Algorithms for optimising heterogeneous Cloud virtual machine clusters

Long Thanh Thai; Blesson Varghese; Adam David Barker

doi:10.1109/CloudCom.2016.0033

Algorithms for optimising heterogeneous Cloud virtual machine clusters

Long Thanh Thai, Blesson Varghese, Adam David Barker

School of Computer Science

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

Abstract

It is challenging to execute an application in a heterogeneous cloud cluster, which consists of multiple types of virtual machines with different performance capabilities and prices. This paper aims to mitigate this challenge by proposing a scheduling mechanism to optimise the execution of Bag-of-Task jobs on a heterogeneous cloud cluster. The proposed scheduler considers two approaches to select suitable cloud resources for executing a user application while satisfying pre-defined Service Level Objectives (SLOs) both in terms of execution deadline and minimising monetary cost. Additionally, a mechanism for dynamic re-assignment of jobs during execution is presented to resolve potential violation of SLOs.
Experimental studies are performed both in simulation and on a public cloud using real-world applications. The results highlight that our scheduling approaches result in cost saving of up to 31% in comparison to naive approaches that only employ a single type of virtual machine in a homogeneous cluster. Dynamic reassignment completely prevents deadline violation in the best-case and reduces deadline violations by 95% in the worst-case scenario.

Original language	English
Title of host publication	2016 IEEE International Conference on Cloud Computing Technology and Science
Publisher	IEEE
Pages	118-125
Number of pages	8
ISBN (Electronic)	9781509014453
ISBN (Print)	9781509014460
DOIs	https://doi.org/10.1109/CloudCom.2016.0033
Publication status	Published - 12 Dec 2016
Event	8th IEEE International Conference on Cloud Computing Technology and Science - Alvisse Parc Hotel, Luxembourg Duration: 12 Dec 2016 → 15 Dec 2016 Conference number: 8 http://2016.cloudcom.org/

Conference

Conference	8th IEEE International Conference on Cloud Computing Technology and Science
Abbreviated title	CloudCom
Country/Territory	Luxembourg
Period	12/12/16 → 15/12/16
Internet address	http://2016.cloudcom.org/

Access to Document

10.1109/CloudCom.2016.0033

Thai_2016_Algorithms_CloudCom_AAM
© 2016, IEEE. This work has been made available online in accordance with the publisher’s policies. This is the author created, accepted version manuscript following peer review and may differ slightly from the final published version. The final published version of this work is available at ieeexplore.ieee.org / 10.1109/CloudCom.2016.0033
Accepted author manuscript, 1.32 MBLicence: Unspecified

Cite this

@inproceedings{97d43d17da1d445ca242446f3ad32d6b,

title = "Algorithms for optimising heterogeneous Cloud virtual machine clusters",

abstract = "It is challenging to execute an application in a heterogeneous cloud cluster, which consists of multiple types of virtual machines with different performance capabilities and prices. This paper aims to mitigate this challenge by proposing a scheduling mechanism to optimise the execution of Bag-of-Task jobs on a heterogeneous cloud cluster. The proposed scheduler considers two approaches to select suitable cloud resources for executing a user application while satisfying pre-defined Service Level Objectives (SLOs) both in terms of execution deadline and minimising monetary cost. Additionally, a mechanism for dynamic re-assignment of jobs during execution is presented to resolve potential violation of SLOs.Experimental studies are performed both in simulation and on a public cloud using real-world applications. The results highlight that our scheduling approaches result in cost saving of up to 31% in comparison to naive approaches that only employ a single type of virtual machine in a homogeneous cluster. Dynamic reassignment completely prevents deadline violation in the best-case and reduces deadline violations by 95% in the worst-case scenario.",

author = "Thai, {Long Thanh} and Blesson Varghese and Barker, {Adam David}",

note = "This research was supported by an Amazon Web Services Education Research grant. ; 8th IEEE International Conference on Cloud Computing Technology and Science, CloudCom ; Conference date: 12-12-2016 Through 15-12-2016",

year = "2016",

month = dec,

day = "12",

doi = "10.1109/CloudCom.2016.0033",

language = "English",

isbn = "9781509014460",

pages = "118--125",

booktitle = "2016 IEEE International Conference on Cloud Computing Technology and Science",

publisher = "IEEE",

url = "http://2016.cloudcom.org/",

}

TY - GEN

T1 - Algorithms for optimising heterogeneous Cloud virtual machine clusters

AU - Thai, Long Thanh

AU - Varghese, Blesson

AU - Barker, Adam David

N1 - Conference code: 8

PY - 2016/12/12

Y1 - 2016/12/12

N2 - It is challenging to execute an application in a heterogeneous cloud cluster, which consists of multiple types of virtual machines with different performance capabilities and prices. This paper aims to mitigate this challenge by proposing a scheduling mechanism to optimise the execution of Bag-of-Task jobs on a heterogeneous cloud cluster. The proposed scheduler considers two approaches to select suitable cloud resources for executing a user application while satisfying pre-defined Service Level Objectives (SLOs) both in terms of execution deadline and minimising monetary cost. Additionally, a mechanism for dynamic re-assignment of jobs during execution is presented to resolve potential violation of SLOs.Experimental studies are performed both in simulation and on a public cloud using real-world applications. The results highlight that our scheduling approaches result in cost saving of up to 31% in comparison to naive approaches that only employ a single type of virtual machine in a homogeneous cluster. Dynamic reassignment completely prevents deadline violation in the best-case and reduces deadline violations by 95% in the worst-case scenario.

AB - It is challenging to execute an application in a heterogeneous cloud cluster, which consists of multiple types of virtual machines with different performance capabilities and prices. This paper aims to mitigate this challenge by proposing a scheduling mechanism to optimise the execution of Bag-of-Task jobs on a heterogeneous cloud cluster. The proposed scheduler considers two approaches to select suitable cloud resources for executing a user application while satisfying pre-defined Service Level Objectives (SLOs) both in terms of execution deadline and minimising monetary cost. Additionally, a mechanism for dynamic re-assignment of jobs during execution is presented to resolve potential violation of SLOs.Experimental studies are performed both in simulation and on a public cloud using real-world applications. The results highlight that our scheduling approaches result in cost saving of up to 31% in comparison to naive approaches that only employ a single type of virtual machine in a homogeneous cluster. Dynamic reassignment completely prevents deadline violation in the best-case and reduces deadline violations by 95% in the worst-case scenario.

U2 - 10.1109/CloudCom.2016.0033

DO - 10.1109/CloudCom.2016.0033

M3 - Conference contribution

SN - 9781509014460

SP - 118

EP - 125

BT - 2016 IEEE International Conference on Cloud Computing Technology and Science

PB - IEEE

T2 - 8th IEEE International Conference on Cloud Computing Technology and Science

Y2 - 12 December 2016 through 15 December 2016

ER -

Algorithms for optimising heterogeneous Cloud virtual machine clusters

Abstract

Conference

Access to Document

Fingerprint

Cite this