Paper 18 Flash: An efficient and portable Web Server Problem The authors of this paper a new web-server architecture called the asymmetric multi process event-driven(AMPED) architecture. They argue that the performance of Flash, a web server implemented with the AMPED architecture, can match or exceed the performance of existing Web servers by upto 50%. Strengths & New-Ideas (1) The paper outlines the steps that a web-server must take to process a incoming request for a file. The entire processing of accepting a connection, reading the request, finding the file, seding the response header, reading the file and sending the data are described clearly. It also explains the potential performance pitfalls that a server must avoid while serving a file. (2) The paper discusses the three widely used server architectures-- a. Multi-process, b. Multi-threaded, c. Single-process event driven and discusses the architectural differences and the pros and cons of each of them. It gives the reader a good high-level overview of the different architectures. (3) The paper proposes a different architecture called the AMPED which is a hybrid of the SPED ad the MP archtiectures. The paper the compares the design of the AMPED with the other architectures and discusses the cost and the benefits of the different architecture for the most commonly performed web-server activities. Through the course of this discussion, it shows how the AMPED architecture is better than the other architectures. (4) The paper also describes the test and evaluation criteria in detail. The tests are well designed and support the authors arguments. Weaknesses (1) The paper is well presented and argued. There are no major weaknesses in the theoretical arguments or the evaluation criteria used in the paper. However, even though Flash is offered for use, it is surprising that it is not widely used. It is also surprising that there are no other popular web-servers that use a similar design. Testing Flash in a commercial enviroment would yield interesting results.