Modeling retrieval parallelism in web content delivery

Abstract

Retrieval latency is one of the most important issues in web content delivery. Many works have been done to understand and improve web retrieval latency. However, most of the previous studies are individual object based, which we think is insufficient and sometimes inaccurate. In this paper, we propose a Web Retrieval Parallelism Model (WRPM) for the study of web retrieval performance. By providing precise capture of the parallelism in web retrieval process, our model reveals some new factors in web retrieval. Our results show that these factors have significant impact on page retrieval latency. We believe that the details revealed in this paper are very important to studies in the web content delivery area.