Exploiting Object Relationships for More Deterministic Management of Distributed Objects

Worcester Polytechnic Institute, Worcester MA

Investigators

Abstract

Caching and replication of objects on the Internet and the World Wide Web are frequently used techniques to reduce latencies for clients accessing these objects while reducing network and server load. This work investigates the potential of managing objects where servers supply more complete information to client caches and use the information to manage replicated content. The idea is to exploit composite objects, which group a set of distinct objects where each object has a uniform type and change characteristic. Relationships between these objects can be "compiled" at a server into information that the server can pass to clients. Such information ranges from explicit expiration times for caching/replicating a list of objects to invalidations based on the object membership in a volume of related objects. A client/server environment consisting of objects with different types and relationships will be used to test the approach. It is expected to yield more effective and predictable caching as well as drive placement and removal of replicated content, particularly when combined with client access counts. The work will have continued impact as distributed object environments continue to grow in size with increasing diversity in the type of objects grouped together and served.