An efficient and fair multi-resource allocation mechanism for heterogeneous servers

Jalal Khamse-Ashari, Ioannis Lambadaris, George Kesidis, Bhuvan Urgaonkar, Yiqiang Zhao

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Efficient and fair allocation of multiple types of resources is a crucial objective in a cloud/distributed computing cluster. Users may have diverse resource needs. Furthermore, diversity in server properties/capabilities may mean that only a subset of servers may be usable by a given user. In platforms with such heterogeneity, we identify important limitations in existing multi-resource fair allocation mechanisms, notably Dominant Resource Fairness and its follow-up work. To overcome such limitations, we propose a new server-based approach; each server allocates resources by maximizing a per-server utility function. We propose a specific class of utility functions which, when appropriately parameterized, adjusts the trade-off between efficiency and fairness, and captures a variety of fairness measures (such as our recently proposed Per-Server Dominant Share Fairness ). We establish conditions for the proposed mechanism to satisfy certain properties that are generally deemed desirable, e.g., envy-freeness, sharing incentive, bottleneck fairness, and Pareto optimality. To implement our resource allocation mechanism, we develop an iterative algorithm which is shown to be globally convergent. Subsequently, we show how the proposed mechanism could be implemented in a distributed fashion. Finally, we carry out extensive trace-driven simulations to show the enhanced performance of our proposed mechanism over the existing ones.

Original languageEnglish (US)
Article number8368291
Pages (from-to)2686-2699
Number of pages14
JournalIEEE Transactions on Parallel and Distributed Systems
Volume29
Issue number12
DOIs
StatePublished - Dec 1 2018

Fingerprint

Resource allocation
Servers
Distributed computer systems

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

@article{e010ef5f1b2448ae942fd99f9026d37d,
title = "An efficient and fair multi-resource allocation mechanism for heterogeneous servers",
abstract = "Efficient and fair allocation of multiple types of resources is a crucial objective in a cloud/distributed computing cluster. Users may have diverse resource needs. Furthermore, diversity in server properties/capabilities may mean that only a subset of servers may be usable by a given user. In platforms with such heterogeneity, we identify important limitations in existing multi-resource fair allocation mechanisms, notably Dominant Resource Fairness and its follow-up work. To overcome such limitations, we propose a new server-based approach; each server allocates resources by maximizing a per-server utility function. We propose a specific class of utility functions which, when appropriately parameterized, adjusts the trade-off between efficiency and fairness, and captures a variety of fairness measures (such as our recently proposed Per-Server Dominant Share Fairness ). We establish conditions for the proposed mechanism to satisfy certain properties that are generally deemed desirable, e.g., envy-freeness, sharing incentive, bottleneck fairness, and Pareto optimality. To implement our resource allocation mechanism, we develop an iterative algorithm which is shown to be globally convergent. Subsequently, we show how the proposed mechanism could be implemented in a distributed fashion. Finally, we carry out extensive trace-driven simulations to show the enhanced performance of our proposed mechanism over the existing ones.",
author = "Jalal Khamse-Ashari and Ioannis Lambadaris and George Kesidis and Bhuvan Urgaonkar and Yiqiang Zhao",
year = "2018",
month = "12",
day = "1",
doi = "10.1109/TPDS.2018.2841915",
language = "English (US)",
volume = "29",
pages = "2686--2699",
journal = "IEEE Transactions on Parallel and Distributed Systems",
issn = "1045-9219",
publisher = "IEEE Computer Society",
number = "12",

}

An efficient and fair multi-resource allocation mechanism for heterogeneous servers. / Khamse-Ashari, Jalal; Lambadaris, Ioannis; Kesidis, George; Urgaonkar, Bhuvan; Zhao, Yiqiang.

In: IEEE Transactions on Parallel and Distributed Systems, Vol. 29, No. 12, 8368291, 01.12.2018, p. 2686-2699.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An efficient and fair multi-resource allocation mechanism for heterogeneous servers

AU - Khamse-Ashari, Jalal

AU - Lambadaris, Ioannis

AU - Kesidis, George

AU - Urgaonkar, Bhuvan

AU - Zhao, Yiqiang

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Efficient and fair allocation of multiple types of resources is a crucial objective in a cloud/distributed computing cluster. Users may have diverse resource needs. Furthermore, diversity in server properties/capabilities may mean that only a subset of servers may be usable by a given user. In platforms with such heterogeneity, we identify important limitations in existing multi-resource fair allocation mechanisms, notably Dominant Resource Fairness and its follow-up work. To overcome such limitations, we propose a new server-based approach; each server allocates resources by maximizing a per-server utility function. We propose a specific class of utility functions which, when appropriately parameterized, adjusts the trade-off between efficiency and fairness, and captures a variety of fairness measures (such as our recently proposed Per-Server Dominant Share Fairness ). We establish conditions for the proposed mechanism to satisfy certain properties that are generally deemed desirable, e.g., envy-freeness, sharing incentive, bottleneck fairness, and Pareto optimality. To implement our resource allocation mechanism, we develop an iterative algorithm which is shown to be globally convergent. Subsequently, we show how the proposed mechanism could be implemented in a distributed fashion. Finally, we carry out extensive trace-driven simulations to show the enhanced performance of our proposed mechanism over the existing ones.

AB - Efficient and fair allocation of multiple types of resources is a crucial objective in a cloud/distributed computing cluster. Users may have diverse resource needs. Furthermore, diversity in server properties/capabilities may mean that only a subset of servers may be usable by a given user. In platforms with such heterogeneity, we identify important limitations in existing multi-resource fair allocation mechanisms, notably Dominant Resource Fairness and its follow-up work. To overcome such limitations, we propose a new server-based approach; each server allocates resources by maximizing a per-server utility function. We propose a specific class of utility functions which, when appropriately parameterized, adjusts the trade-off between efficiency and fairness, and captures a variety of fairness measures (such as our recently proposed Per-Server Dominant Share Fairness ). We establish conditions for the proposed mechanism to satisfy certain properties that are generally deemed desirable, e.g., envy-freeness, sharing incentive, bottleneck fairness, and Pareto optimality. To implement our resource allocation mechanism, we develop an iterative algorithm which is shown to be globally convergent. Subsequently, we show how the proposed mechanism could be implemented in a distributed fashion. Finally, we carry out extensive trace-driven simulations to show the enhanced performance of our proposed mechanism over the existing ones.

UR - http://www.scopus.com/inward/record.url?scp=85047828547&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047828547&partnerID=8YFLogxK

U2 - 10.1109/TPDS.2018.2841915

DO - 10.1109/TPDS.2018.2841915

M3 - Article

AN - SCOPUS:85047828547

VL - 29

SP - 2686

EP - 2699

JO - IEEE Transactions on Parallel and Distributed Systems

JF - IEEE Transactions on Parallel and Distributed Systems

SN - 1045-9219

IS - 12

M1 - 8368291

ER -