Abstract
In the past few years, the Internet and World Wide Web (www)have become more and more popular. At the same time, The Web has become an improtant infrastructure for distributed computing, as it provides rich machine resources connected via Internet. How to partition the problems and to distribute tasks to machine resources on internet are recently the important issues. There are two opposite sides, server side and client side, to perform the informatiom and computing power sharing on the Web. The Common Gateway Interface(CGI) on server side cooperate with the HTTP daemon to proces the request from Web client. On the client side, the Java technology has gained an important role for developments of future Web and Internet softwares as well as for the support of parallel and distibuted computings. In this thesis, we investigate how the emerging Web technologies can be integrated with parallel compilers, which have long been a major focus in our lab. We come up with tow innovative ideas in integrating Web technologies with parallel compilers. First, we demonstrate a server side compiler and environment to optimize HPF programs. The environment is machine-independent and with interactive environ-ment. Second, we propose optimizations with parallel Java environments on the client side. A collection of generic matrix classes written in Java and a runtime environment are presented to support automatic parallelization of sparse computations on distributed environments.