Cache performance depends heavily on replacement algorithms, which dynamically select a suitable subset of objects for caching in a finite space. Developing such algorithms for wide-area distributed environments is challenging because, unlike traditional paging systems, retrieval costs and object sizes are not necessarily uniform. This paper presents the least-unified value algorithm, which performs better than existing algorithms for replacing nonuniform data objects in wide-area distributed environments.
Bibliographical noteFunding Information:
We thank NLANR and DEC for making their proxy traces available; without their generosity, this work would not exist. Many thanks also to the reviewers for their constructive and enlightening comments. This article was supported in part by Korea’s Ministry of Education under the Brain Korea 21 program and by the Ministry of Science and Technology under the National Research Laboratory program.