Paper #: Week 9 Paper 18
Title: Flash: An efficient and portable web server.

(1) Problems
This paper discusses the design and performance of a Web server architecture called the Asymmetric multiprocess event driven (AMPED). The authors have implemented Flash which is a web server that follows this architecture. This new architecture combines good performance aspects from single process event driven servers and from multi process and multithreaded servers and thus yields a superior performance.

  
(2) New Idea and Strengths
The paper has a lot of strengths:
- This paper is very well written and provides a very clear description of web server architecture in general and their drawbacks.
- Then the authors also describe in a clear fashion their own architecture and its implementation - Flash.
- The use of simple diagrams to explain the working of the different models enhances the user knowledge and understanding.
- The idea is pretty nice. By combining the good aspects of architectures that perform well on cached workloads with architectures that perform well on disk bound workloads the authors come up with an architecture that yields excellent results.
- Another added advantage is that Flash uses facilities (standard APIs) that are available in all modern operating systems, thus it is portable. 
- The paper also has backed up their design with good experimental results. 

(3) Weaknesses and Extensions
It is very hard to find weaknesses in this paper. The authors have done a marvellous job in keeping the readers interested. The language is simple to understand and the design is clear. They have explained the motivation, the background and similar work very well. Also, they have actually implemented a system based on the AMPED design to prove their results. Thus, I think they have left no stone unturned as such.


All in all I think this is a very good paper and the authors have done a great job of making it easy to read and understand.

-- END --