Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management

Laurens Versluis; Alexandru Iosup

doi:10.1145/3491204.3527466

Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management

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

Abstract

Datacenters need to become more power efficient for political and climate reasons. In this work, we introduce an idea for the community to further explore. We embed the idea in TaskFlow: a makespan conservative, energy-aware task placement policy for workflow scheduling. Using static, rough numbers and simulation, we obtain energy savings between [4.24, 47.00]% and [0.1, 13.6]%, respectively. We also present some pitfalls that should be investigated further, notably starvation of large tasks when using TaskFlow.

Original language	English
Title of host publication	ICPE 2022
Subtitle of host publication	Companion of the 2022 ACM/SPEC International Conference on Performance Engineering
Publisher	Association for Computing Machinery, Inc
Pages	81-88
Number of pages	8
ISBN (Electronic)	9781450391597
DOIs	https://doi.org/10.1145/3491204.3527466
Publication status	Unpublished - 19 Jul 2022
Event	13th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2022 - Virtual, Online, China Duration: 9 Apr 2022 → 13 Apr 2022

Conference

Conference	13th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2022
Country/Territory	China
City	Virtual, Online
Period	9/04/22 → 13/04/22

Bibliographical note

Publisher Copyright:
© 2022 ACM.

Keywords

cloud
datacenter
delay management
distributed system
energy
energy reduction
workflow
workflow management systems
workflow scheduling

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1145/3491204.3527466

Persistent URL (handle)

Persistent link

Cite this

@inproceedings{07e2e5713b6c4aad9d22e97660629f70,

title = "Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management",

abstract = "Datacenters need to become more power efficient for political and climate reasons. In this work, we introduce an idea for the community to further explore. We embed the idea in TaskFlow: a makespan conservative, energy-aware task placement policy for workflow scheduling. Using static, rough numbers and simulation, we obtain energy savings between [4.24, 47.00]% and [0.1, 13.6]%, respectively. We also present some pitfalls that should be investigated further, notably starvation of large tasks when using TaskFlow. ",

keywords = "cloud, datacenter, delay management, distributed system, energy, energy reduction, workflow, workflow management systems, workflow scheduling",

author = "Laurens Versluis and Alexandru Iosup",

note = "Publisher Copyright: {\textcopyright} 2022 ACM.; 13th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2022 ; Conference date: 09-04-2022 Through 13-04-2022",

year = "2022",

month = jul,

day = "19",

doi = "10.1145/3491204.3527466",

language = "English",

pages = "81--88",

booktitle = "ICPE 2022",

publisher = "Association for Computing Machinery, Inc",

}

Versluis, L & Iosup, A 2022, Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management. in ICPE 2022: Companion of the 2022 ACM/SPEC International Conference on Performance Engineering. Association for Computing Machinery, Inc, pp. 81-88, 13th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2022, Virtual, Online, China, 9/04/22. https://doi.org/10.1145/3491204.3527466

Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management. / Versluis, Laurens; Iosup, Alexandru.
ICPE 2022: Companion of the 2022 ACM/SPEC International Conference on Performance Engineering. Association for Computing Machinery, Inc, 2022. p. 81-88.

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

TY - GEN

T1 - Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management

AU - Versluis, Laurens

AU - Iosup, Alexandru

PY - 2022/7/19

Y1 - 2022/7/19

N2 - Datacenters need to become more power efficient for political and climate reasons. In this work, we introduce an idea for the community to further explore. We embed the idea in TaskFlow: a makespan conservative, energy-aware task placement policy for workflow scheduling. Using static, rough numbers and simulation, we obtain energy savings between [4.24, 47.00]% and [0.1, 13.6]%, respectively. We also present some pitfalls that should be investigated further, notably starvation of large tasks when using TaskFlow.

AB - Datacenters need to become more power efficient for political and climate reasons. In this work, we introduce an idea for the community to further explore. We embed the idea in TaskFlow: a makespan conservative, energy-aware task placement policy for workflow scheduling. Using static, rough numbers and simulation, we obtain energy savings between [4.24, 47.00]% and [0.1, 13.6]%, respectively. We also present some pitfalls that should be investigated further, notably starvation of large tasks when using TaskFlow.

KW - cloud

KW - datacenter

KW - delay management

KW - distributed system

KW - energy

KW - energy reduction

KW - workflow

KW - workflow management systems

KW - workflow scheduling

UR - http://www.scopus.com/inward/record.url?scp=85136012694&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85136012694&partnerID=8YFLogxK

U2 - 10.1145/3491204.3527466

DO - 10.1145/3491204.3527466

M3 - Conference contribution

AN - SCOPUS:85136012694

SP - 81

EP - 88

BT - ICPE 2022

PB - Association for Computing Machinery, Inc

T2 - 13th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2022

Y2 - 9 April 2022 through 13 April 2022

ER -

Taskflow: An energy- and makespan-aware task placement policy forworkflow scheduling through delay management

Abstract

Conference

Bibliographical note

Keywords

UN SDGs

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this