Efficient back-end prefetching scheme in cluster-based web servers = 클러스터 기반 웹 서버 상에서 실제 서버간 효율적인 선인출 기법

With the explosive growth of WWW traffic, there is an increasing demand for the high performance Web servers to provide a stable Web service to users. The cluster-based Web server is a solution to cope with the heavy access from users, easily scaling the server according to the loads. In the cluster-based Web server, a back-end node may not be able to serve some HTTP requests directly because it does not have the requested contents in its main memory. In this case, the back-end node has to retrieve the requested contents from its local disk or other back-end nodes in the cluster. To reduce service latency, we introduce a new prefetch scheme. The back-end nodes predict the next HTTP requests and prefetch the contents of predicted requests before the next requests arrive. We develop three prefetch algorithms based on some useful information gathered from many clients’ HTTP requests. Through trace-driven simulation, Time and Access Probability-based Prefetch ($TAP^2$) algorithm, which uses the access probability and the inter-reference time of Web objects, shows the best performance among the proposed prefetch algorithms. With $TAP^2$ algorithm, the service latency is reduced by 20.1% in a small sized memory and 1.5% in a large sized memory, comparing with none-prefetch mechanism.
Advisors
Cho, Jung-WanresearcherLee, Joon-Wonresearcher조정완researcher이준원researcher
Publisher
한국과학기술원
Issue Date
2001
Identifier
165487/325007 / 000993223
Language
eng
Description

학위논문(석사) - 한국과학기술원: 전산학전공, 2001.2, [ ii, 39 p. ]

Keywords

Web server; prefetching; 클러스터; 웹 서버; cluster; 선인출

URI
http://hdl.handle.net/10203/34423
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=165487&flag=t
Appears in Collection
CS-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.
  • Hit : 116
  • Download : 0
  • Cited 0 times in thomson ci

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0