STAAF: Spatial-Temporal Correlations Aware AutoScaling Framework for Microservices

  • Jianxiong Liao*
  • , Zhi Zhou
  • , Fei Xu
  • , Xu Chen
  • *Corresponding author for this work

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

1 Scopus citations

Abstract

Microservice architecture has gained widespread popularity for simplifying the development of online applications. However, the complicated dependencies among microservices and fluctuating nature of online application workloads can pose significant challenges in efficiently provisioning resources to microservices to improve the performance of applications and optimize resource utilization. In this paper, we present STAAF, a proactive microservice autoscaling framework that takes the spatial-temporal correlations of microservices into account to globally provision resources. STAAF predicts the workloads of the entering microservice by analyzing the temporal pattern of workloads and estimates the actual workloads of all microservices to solve the problems brought by spatial correlations of microservices. In addition, STAAF profiles the mapping between resource utilization and request arrival rate to proactively allocate resources while optimizing resource usage. Experiments conducted with a microservice benchmark indicate that STAAF can make appropriate autoscaling decisions during traffic surges and reduce tail latency by 74% on average compared to the Kubernetes autoscaler.

Original languageEnglish
Title of host publicationProceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350319804
DOIs
StatePublished - 2023
Event9th IEEE Smart World Congress, SWC 2023 - Portsmouth, United Kingdom
Duration: 28 Aug 202331 Aug 2023

Publication series

NameProceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023

Conference

Conference9th IEEE Smart World Congress, SWC 2023
Country/TerritoryUnited Kingdom
CityPortsmouth
Period28/08/2331/08/23

Keywords

  • autoscaling
  • cloud computing
  • microservices
  • spatial and temporal correlations

Fingerprint

Dive into the research topics of 'STAAF: Spatial-Temporal Correlations Aware AutoScaling Framework for Microservices'. Together they form a unique fingerprint.

Cite this