Accelerating Web Page Retrieval Through Object Usage Declaration
ANSS '04 Proceedings of the 37th annual symposium on Simulation
Proxy-cache aware object bundling for web access acceleration
Web content caching and distribution
Web caching performance: how much is lost unwarily?
HSI'03 Proceedings of the 2nd international conference on Human.society@internet
| Hi-index | 0.00 |
In this paper, we propose a chunk-level client's latencydependence model (C-LDM) to describe the effect ofnetwork protocol, streaming data transfer mechanism,and web page structure on the latency perceived by a webclient. Object retrieval latency is made up of fourcomponents: (i) definition - finding the definition ofembedded objects in a page, (ii) queuing - waiting ofobject request dissemination into the network once itsexistence is defined, (iii) connection - setting up networkconnection for data transfer, and (iv) chunk transfer -transferring actual data from server to client once theconnection is setup. We show that for typical networkconnectivity, user's perceived page latency is mainly dueto the definition and queuing of embedded objectrequests inside a web page, which are related to thecontent structure of a web page and the parallelism widthof the browser for object fetching. Such understanding isimportant because it opens opportunities to improve theretrieval speed of web surfing through the minimizationof data chunk dependence.