A cloud-agnostic queuing system to support the implementation of deadline-based application execution policies

Tamas Kiss, James DesLauriers, Gregoire Gesmier, Gabor Terstyanszky, Gabriele Pierantoni, Osama Abu Oun, Simon J.E. Taylor, Anastasia Anagnostou, J. Kovács

Research output: Contribution to journalArticle

Abstract

There are many scientific and commercial applications that require the execution of a large number of independent jobs resulting in significant overall execution time. Therefore, such applications typically require distributed computing infrastructures and science gateways to run efficiently and to be easily accessible for end-users. Optimising the execution of such applications in a cloud computing environment by keeping resource utilisation at minimum but still completing the experiment by a set deadline has paramount importance. As container-based technologies are becoming more widespread, support for job-queuing and auto-scaling in such environments is becoming important. Current container management technologies, such as Docker Swarm or Kubernetes, while provide auto-scaling based on resource consumption, do not support job queuing and deadline-based execution policies directly. This paper presents JQueuer, a cloud-agnostic queuing system that supports the scheduling of a large number of jobs in containerised cloud environments. The paper also demonstrates how JQueuer, when integrated with a cloud application-level orchestrator and auto-scaling framework, called MiCADO, can be used to implement deadline-based execution policies. This novel technical solution provides an important step towards the cost-optimisation of batch processing and job submission applications. In order to test and prove the effectiveness of the solution, the paper presents experimental results when executing an agent-based simulation application using the open source REPAST simulation framework.

Original languageEnglish
Pages (from-to)99-111
Number of pages13
JournalFuture Generation Computer Systems
Volume101
DOIs
Publication statusPublished - Dec 1 2019

Fingerprint

Containers
Gateways (computer networks)
Distributed computer systems
Cloud computing
Scheduling
Costs
Experiments

Keywords

  • Agent-based simulation
  • Cloud computing
  • Container technologies
  • Deadline-based auto-scaling
  • JQueuer
  • MiCADO

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

A cloud-agnostic queuing system to support the implementation of deadline-based application execution policies. / Kiss, Tamas; DesLauriers, James; Gesmier, Gregoire; Terstyanszky, Gabor; Pierantoni, Gabriele; Oun, Osama Abu; Taylor, Simon J.E.; Anagnostou, Anastasia; Kovács, J.

In: Future Generation Computer Systems, Vol. 101, 01.12.2019, p. 99-111.

Research output: Contribution to journalArticle

Kiss, T, DesLauriers, J, Gesmier, G, Terstyanszky, G, Pierantoni, G, Oun, OA, Taylor, SJE, Anagnostou, A & Kovács, J 2019, 'A cloud-agnostic queuing system to support the implementation of deadline-based application execution policies', Future Generation Computer Systems, vol. 101, pp. 99-111. https://doi.org/10.1016/j.future.2019.05.062
Kiss, Tamas ; DesLauriers, James ; Gesmier, Gregoire ; Terstyanszky, Gabor ; Pierantoni, Gabriele ; Oun, Osama Abu ; Taylor, Simon J.E. ; Anagnostou, Anastasia ; Kovács, J. / A cloud-agnostic queuing system to support the implementation of deadline-based application execution policies. In: Future Generation Computer Systems. 2019 ; Vol. 101. pp. 99-111.
@article{1f6304c0d4aa4086be6e55a75e3814f3,
title = "A cloud-agnostic queuing system to support the implementation of deadline-based application execution policies",
abstract = "There are many scientific and commercial applications that require the execution of a large number of independent jobs resulting in significant overall execution time. Therefore, such applications typically require distributed computing infrastructures and science gateways to run efficiently and to be easily accessible for end-users. Optimising the execution of such applications in a cloud computing environment by keeping resource utilisation at minimum but still completing the experiment by a set deadline has paramount importance. As container-based technologies are becoming more widespread, support for job-queuing and auto-scaling in such environments is becoming important. Current container management technologies, such as Docker Swarm or Kubernetes, while provide auto-scaling based on resource consumption, do not support job queuing and deadline-based execution policies directly. This paper presents JQueuer, a cloud-agnostic queuing system that supports the scheduling of a large number of jobs in containerised cloud environments. The paper also demonstrates how JQueuer, when integrated with a cloud application-level orchestrator and auto-scaling framework, called MiCADO, can be used to implement deadline-based execution policies. This novel technical solution provides an important step towards the cost-optimisation of batch processing and job submission applications. In order to test and prove the effectiveness of the solution, the paper presents experimental results when executing an agent-based simulation application using the open source REPAST simulation framework.",
keywords = "Agent-based simulation, Cloud computing, Container technologies, Deadline-based auto-scaling, JQueuer, MiCADO",
author = "Tamas Kiss and James DesLauriers and Gregoire Gesmier and Gabor Terstyanszky and Gabriele Pierantoni and Oun, {Osama Abu} and Taylor, {Simon J.E.} and Anastasia Anagnostou and J. Kov{\'a}cs",
year = "2019",
month = "12",
day = "1",
doi = "10.1016/j.future.2019.05.062",
language = "English",
volume = "101",
pages = "99--111",
journal = "Future Generation Computer Systems",
issn = "0167-739X",
publisher = "Elsevier",

