Job assignment in large-scale service systems with affinity relations

Ellen Cardinaels (Corresponding author), Sem C. Borst, Johan S.H. van Leeuwaarden

Research output: Contribution to journalArticleAcademicpeer-review

6 Downloads (Pure)

Abstract

We consider load balancing in service systems with affinity relations between jobs and servers. Specifically, an arriving job can be assigned to a fast, primary server from a particular selection associated with this job or to a secondary server to be processed at a slower rate. Such job–server affinity relations can model network topologies based on geographical proximity, or data locality in cloud scenarios. We introduce load balancing schemes that assign jobs to primary servers if available, and otherwise to secondary servers. A novel coupling construction is developed to obtain stability conditions and performance bounds. We also conduct a fluid limit analysis for symmetric model instances, which reveals a delicate interplay between the model parameters and load balancing performance.

Original languageEnglish
Pages (from-to)227-268
Number of pages42
JournalQueueing Systems
Volume93
Issue number3-4
DOIs
Publication statusPublished - Dec 2019

Fingerprint

Affine transformation
Assignment
Servers
Server
Load Balancing
Resource allocation
Fluid Limits
Data Locality
Limit Analysis
Performance Bounds
Stability Condition
Network Topology
Proximity
Assign
Topology
Service system
Job assignment
Model
Scenarios
Fluids

Keywords

  • Fluid limit
  • Job scheduling
  • Load balancing
  • Network topology
  • Stochastic coupling

Cite this

@article{d0fe1b7f5e5d4557a5c01532fae20b41,
title = "Job assignment in large-scale service systems with affinity relations",
abstract = "We consider load balancing in service systems with affinity relations between jobs and servers. Specifically, an arriving job can be assigned to a fast, primary server from a particular selection associated with this job or to a secondary server to be processed at a slower rate. Such job–server affinity relations can model network topologies based on geographical proximity, or data locality in cloud scenarios. We introduce load balancing schemes that assign jobs to primary servers if available, and otherwise to secondary servers. A novel coupling construction is developed to obtain stability conditions and performance bounds. We also conduct a fluid limit analysis for symmetric model instances, which reveals a delicate interplay between the model parameters and load balancing performance.",
keywords = "Fluid limit, Job scheduling, Load balancing, Network topology, Stochastic coupling",
author = "Ellen Cardinaels and Borst, {Sem C.} and {van Leeuwaarden}, {Johan S.H.}",
year = "2019",
month = "12",
doi = "10.1007/s11134-019-09633-y",
language = "English",
volume = "93",
pages = "227--268",
journal = "Queueing Systems: Theory and Applications",
issn = "0257-0130",
publisher = "Springer",
number = "3-4",

}

Job assignment in large-scale service systems with affinity relations. / Cardinaels, Ellen (Corresponding author); Borst, Sem C.; van Leeuwaarden, Johan S.H.

In: Queueing Systems, Vol. 93, No. 3-4, 12.2019, p. 227-268.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Job assignment in large-scale service systems with affinity relations

AU - Cardinaels, Ellen

AU - Borst, Sem C.

AU - van Leeuwaarden, Johan S.H.

PY - 2019/12

Y1 - 2019/12

N2 - We consider load balancing in service systems with affinity relations between jobs and servers. Specifically, an arriving job can be assigned to a fast, primary server from a particular selection associated with this job or to a secondary server to be processed at a slower rate. Such job–server affinity relations can model network topologies based on geographical proximity, or data locality in cloud scenarios. We introduce load balancing schemes that assign jobs to primary servers if available, and otherwise to secondary servers. A novel coupling construction is developed to obtain stability conditions and performance bounds. We also conduct a fluid limit analysis for symmetric model instances, which reveals a delicate interplay between the model parameters and load balancing performance.

AB - We consider load balancing in service systems with affinity relations between jobs and servers. Specifically, an arriving job can be assigned to a fast, primary server from a particular selection associated with this job or to a secondary server to be processed at a slower rate. Such job–server affinity relations can model network topologies based on geographical proximity, or data locality in cloud scenarios. We introduce load balancing schemes that assign jobs to primary servers if available, and otherwise to secondary servers. A novel coupling construction is developed to obtain stability conditions and performance bounds. We also conduct a fluid limit analysis for symmetric model instances, which reveals a delicate interplay between the model parameters and load balancing performance.

KW - Fluid limit

KW - Job scheduling

KW - Load balancing

KW - Network topology

KW - Stochastic coupling

UR - http://www.scopus.com/inward/record.url?scp=85074490496&partnerID=8YFLogxK

U2 - 10.1007/s11134-019-09633-y

DO - 10.1007/s11134-019-09633-y

M3 - Article

AN - SCOPUS:85074490496

VL - 93

SP - 227

EP - 268

JO - Queueing Systems: Theory and Applications

JF - Queueing Systems: Theory and Applications

SN - 0257-0130

IS - 3-4

ER -