Adaptive software cache management

Gil Einziger, Ohad Eytan, Roy Friedman, Ben Manes

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

27 Scopus citations


Developing a silver bullet software cache management policy is a daunting task due to the variety of potential workloads. In this paper, we investigate an adaptivity mechanism for software cache management schemes which offer tuning parameters targeted at the frequency vs. recency bias in the workload. The goal is automatic tuning of the parameters for best performance based on the workload without any manual intervention. We study two approaches for this problem, a hill climbing solution and an indicator based solution. In hill climbing, we repeatedly reconfigure the system hoping to find its best setting. In the indicator approach, we estimate the workloads’ frequency vs. recency bias and adjust the parameters accordingly in a single swoop. We apply these adaptive mechanisms to two recent software management schemes. We perform an extensive evaluation of the schemes and adaptation mechanisms over a large selection of workloads with varying characteristics. With these, we derive a parame-terless software cache management policy that is competitive for all tested workloads.

Original languageEnglish
Title of host publicationProceedings of the 19th International Middleware Conference, Middleware 2018
PublisherAssociation for Computing Machinery, Inc
Number of pages13
ISBN (Electronic)9781450357029
StatePublished - 26 Nov 2018
Event19th ACM/IFIP/USENIX International Middleware Conference, Middleware 2018 - Rennes, Brittany, France
Duration: 10 Dec 201814 Dec 2018

Publication series

NameProceedings of the 19th International Middleware Conference, Middleware 2018


Conference19th ACM/IFIP/USENIX International Middleware Conference, Middleware 2018
CityRennes, Brittany

ASJC Scopus subject areas

  • Software
  • Information Systems


Dive into the research topics of 'Adaptive software cache management'. Together they form a unique fingerprint.

Cite this