Using speculation to reduce server load and service time on the WWW
CIKM '95 Proceedings of the fourth international conference on Information and knowledge management
EW 7 Proceedings of the 7th workshop on ACM SIGOPS European workshop: Systems support for worldwide applications
Web object-based storage management in proxy caches
Future Generation Computer Systems
USITS'99 Proceedings of the 2nd conference on USENIX Symposium on Internet Technologies and Systems - Volume 2
Improving the delivery of multimedia embedded in HTML over HTTP on wireless networks
Proceedings of the 3rd international conference on Mobile multimedia communications
Hi-index | 0.00 |
The purpose of this research is to improve the performance of the Apache HTTP Server for serving web pages which include embedded objects, such as images or multimedia components. By prefetching the embedded objects of a web page from server disk to its cache memory, we reduce page load latency without modifying current web protocols. We have implemented this strategy within the Apache HTTP Server that is referred to in this paper as modified Apache. Apache HTTP Server is a popular open source web server. In this paper, we present the strategy that we used to prefetch embedded objects in modified Apache from disk into main memory, and report on the performance achieved in trace driven simulations. The performance was measured by processing a log file from IRCache to simulate realistic fetches on an object-by-object basis. The memory cache hit ratio and byte hit ratio were recorded for each object accessed through normal and modified Apache, and compared.