Network Computing
by Tom Chen
Scott McNealy of Sun Microsystems is famous for saying that the "network
is the computer" for several years before most people understood what
it meant. This observation was remarkable in predicting the profound impact
of the Internet on modern computing. Historically, networking (data communications)
and computing (data processing) technologies have developed separately,
with the network simply acting as a "dumb" data carrier between powerful
computing hosts. The Internet, however, has moved to the center of computing.
Instead of stand-alone personal computers, today we are seeing a migration
of computing into the network. The best example of this modern paradigm
may be the World Wide Web which is a vast interconnected repository of
data accessible any time, anywhere through the Internet. At present, processing
and storage have not moved into the network, which we believe will be
the next step in the Internet evolution. This research project is directed
towards a vision of Internet computing where the Internet will evolve
to a coordinated collection of high-performance networks, distributed
computers, storage systems, and middleware (sometimes called a computation
grid). In this vision, the internet will be a powerful and pervasive computing
utility, where access to the combined distributed computing resources
of the network will be as simple and commonplace as electric power or
telephony today. Many of the necessary basic technologies - high performance
networks, distributed databases, supercomputers - exist today but have
not been designed for the enormous scale and heterogeneity of the Internet
environment. Resource management in high performance networks is a major
activity in this research project. Another major research activity is
in middleware, the crucial layer of software for discovering and coordinating
resources in support of distributed applications.
Projects
2002-2003
2001-2002
Technical Reports
Special Emphasis
