CaltechAUTHORS
  A Caltech Library Service

Asymptotically optimal load balancing in large-scale heterogeneous systems with multiple dispatchers

Zhou, Xingyu and Shroff, Ness and Wierman, Adam (2021) Asymptotically optimal load balancing in large-scale heterogeneous systems with multiple dispatchers. Performance Evaluation, 145 . Art. No. 102146. ISSN 0166-5316. doi:10.1016/j.peva.2020.102146. https://resolver.caltech.edu/CaltechAUTHORS:20200526-152856053

[img] PDF - Accepted Version
See Usage Policy.

586kB
[img] PDF - Submitted Version
Creative Commons Attribution.

344kB
[img] PDF (Supplementary data) - Supplemental Material
See Usage Policy.

198kB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200526-152856053

Abstract

We consider the load balancing problem in large-scale heterogeneous systems with multiple dispatchers. We introduce a general framework called Local-Estimation-Driven (LED). Under this framework, each dispatcher keeps local (possibly outdated) estimates of the queue lengths for all the servers, and the dispatching decision is made purely based on these local estimates. The local estimates are updated via infrequent communications between dispatchers and servers. We derive sufficient conditions for LED policies to achieve throughput optimality and delay optimality in heavy-traffic, respectively. These conditions directly imply delay optimality for many previous local-memory based policies in heavy traffic. Moreover, the results enable us to design new delay optimal policies for heterogeneous systems with multiple dispatchers. Finally, the heavy-traffic delay optimality of the LED framework also sheds light on a recent open question on how to design optimal load balancing schemes using delayed information.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1016/j.peva.2020.102146DOIArticle
https://arxiv.org/abs/2002.08908arXivDiscussion Paper
https://resolver.caltech.edu/CaltechAUTHORS:20210308-133521968Related ItemPerformance Evaluation Review Journal (ACM)
Additional Information:© 2020 Elsevier B.V. Received 24 September 2020, Accepted 25 September 2020, Available online 8 October 2020. This project has been funded in part through NSF, USA grants: CNS-2007231, CNS-1719371, and CNS-1717060 and NSF, USA grants AitF-1637598 and CNS-1518941. Declaration of Competing Interest: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: S. Theja Maguluri, Georgia Institute of Technology, Atlanta, Georgia, United States C.H. Xia, OHIO STATE UNIVERSITY, Columbus, Ohio, United States.
Funders:
Funding AgencyGrant Number
NSFCNS-2007231
NSFCNS-1719371
NSFCNS-1717060
NSFCCF-1637598
NSFCNS-1518941
Subject Keywords:Asymptotically optimal; Load balancing; Heterogeneous systems; Multiple dispatchers; Delayed information
DOI:10.1016/j.peva.2020.102146
Record Number:CaltechAUTHORS:20200526-152856053
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200526-152856053
Official Citation:Xingyu Zhou, Ness Shroff, Adam Wierman, Asymptotically optimal load balancing in large-scale heterogeneous systems with multiple dispatchers, Performance Evaluation, Volume 145, 2021, 102146, ISSN 0166-5316, https://doi.org/10.1016/j.peva.2020.102146. (http://www.sciencedirect.com/science/article/pii/S0166531620300663)
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:103477
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:26 May 2020 22:32
Last Modified:16 Nov 2021 18:21

Repository Staff Only: item control page