During the last few years, research on consistency models, protocols and applications has allowed the software DSM field to become relatively mature. At the same time, general-purpose networks and communication packages have advanced to the point where clusters of workstations can be coupled almost as closely as hardware multi-processors.

The theme of this workshop, then, will be an attempt to answer the questions of What remains to be done in DSM? And more to the point, Will DSM ever move into the mainstream?

To this end, we solicit abstracts of original DSM research including, but not limited to, the following areas: applications, clusters of multiprocessors, compiler support, experience with real applications, fault tolerance, hardware support, memory models, multi-threading, online adaptation, and performance/debugging tools.

Important Dates: April 15, 1999: Papers due. April 25, 1999: Automatic extension. April 30, 1999: Notification sent to authors. May 18, 1999: Camera-ready papers due.

Program Chairs: Liviu Iftode (iftode@cs.rutgers.edu) Peter Keleher (keleher@cs.umd.edu)

PDF or PS submissions should be emailed to wsdsm@cs.umd.edu, along with complete postal address, e-mail address, phone number, and fax number for the contact author.