A CPU contention predictor for business-critical workloads in cloud datacenters

Vincent Van Beek, Giorgos Oikonomou, Alexandru Iosup

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

Resource contention is one of the major problems in cloud datacenters. Many types of resource contention occur, with important impact on the performance and sometimes even the reliability of applications running in cloud datacenters. Cloud applications run together on the same physical machines with different workloads resulting in non-synchronized accesses to the shared resources. This leads to cases where co-hosted applications are contending for the common resources and not receiving the demanded resource amounts. In this work, we investigate the contention in CPU resources, as CPU is allowed to be over-committed by typical SLAs. We propose a CPU-contention predictor for the demanding business-critical workloads, which require low resource contention to deliver the required performance to customers. Our predictor is based on a set of regression models and metrics which we evaluate extensively. We tune the predictor with data collected from a real-world cloud operation spanning multiple datacenters and servicing business-critical workloads.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages56-61
Number of pages6
ISBN (Electronic)9781728124063
DOIs
Publication statusPublished - 8 Aug 2019
Event4th IEEE International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019 - Umea, Sweden
Duration: 16 Jun 201920 Jun 2019

Conference

Conference4th IEEE International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019
CountrySweden
CityUmea
Period16/06/1920/06/19

Fingerprint

Contention
Program processors
Workload
Predictors
Resources
Industry
Business
Regression Model
Customers
Metric
Evaluate

Keywords

  • Business critical workloads
  • CPU contention
  • Resource contention

Cite this

Van Beek, V., Oikonomou, G., & Iosup, A. (2019). A CPU contention predictor for business-critical workloads in cloud datacenters. In Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019 (pp. 56-61). [8791987] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/FAS-W.2019.00027
Van Beek, Vincent ; Oikonomou, Giorgos ; Iosup, Alexandru. / A CPU contention predictor for business-critical workloads in cloud datacenters. Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 56-61
@inproceedings{83729555a66d4152ae9f516db13f5daa,
title = "A CPU contention predictor for business-critical workloads in cloud datacenters",
abstract = "Resource contention is one of the major problems in cloud datacenters. Many types of resource contention occur, with important impact on the performance and sometimes even the reliability of applications running in cloud datacenters. Cloud applications run together on the same physical machines with different workloads resulting in non-synchronized accesses to the shared resources. This leads to cases where co-hosted applications are contending for the common resources and not receiving the demanded resource amounts. In this work, we investigate the contention in CPU resources, as CPU is allowed to be over-committed by typical SLAs. We propose a CPU-contention predictor for the demanding business-critical workloads, which require low resource contention to deliver the required performance to customers. Our predictor is based on a set of regression models and metrics which we evaluate extensively. We tune the predictor with data collected from a real-world cloud operation spanning multiple datacenters and servicing business-critical workloads.",
keywords = "Business critical workloads, CPU contention, Resource contention",
author = "{Van Beek}, Vincent and Giorgos Oikonomou and Alexandru Iosup",
year = "2019",
month = "8",
day = "8",
doi = "10.1109/FAS-W.2019.00027",
language = "English",
pages = "56--61",
booktitle = "Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Van Beek, V, Oikonomou, G & Iosup, A 2019, A CPU contention predictor for business-critical workloads in cloud datacenters. in Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019., 8791987, Institute of Electrical and Electronics Engineers Inc., pp. 56-61, 4th IEEE International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019, Umea, Sweden, 16/06/19. https://doi.org/10.1109/FAS-W.2019.00027

A CPU contention predictor for business-critical workloads in cloud datacenters. / Van Beek, Vincent; Oikonomou, Giorgos; Iosup, Alexandru.

Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 56-61 8791987.

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - A CPU contention predictor for business-critical workloads in cloud datacenters

AU - Van Beek, Vincent

AU - Oikonomou, Giorgos

AU - Iosup, Alexandru

PY - 2019/8/8

Y1 - 2019/8/8

N2 - Resource contention is one of the major problems in cloud datacenters. Many types of resource contention occur, with important impact on the performance and sometimes even the reliability of applications running in cloud datacenters. Cloud applications run together on the same physical machines with different workloads resulting in non-synchronized accesses to the shared resources. This leads to cases where co-hosted applications are contending for the common resources and not receiving the demanded resource amounts. In this work, we investigate the contention in CPU resources, as CPU is allowed to be over-committed by typical SLAs. We propose a CPU-contention predictor for the demanding business-critical workloads, which require low resource contention to deliver the required performance to customers. Our predictor is based on a set of regression models and metrics which we evaluate extensively. We tune the predictor with data collected from a real-world cloud operation spanning multiple datacenters and servicing business-critical workloads.

AB - Resource contention is one of the major problems in cloud datacenters. Many types of resource contention occur, with important impact on the performance and sometimes even the reliability of applications running in cloud datacenters. Cloud applications run together on the same physical machines with different workloads resulting in non-synchronized accesses to the shared resources. This leads to cases where co-hosted applications are contending for the common resources and not receiving the demanded resource amounts. In this work, we investigate the contention in CPU resources, as CPU is allowed to be over-committed by typical SLAs. We propose a CPU-contention predictor for the demanding business-critical workloads, which require low resource contention to deliver the required performance to customers. Our predictor is based on a set of regression models and metrics which we evaluate extensively. We tune the predictor with data collected from a real-world cloud operation spanning multiple datacenters and servicing business-critical workloads.

KW - Business critical workloads

KW - CPU contention

KW - Resource contention

UR - http://www.scopus.com/inward/record.url?scp=85071467904&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071467904&partnerID=8YFLogxK

U2 - 10.1109/FAS-W.2019.00027

DO - 10.1109/FAS-W.2019.00027

M3 - Conference contribution

SP - 56

EP - 61

BT - Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Van Beek V, Oikonomou G, Iosup A. A CPU contention predictor for business-critical workloads in cloud datacenters. In Proceedings - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems, FAS*W 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 56-61. 8791987 https://doi.org/10.1109/FAS-W.2019.00027