A trace-based performance study of autoscaling workloads of workflows in datacenters

Laurens Versluis, Mihai Neacsu, Alexandru Iosup

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

99 Downloads (Pure)

Abstract

To improve customer experience, datacenter operators offer support for simplifying application and resource management. For example, running workloads of workflows on behalf of customers is desirable, but requires increasingly more sophisticated autoscaling policies, that is, policies that dynamically provision resources for the customer. Although selecting and tuning autoscaling policies is a challenging task for datacenter operators, so far relatively few studies investigate the performance of autoscaling for workloads of workflows. Complementing previous knowledge, in this work we propose the first comprehensive performance study in the field. Using trace-based simulation, we compare state-of-the-art autoscaling policies across multiple application domains, workload arrival patterns (e.g., burstiness), and system utilization levels. We further investigate the interplay between autoscaling and regular allocation policies, and the complexity cost of autoscaling. Our quantitative study focuses not only on traditional performance metrics and on state-of-the-art elasticity metrics, but also on time-and memory-related autoscaling-complexity metrics. Our main results give strong and quantitative evidence about previously unreported operational behavior, for example, that autoscaling policies perform differently across application domains and allocation and provisioning policies should be co-designed.

Original languageEnglish
Title of host publicationProceedings - 18th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, CCGRID 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages223-232
Number of pages10
ISBN (Electronic)9781538658154
DOIs
Publication statusPublished - 13 Jul 2018
Event18th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, CCGRID 2018 - Washington, United States
Duration: 1 May 20184 May 2018

Conference

Conference18th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, CCGRID 2018
CountryUnited States
CityWashington
Period1/05/184/05/18

Keywords

  • Autoscaler
  • Autoscaling
  • Datacenters
  • Real world Traces
  • Resources
  • Scheduling
  • Simulation

Fingerprint Dive into the research topics of 'A trace-based performance study of autoscaling workloads of workflows in datacenters'. Together they form a unique fingerprint.

Cite this