Distributed and clustered systems

The Distributed Systems & Clusters group at IBM is dedicated to providing state-of-the-art solutions to real-world distributed computing problems. As researchers at the IBM Thomas J. Watson Research center, our work centers on solving hard problems arising from the business needs of IBM and its customers. Our work to date has required the development and implementation of cutting edge technology in the areas of clustered computing, fault tolerance, system management,  and  internet related technologies. The results of our work have been implemented as products on a variety of platforms including IBM's RS6000 SP line of clusters, Windows NT clustering, OS/2, and Java.

We prototyped and participated in the development of several key components of the Phoenix cluster software stack.  Phoenix provides infrastructure for high availability and scalability on the RS6000 SP by implementing reliable messaging, synchronization and coordination primitives, and group membership services to higher level applications. We are extending the scope of our previous work on Phoenix to span the internet, and to provide novel cluster management abilities.

Comments and further information please contact:  Srirama Krishnakumar

[Research home page]  [Distributed and Clustered Systems Home Page]

[ IBM home page | Order | Search | Contact IBM | Legal ]