Adaptive Partitioning and Scheduling for Enhancing WWW Application Performance

A partitioning and scheduling mechanism that adapts to resource changes and optimizes resource utilization is proposed and the application of this mechanism for online information browsing is demonstrated.

Abstract

This paper studies runtime partitioning, scheduling and load balancing techniques for improving performance of online WWW-based information systems such as digital libraries. The main performance bottlenecks of such a system are caused by the server computing capability and Internet bandwidth. Our observations and solutions are based on our experience with the Alexandria Digital Library (ADL) testbed at UCSB, which provides online browsing and processing of documents, digitized maps, and other geo-spatially mapped data via the WWW. A proper partitioning and scheduling of computation and communication in processing a user request on a multiprocessor server and transferring some computation to client-site machines can reduce network traffic and substantially improve system response time. We propose a partitioning and scheduling mechanism that adapts to resource changes and optimizes resource utilization and demonstrate the application of this mechanism for online information browsing. We also provide a performance analysis and experimental results to study the impact of resource availability and the effectiveness of our scheduling techniques.