Analysis of SITA policies

Eitan Bachmat; Hagit Sarfati

doi:10.1016/j.peva.2009.09.007

Analysis of SITA policies

Eitan Bachmat, Hagit Sarfati

Department of Computer Science

Research output: Contribution to journal › Article › peer-review

24 Scopus citations

Abstract

We analyze the performance of Size Interval Task Assignment (SITA) policies, for multi-host assignment in a non-preemptive environment. Assuming Poisson arrivals, we provide general bounds on the average waiting time independent of the job size distribution. We establish a general duality theory for the performance of SITA policies. We provide a detailed analysis of the performance of SITA systems when the job size distribution is Bounded Pareto and the range of job sizes tends to infinity. In particular, we determine asymptotically optimal cutoff values and provide asymptotic formulas for average waiting time and slowdown. We compare the results with the Least Work Remaining policy and compute which policy is asymptotically better for any given set of parameters. In the case of inhomogeneous hosts, we determine their optimal ordering.

Original language	English
Pages (from-to)	102-120
Number of pages	19
Journal	Performance Evaluation
Volume	67
Issue number	2
DOIs	https://doi.org/10.1016/j.peva.2009.09.007
State	Published - 1 Feb 2010

Keywords

Heavy-tailed distributions
Multiple host scheduling policies
Queueing theory
Size interval splitting policies

ASJC Scopus subject areas

Software
Modeling and Simulation
Hardware and Architecture
Computer Networks and Communications

Access to Document

10.1016/j.peva.2009.09.007

Cite this

@article{5d255d1c0f57423f9e6103f90aa87311,

title = "Analysis of SITA policies",

abstract = "We analyze the performance of Size Interval Task Assignment (SITA) policies, for multi-host assignment in a non-preemptive environment. Assuming Poisson arrivals, we provide general bounds on the average waiting time independent of the job size distribution. We establish a general duality theory for the performance of SITA policies. We provide a detailed analysis of the performance of SITA systems when the job size distribution is Bounded Pareto and the range of job sizes tends to infinity. In particular, we determine asymptotically optimal cutoff values and provide asymptotic formulas for average waiting time and slowdown. We compare the results with the Least Work Remaining policy and compute which policy is asymptotically better for any given set of parameters. In the case of inhomogeneous hosts, we determine their optimal ordering.",

keywords = "Heavy-tailed distributions, Multiple host scheduling policies, Queueing theory, Size interval splitting policies",

author = "Eitan Bachmat and Hagit Sarfati",

year = "2010",

month = feb,

day = "1",

doi = "10.1016/j.peva.2009.09.007",

language = "English",

volume = "67",

pages = "102--120",

journal = "Performance Evaluation",

issn = "0166-5316",

publisher = "Elsevier B.V.",

number = "2",

}

TY - JOUR

T1 - Analysis of SITA policies

AU - Bachmat, Eitan

AU - Sarfati, Hagit

PY - 2010/2/1

Y1 - 2010/2/1

N2 - We analyze the performance of Size Interval Task Assignment (SITA) policies, for multi-host assignment in a non-preemptive environment. Assuming Poisson arrivals, we provide general bounds on the average waiting time independent of the job size distribution. We establish a general duality theory for the performance of SITA policies. We provide a detailed analysis of the performance of SITA systems when the job size distribution is Bounded Pareto and the range of job sizes tends to infinity. In particular, we determine asymptotically optimal cutoff values and provide asymptotic formulas for average waiting time and slowdown. We compare the results with the Least Work Remaining policy and compute which policy is asymptotically better for any given set of parameters. In the case of inhomogeneous hosts, we determine their optimal ordering.

AB - We analyze the performance of Size Interval Task Assignment (SITA) policies, for multi-host assignment in a non-preemptive environment. Assuming Poisson arrivals, we provide general bounds on the average waiting time independent of the job size distribution. We establish a general duality theory for the performance of SITA policies. We provide a detailed analysis of the performance of SITA systems when the job size distribution is Bounded Pareto and the range of job sizes tends to infinity. In particular, we determine asymptotically optimal cutoff values and provide asymptotic formulas for average waiting time and slowdown. We compare the results with the Least Work Remaining policy and compute which policy is asymptotically better for any given set of parameters. In the case of inhomogeneous hosts, we determine their optimal ordering.

KW - Heavy-tailed distributions

KW - Multiple host scheduling policies

KW - Queueing theory

KW - Size interval splitting policies

UR - http://www.scopus.com/inward/record.url?scp=70450245332&partnerID=8YFLogxK

U2 - 10.1016/j.peva.2009.09.007

DO - 10.1016/j.peva.2009.09.007

M3 - Article

AN - SCOPUS:70450245332

SN - 0166-5316

VL - 67

SP - 102

EP - 120

JO - Performance Evaluation

JF - Performance Evaluation

IS - 2

ER -

Analysis of SITA policies

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this