Web performance

Abstract: From the Publisher: The #1 guide to Web capacity planning — now completely updated! A quantitative analysis of Web service availability An integrated coverage of benchmarking, load testing, workload forecasting, and performance modeling of Web services Example and case studies show how to use each technique in the latest Web services, portals, search engines, mobile and streaming-media applications A quantitative framework for planning the capacity of Web services and understanding their behavior The world's #1 book on Web capacity planning now covers the latest Web services, e-business, and mobile applications! Capacity Planning for Web Services: Metrics, Models, and Methods introduces quantitative performance predictive models for every major Web scenario, showing precisely how to identify and address both potential and actual performance problems. Coverage includes: Web services: protocols, interaction models, and unique performance, reliability, and availability challenges State-of-the-art capacity planning methodologies Spreadsheets implement the solutions of the models presenteed in the book Specific issues and workloads associated with HTTP and TCP/IP protocols Benchmarking current performance at system and component levels From accommodating current usage peaks to defining service provider SLAs, Daniel A. Menasce and Virgilio Almeida cover every aspect of capacity planning — helping you optimize every tradeoff between cost and performance. "This bookis the best guide available to understanding the uniqinvolved in delivering today's Web services." — Mark Crovella,Associate Professor, Boston University; Technical Director, Network Appliance "...asuperb starting point for anyone wishing to explore the world of Web performance." — Jeffrey P. Buzen, President of CMG; Co-Founder, BGS Systems "There is no other book like this. It is a first." — Peter J. Denning, Professor of Computer Science, George Mason University and former President of the ACM "Web servers have bursty andhighly-skewed load characteristics. This book presents a new way tomodel, analyze, and plan for these new performance problems." — Jim Gray, Senior Researcher, Microsoft Research; 1998 ACM Turing Award Recipient "...a welcome approach to the performance analysis of today'sWeb-based Internet. ... no simple and practical treatment has been offered before, andtheirs is a timely contribution." — Leonard Kleinrock, Professor of Computer Science, UCLA; Chairman and Founder, Nomadix, Inc.

Abstract: The present invention provides a method and apparatus for increasing the performance of world-wide-web traffic over the Internet. A distributed network of specialized nodes of two types is dispersed around the Internet. A web client's requests are directed to a node of the first type chosen to be close to the client, and the client communicates with this node using a standard protocol such as HTTP. This first node receives the request, and communicates the request to a node of the second type chosen to be close to the request's ultimate destination (e.g., a web server capable of generating a response to the request.) The first node communicates the request to the second node using a different, specialized, protocol that has been designed for improved performance and specifically to reduce traffic volume and to reduce latency. The second node receives communication from the first node using this specialized protocol, converts it back to a standard protocol such as HTTP, and forwards the request to the destination computer or server. Responses from the destination to the client take the corresponding reverse route, and also are carried over a specialized protocol between the two nodes. In addition, these nodes can employ other techniques such as web caches that avoid or improve some communication steps. Thus, specialized, proprietary, or complex protocols and techniques can be quickly deployed to enhance web performance without requiring significant changes to the clients or servers.

Abstract: We describe our investigation of the effect of persistent connections, pipelining and link level document compression on our client and server HTTP implementations. A simple test setup is used to verify HTTP/1.1's design and understand HTTP/1.1 implementation strategies. We present TCP and real time performance data between the libwww robot [27] and both the W3C's Jigsaw [28] and Apache [29] HTTP servers using HTTP/1.0, HTTP/1.1 with persistent connections, HTTP/1.1 with pipelined requests, and HTTP/1.1 with pipelined requests and deflate data compression [22]. We also investigate whether the TCP Nagle algorithm has an effect on HTTP/1.1 performance. While somewhat artificial and possibly overstating the benefits of HTTP/1.1, we believe the tests and results approximate some common behavior seen in browsers. The results confirm that HTTP/1.1 is meeting its major design goals. Our experience has been that implementation details are very important to achieve all of the benefits of HTTP/1.1.For all our tests, a pipelined HTTP/1.1 implementation outperformed HTTP/1.0, even when the HTTP/1.0 implementation used multiple connections in parallel, under all network environments tested. The savings were at least a factor of two, and sometimes as much as a factor of ten, in terms of packets transmitted. Elapsed time improvement is less dramatic, and strongly depends on your network connection.Some data is presented showing further savings possible by changes in Web content, specifically by the use of CSS style sheets [10], and the more compact PNG [20] image representation, both recent recommendations of W3C. Time did not allow full end to end data collection on these cases. The results show that HTTP/1.1 and changes in Web content will have dramatic results in Internet and Web performance as HTTP/1.1 and related technologies deploy over the near future. Universal use of style sheets, even without deployment of HTTP/1.1, would cause a very significant reduction in network traffic.This paper does not investigate further performance and network savings enabled by the improved caching facilities provided by the HTTP/1.1 protocol, or by sophisticated use of range requests.