleigh, NC, United States
Duration: Aug 15 1994Aug 19 1994

Fingerprint

Hypercube
Overflow
Queueing Model
Multiprocessor
Fragmentation
Simulation Study
Sufficient
Predict
Simulation
Policy

All Science Journal Classification (ASJC) codes

  • Software
  • Mathematics(all)
  • Hardware and Architecture

Cite this

@article{bb7ef809dd4e441f869f768e62a544b5,
title = "Limit allocation: An efficient processor management scheme for hypercubes",
abstract = "Efficient task management in a hypercube multi-processor becomes difficult due to system overflow, where an incoming job cannot be allocated in spite of a sufficient number of free processors. Overflow occurs either due to the inability of recognizing a free subcube or due to external fragmentation. In this paper, we propose an allocation strategy that tries to scale down an incoming job size if it cannot fit into a fragmented hypercube. We call it limit allocation. We discuss three simple schemes, Limit-k, Greedy and Average. We conduct both analysis and simulation to characterize and compare various allocation policies. An M/M/m queueing model is developed to predict the behavior of buddy, free list and limit-k policies. The simulation study shows that the two adaptive schemes, greedy and average, outperform all other schemes reported so far in the literature.",
author = "Chansu Yu and Chitaranjan Das",
year = "1994",
month = "1",
day = "1",
doi = "10.1109/ICPP.1994.124",
language = "English (US)",
volume = "2",
journal = "Proceedings of the International Conference on Parallel Processing",
issn = "0190-3918",

}

Limit allocation : An efficient processor management scheme for hypercubes. / Yu, Chansu; Das, Chitaranjan.

In: Proceedings of the International Conference on Parallel Processing, Vol. 2, 5727776, 01.01.1994.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Limit allocation

T2 - An efficient processor management scheme for hypercubes

AU - Yu, Chansu

AU - Das, Chitaranjan

PY - 1994/1/1

Y1 - 1994/1/1

N2 - Efficient task management in a hypercube multi-processor becomes difficult due to system overflow, where an incoming job cannot be allocated in spite of a sufficient number of free processors. Overflow occurs either due to the inability of recognizing a free subcube or due to external fragmentation. In this paper, we propose an allocation strategy that tries to scale down an incoming job size if it cannot fit into a fragmented hypercube. We call it limit allocation. We discuss three simple schemes, Limit-k, Greedy and Average. We conduct both analysis and simulation to characterize and compare various allocation policies. An M/M/m queueing model is developed to predict the behavior of buddy, free list and limit-k policies. The simulation study shows that the two adaptive schemes, greedy and average, outperform all other schemes reported so far in the literature.

AB - Efficient task management in a hypercube multi-processor becomes difficult due to system overflow, where an incoming job cannot be allocated in spite of a sufficient number of free processors. Overflow occurs either due to the inability of recognizing a free subcube or due to external fragmentation. In this paper, we propose an allocation strategy that tries to scale down an incoming job size if it cannot fit into a fragmented hypercube. We call it limit allocation. We discuss three simple schemes, Limit-k, Greedy and Average. We conduct both analysis and simulation to characterize and compare various allocation policies. An M/M/m queueing model is developed to predict the behavior of buddy, free list and limit-k policies. The simulation study shows that the two adaptive schemes, greedy and average, outperform all other schemes reported so far in the literature.

UR - http://www.scopus.com/inward/record.url?scp=84904310496&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904310496&partnerID=8YFLogxK

U2 - 10.1109/ICPP.1994.124

DO - 10.1109/ICPP.1994.124

M3 - Conference article

AN - SCOPUS:84904310496

VL - 2

JO - Proceedings of the International Conference on Parallel Processing

JF - Proceedings of the International Conference on Parallel Processing

SN - 0190-3918

M1 - 5727776

ER -