Web server performance is a critical issue for sites which service a high volume of requests. This paper examines the performance of Web servers under high CPU loads. Simulations are used in conjunction with workloads which were obtained by analyzing Web logs and performance data from several real sites. Performance is significantly affected by the percentage of requests for dynamic HTML pages; dynamic HTML pages adversely affect server performance. In order to optimize performance, the number of dynamic pages should be kept as low as possible. When dynamic pages are required, techniques such as fast API's for invoking server programs and caching can be employed to keep the overhead of server programs generating the dynamic pages as low as possible. When the server is operating at or near peak capacity, there is a trade-off between average latencies and the percentage of requests rejected; performance is improved by rejecting a higher percentage of requests. For the real request distri...

This paper describes the design and implementation of an on-line, distributed data repository, known as an electronic vault (e-Vault), which stores information across a network securely so that integrity is preserved even when some of the servers malfunction. After a document has been deposited by a client, the e-Vault returns a receipt to the client which can be used to verify that the document has been properly received by all (correct) servers. The e-Vault disperses information across the servers using Rabin's Information Dispersal Algorithm (IDA). Shared distributed digital signatures are used to provide proof to the client that a document has been correctly received by all servers. Cryptographic hash functions are used to fingerprint information in order to determine if the information is later corrupted. Our e-Vault has been implemented on multiple nodes of an IBM SP2 multiprocessor. Clients communicate with the e-Vault using SSL-enabled browsers. Design extensions for confidenti...

. This paper describes techniques for improving performance at Web sites which receive significant traffic. Poor performance can be caused by dynamic data, insufficientnetwork bandwidth, and poor Web page design. Dynamic data overheads can often be reduced bycaching dynamic pages and using fast interfaces to invokeserver programs. Web server acceleration can significantly improve performance and reduce the hardware needed at a Web site. We discuss techniques for balancing load among multiple serversataWeb site. WealsoshowhowWeb pages can be designed to minimize traffic to the site. 1 Introduction Performance is a critical factor at manyWeb sites. Popular Web sites can receive hundreds of thousands of hits per minute. A Web site which only receives a moderate amount of traffic can also suffer from slow response times if a significant percent of the requests are for dynamic data which are expensive to create since dynamic data can consume orders of magnitude more CPU time to cr...

Web server performance is a critical issue for sites which service a high volume of requests. This paper examines the performance of Web servers under high CPU loads. Simulations are used in conjunction with workloads which were obtained by analyzing Web logs and performance data from several real sites. Performance is significantly affected by the percentage of requests for dynamic HTML pages; dynamic HTML pages adversely affect server performance. In order to optimize performance, the number of dynamic pages should be kept as low as possible. When dynamic pages are required, techniques such as fast API's for invoking server programs and caching can be employed to keep the overhead of server programs generating the dynamic pages as low as possible. When the server is operating at or near peak capacity, there is a trade-off between average latencies and the percentage of requests rejected; performance is improved by rejecting a higher percentage of requests. For the real request distri...