Title

A Hybrid Cache Architecture for Multi-tenant Caching

Author

Taejoon Kim

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science and Info Sys

Date of Award

Spring 2019

Abstract

In-memory caching is an essential piece in a cloud system where multiple tenants share the computation and storage resources. In such a shared environment, individual tenants may have different performance goals including the performance of data access. The existing caching techniques would be either global caching which is tenant-unaware manner or static caching which is hard to determine the cache size for individual tenants. The core of this research is the design of a new type of cache architecture that incorporates global and static caching to meet the data access performance requirement. In addition, new eviction techniques will be presented which are developed to effectively manage the cache system based on the proposed cache architecture. Finally, the extensive evaluation results will be reported and conducted on the simulation and emulation settings, which confirmed the validity of the proposed cache architecture with the eviction algorithm.

Advisor

Jinoh Kim

Subject Categories

Computer Sciences | Physical Sciences and Mathematics

Share

COinS