On cache replacement policies for servicing mixed data intensive query workloads

TitleOn cache replacement policies for servicing mixed data intensive query workloads
Publication TypeConference Papers
Year of Publication2002
AuthorsAndrade H, Kurc T, Sussman A, Borovikov E, Saltz J
Conference NameProceedings of the 2nd Workshop on Caching, Coherence, and Consistency, held in conjunction with the 16th ACM International Conference on Supercomputing
Date Published2002///
Abstract

When data analysis applications are employed in a multi-client environment, a data server must service multiple si-
multaneous queries, each of which may employ complex
user-defined data structures and operations on the data. It
is then necessary to harness inter- and intra-query com-
monalities and system resources to improve the performance
of the data server. We have developed a framework and
customizable middleware to enable reuse of intermediate
and final results among queries, through an in-memory ac-
tive semantic cache and user-defined transformation func-
tions. Since resources such as processing power and mem-
ory space are limited on the machine hosting the server, ef-
fective scheduling of incoming queries and efficient cache
replacement policies are challenging issues that must be ad-
dressed. We have worked on the scheduling problem in ear-
lier work, and in this paper we describe and evaluate several
cache replacement policies. We present experimental eval-
uation of the policies on a shared-memory parallel system
using two applications from different application domains.