Stochastic resource allocation

Liran Funaro; Orna Agmon Ben-Yehuda; Assaf Schuster

doi:10.1145/3313808.3313815

Stochastic resource allocation

Liran Funaro, Orna Agmon Ben-Yehuda, Assaf Schuster

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Suboptimal resource utilization among public and private cloud providers prevents them from maximizing their economic potential. Long-term allocated resources are often idle when they might have been subleased for a short period. Alternatively, arbitrary resource overcommitment may lead to unpredictable client performance. We propose a mechanism for fixed availability (traditional) resource allocation alongside stochastic resource allocation in the form of shares. We show its benefit for private and public cloud providers and for a wide range of clients. Our simulations show that our mechanism can increase server consolidation by 5.6 times on average compared with selling only fixed performance resources, and by 1.7 times compared with burstable instances, which is the most prevalent flexible allocation method. Our mechanism also yields better performance (i.e., higher revenues) or a lower cost than burstable instances for a wide range of clients, making it more profitable for them.

Original language	English
Title of host publication	VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments
Editors	Jennifer B. Sartor, Mayur Naik, Jennifer B. Sartor
Publisher	Association for Computing Machinery, Inc
Pages	122-136
Number of pages	15
ISBN (Electronic)	9781450360203
DOIs	https://doi.org/10.1145/3313808.3313815
State	Published - 14 Apr 2019
Externally published	Yes
Event	15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, VEE 2019 - Providence, United States Duration: 14 Apr 2019 → …

Publication series

Name	VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments

Conference

Conference	15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, VEE 2019
Country/Territory	United States
City	Providence
Period	14/04/19 → …

Bibliographical note

Publisher Copyright:
© 2019 Copyright held by the owner/author(s). Publication rights licensed to ACM.

Keywords

Cloud
Pricing
Resource-Allocation
Scheduler

ASJC Scopus subject areas

Software
Computer Networks and Communications
Hardware and Architecture

Access to Document

10.1145/3313808.3313815

Cite this

Funaro, L., Ben-Yehuda, O. A., & Schuster, A. (2019). Stochastic resource allocation. In J. B. Sartor, M. Naik, & J. B. Sartor (Eds.), VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (pp. 122-136). (VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments). Association for Computing Machinery, Inc. https://doi.org/10.1145/3313808.3313815

Funaro, Liran ; Ben-Yehuda, Orna Agmon ; Schuster, Assaf. / Stochastic resource allocation. VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments. editor / Jennifer B. Sartor ; Mayur Naik ; Jennifer B. Sartor. Association for Computing Machinery, Inc, 2019. pp. 122-136 (VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments).

@inproceedings{3a509a0f9dce493fbc1f355462065413,

title = "Stochastic resource allocation",

abstract = "Suboptimal resource utilization among public and private cloud providers prevents them from maximizing their economic potential. Long-term allocated resources are often idle when they might have been subleased for a short period. Alternatively, arbitrary resource overcommitment may lead to unpredictable client performance. We propose a mechanism for fixed availability (traditional) resource allocation alongside stochastic resource allocation in the form of shares. We show its benefit for private and public cloud providers and for a wide range of clients. Our simulations show that our mechanism can increase server consolidation by 5.6 times on average compared with selling only fixed performance resources, and by 1.7 times compared with burstable instances, which is the most prevalent flexible allocation method. Our mechanism also yields better performance (i.e., higher revenues) or a lower cost than burstable instances for a wide range of clients, making it more profitable for them.",

keywords = "Cloud, Pricing, Resource-Allocation, Scheduler",

author = "Liran Funaro and Ben-Yehuda, {Orna Agmon} and Assaf Schuster",

note = "Publisher Copyright: {\textcopyright} 2019 Copyright held by the owner/author(s). Publication rights licensed to ACM.; 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, VEE 2019 ; Conference date: 14-04-2019",

year = "2019",

month = apr,

day = "14",

doi = "10.1145/3313808.3313815",

language = "English",

series = "VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments",

publisher = "Association for Computing Machinery, Inc",

pages = "122--136",

editor = "Sartor, {Jennifer B.} and Mayur Naik and Sartor, {Jennifer B.}",

booktitle = "VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments",

}

Funaro, L, Ben-Yehuda, OA & Schuster, A 2019, Stochastic resource allocation. in JB Sartor, M Naik & JB Sartor (eds), VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments. VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, Association for Computing Machinery, Inc, pp. 122-136, 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, VEE 2019, Providence, United States, 14/04/19. https://doi.org/10.1145/3313808.3313815

Stochastic resource allocation. / Funaro, Liran; Ben-Yehuda, Orna Agmon; Schuster, Assaf.
VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments. ed. / Jennifer B. Sartor; Mayur Naik; Jennifer B. Sartor. Association for Computing Machinery, Inc, 2019. p. 122-136 (VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Stochastic resource allocation

AU - Funaro, Liran

AU - Ben-Yehuda, Orna Agmon

AU - Schuster, Assaf

PY - 2019/4/14

Y1 - 2019/4/14

N2 - Suboptimal resource utilization among public and private cloud providers prevents them from maximizing their economic potential. Long-term allocated resources are often idle when they might have been subleased for a short period. Alternatively, arbitrary resource overcommitment may lead to unpredictable client performance. We propose a mechanism for fixed availability (traditional) resource allocation alongside stochastic resource allocation in the form of shares. We show its benefit for private and public cloud providers and for a wide range of clients. Our simulations show that our mechanism can increase server consolidation by 5.6 times on average compared with selling only fixed performance resources, and by 1.7 times compared with burstable instances, which is the most prevalent flexible allocation method. Our mechanism also yields better performance (i.e., higher revenues) or a lower cost than burstable instances for a wide range of clients, making it more profitable for them.

AB - Suboptimal resource utilization among public and private cloud providers prevents them from maximizing their economic potential. Long-term allocated resources are often idle when they might have been subleased for a short period. Alternatively, arbitrary resource overcommitment may lead to unpredictable client performance. We propose a mechanism for fixed availability (traditional) resource allocation alongside stochastic resource allocation in the form of shares. We show its benefit for private and public cloud providers and for a wide range of clients. Our simulations show that our mechanism can increase server consolidation by 5.6 times on average compared with selling only fixed performance resources, and by 1.7 times compared with burstable instances, which is the most prevalent flexible allocation method. Our mechanism also yields better performance (i.e., higher revenues) or a lower cost than burstable instances for a wide range of clients, making it more profitable for them.

KW - Cloud

KW - Pricing

KW - Resource-Allocation

KW - Scheduler

UR - http://www.scopus.com/inward/record.url?scp=85066632511&partnerID=8YFLogxK

U2 - 10.1145/3313808.3313815

DO - 10.1145/3313808.3313815

M3 - Conference contribution

AN - SCOPUS:85066632511

T3 - VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments

SP - 122

EP - 136

BT - VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments

A2 - Sartor, Jennifer B.

A2 - Naik, Mayur

A2 - Sartor, Jennifer B.

PB - Association for Computing Machinery, Inc

T2 - 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, VEE 2019

Y2 - 14 April 2019

ER -

Funaro L, Ben-Yehuda OA, Schuster A. Stochastic resource allocation. In Sartor JB, Naik M, Sartor JB, editors, VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments. Association for Computing Machinery, Inc. 2019. p. 122-136. (VEE 2019 - Proceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments). doi: 10.1145/3313808.3313815

Stochastic resource allocation

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this