}

TY - JOUR

T1 - A cloud-agnostic queuing system to support the implementation of deadline-based application execution policies

AU - Kiss, Tamas

AU - DesLauriers, James

AU - Gesmier, Gregoire

AU - Terstyanszky, Gabor

AU - Pierantoni, Gabriele

AU - Oun, Osama Abu

AU - Taylor, Simon J.E.

AU - Anagnostou, Anastasia

AU - Kovács, J.

PY - 2019/12/1

Y1 - 2019/12/1

N2 - There are many scientific and commercial applications that require the execution of a large number of independent jobs resulting in significant overall execution time. Therefore, such applications typically require distributed computing infrastructures and science gateways to run efficiently and to be easily accessible for end-users. Optimising the execution of such applications in a cloud computing environment by keeping resource utilisation at minimum but still completing the experiment by a set deadline has paramount importance. As container-based technologies are becoming more widespread, support for job-queuing and auto-scaling in such environments is becoming important. Current container management technologies, such as Docker Swarm or Kubernetes, while provide auto-scaling based on resource consumption, do not support job queuing and deadline-based execution policies directly. This paper presents JQueuer, a cloud-agnostic queuing system that supports the scheduling of a large number of jobs in containerised cloud environments. The paper also demonstrates how JQueuer, when integrated with a cloud application-level orchestrator and auto-scaling framework, called MiCADO, can be used to implement deadline-based execution policies. This novel technical solution provides an important step towards the cost-optimisation of batch processing and job submission applications. In order to test and prove the effectiveness of the solution, the paper presents experimental results when executing an agent-based simulation application using the open source REPAST simulation framework.

AB - There are many scientific and commercial applications that require the execution of a large number of independent jobs resulting in significant overall execution time. Therefore, such applications typically require distributed computing infrastructures and science gateways to run efficiently and to be easily accessible for end-users. Optimising the execution of such applications in a cloud computing environment by keeping resource utilisation at minimum but still completing the experiment by a set deadline has paramount importance. As container-based technologies are becoming more widespread, support for job-queuing and auto-scaling in such environments is becoming important. Current container management technologies, such as Docker Swarm or Kubernetes, while provide auto-scaling based on resource consumption, do not support job queuing and deadline-based execution policies directly. This paper presents JQueuer, a cloud-agnostic queuing system that supports the scheduling of a large number of jobs in containerised cloud environments. The paper also demonstrates how JQueuer, when integrated with a cloud application-level orchestrator and auto-scaling framework, called MiCADO, can be used to implement deadline-based execution policies. This novel technical solution provides an important step towards the cost-optimisation of batch processing and job submission applications. In order to test and prove the effectiveness of the solution, the paper presents experimental results when executing an agent-based simulation application using the open source REPAST simulation framework.

KW - Agent-based simulation

KW - Cloud computing

KW - Container technologies

KW - Deadline-based auto-scaling

KW - JQueuer

KW - MiCADO

UR - http://www.scopus.com/inward/record.url?scp=85067198184&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067198184&partnerID=8YFLogxK

U2 - 10.1016/j.future.2019.05.062

DO - 10.1016/j.future.2019.05.062

M3 - Article

VL - 101

SP - 99

EP - 111

JO - Future Generation Computer Systems

JF - Future Generation Computer Systems

SN - 0167-739X

ER -