Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors

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

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Cloud computing platforms are increasingly deploying multi-processors that are heterogeneous in the resource capacities or functionality of their processors (Instruction Set Architecture, or ISA). ISA heterogeneity (e.g., CPU vs GPU) or administrative policies can additionally create placement constraints whereby certain threads may only execute on a subset of the available cores. Fair CPU scheduling in such settings poses novel challenges that we address in this paper. First, we describe the conditions for a feasible allocation. We then develop a general utility optimal scheduling framework that, when appropriately parameterized, adjusts the trade-off between fairness and throughput, and captures a variety of notions of fairness (proportional fair, max-min fair, etc.). Finally, we design a low-complexity quantum-level scheduling algorithm, called CMFS. We evaluate the efficacy of CMFS via simulations and identify promising future directions.

Original languageEnglish (US)
Title of host publication2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages48-53
Number of pages6
ISBN (Electronic)9781538627846
DOIs
StatePublished - Nov 20 2017
Event2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017 - Atlanta, United States
Duration: May 1 2017May 4 2017

Publication series

Name2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017

Other

Other2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017
CountryUnited States
CityAtlanta
Period5/1/175/4/17

Fingerprint

Multiprocessor
Thread
Placement
Program processors
Scheduling
Cloud computing
Scheduling algorithms
Proportional Fairness
Optimal Scheduling
Throughput
Min-max
Cloud Computing
Fairness
Scheduling Algorithm
Low Complexity
Efficacy
Trade-offs
Resources
Subset
Evaluate

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Control and Optimization
  • Artificial Intelligence
  • Computer Networks and Communications

Cite this

Khamse-Ashari, J., Kesidis, G., Lambadaris, I., Urgaonkar, B., & Zhao, Y. (2017). Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors. In 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017 (pp. 48-53). [8116351] (2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INFCOMW.2017.8116351
Khamse-Ashari, Jalal ; Kesidis, George ; Lambadaris, Ioannis ; Urgaonkar, Bhuvan ; Zhao, Yiqiang. / Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors. 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 48-53 (2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017).
@inproceedings{f4b1823b38d14db98a45559e74204deb,
title = "Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors",
abstract = "Cloud computing platforms are increasingly deploying multi-processors that are heterogeneous in the resource capacities or functionality of their processors (Instruction Set Architecture, or ISA). ISA heterogeneity (e.g., CPU vs GPU) or administrative policies can additionally create placement constraints whereby certain threads may only execute on a subset of the available cores. Fair CPU scheduling in such settings poses novel challenges that we address in this paper. First, we describe the conditions for a feasible allocation. We then develop a general utility optimal scheduling framework that, when appropriately parameterized, adjusts the trade-off between fairness and throughput, and captures a variety of notions of fairness (proportional fair, max-min fair, etc.). Finally, we design a low-complexity quantum-level scheduling algorithm, called CMFS. We evaluate the efficacy of CMFS via simulations and identify promising future directions.",
author = "Jalal Khamse-Ashari and George Kesidis and Ioannis Lambadaris and Bhuvan Urgaonkar and Yiqiang Zhao",
year = "2017",
month = "11",
day = "20",
doi = "10.1109/INFCOMW.2017.8116351",
language = "English (US)",
series = "2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "48--53",
booktitle = "2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017",
address = "United States",

}

Khamse-Ashari, J, Kesidis, G, Lambadaris, I, Urgaonkar, B & Zhao, Y 2017, Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors. in 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017., 8116351, 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017, Institute of Electrical and Electronics Engineers Inc., pp. 48-53, 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017, Atlanta, United States, 5/1/17. https://doi.org/10.1109/INFCOMW.2017.8116351

Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors. / Khamse-Ashari, Jalal; Kesidis, George; Lambadaris, Ioannis; Urgaonkar, Bhuvan; Zhao, Yiqiang.

2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 48-53 8116351 (2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors

AU - Khamse-Ashari, Jalal

AU - Kesidis, George

AU - Lambadaris, Ioannis

AU - Urgaonkar, Bhuvan

AU - Zhao, Yiqiang

PY - 2017/11/20

Y1 - 2017/11/20

N2 - Cloud computing platforms are increasingly deploying multi-processors that are heterogeneous in the resource capacities or functionality of their processors (Instruction Set Architecture, or ISA). ISA heterogeneity (e.g., CPU vs GPU) or administrative policies can additionally create placement constraints whereby certain threads may only execute on a subset of the available cores. Fair CPU scheduling in such settings poses novel challenges that we address in this paper. First, we describe the conditions for a feasible allocation. We then develop a general utility optimal scheduling framework that, when appropriately parameterized, adjusts the trade-off between fairness and throughput, and captures a variety of notions of fairness (proportional fair, max-min fair, etc.). Finally, we design a low-complexity quantum-level scheduling algorithm, called CMFS. We evaluate the efficacy of CMFS via simulations and identify promising future directions.

AB - Cloud computing platforms are increasingly deploying multi-processors that are heterogeneous in the resource capacities or functionality of their processors (Instruction Set Architecture, or ISA). ISA heterogeneity (e.g., CPU vs GPU) or administrative policies can additionally create placement constraints whereby certain threads may only execute on a subset of the available cores. Fair CPU scheduling in such settings poses novel challenges that we address in this paper. First, we describe the conditions for a feasible allocation. We then develop a general utility optimal scheduling framework that, when appropriately parameterized, adjusts the trade-off between fairness and throughput, and captures a variety of notions of fairness (proportional fair, max-min fair, etc.). Finally, we design a low-complexity quantum-level scheduling algorithm, called CMFS. We evaluate the efficacy of CMFS via simulations and identify promising future directions.

UR - http://www.scopus.com/inward/record.url?scp=85041340766&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041340766&partnerID=8YFLogxK

U2 - 10.1109/INFCOMW.2017.8116351

DO - 10.1109/INFCOMW.2017.8116351

M3 - Conference contribution

AN - SCOPUS:85041340766

T3 - 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017

SP - 48

EP - 53

BT - 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Khamse-Ashari J, Kesidis G, Lambadaris I, Urgaonkar B, Zhao Y. Efficient and fair scheduling of placement constrained threads on heterogeneous multi-processors. In 2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 48-53. 8116351. (2017 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2017). https://doi.org/10.1109/INFCOMW.2017.8116351