High level collective operations, i.e. function/method calls executed in
parallel on many cores, are easily applicable mechanisms to coordinate parallel activities.
 On top of architectures with ultra low latency networks like the SCC it is
 possible to coordinate groups of cores within a few microseconds only.
 Thus efficient consistency control protocols can be implemented in software
 (cache coherence, DSM etc.) on a high level of abstraction with modest
 programming effort.
 TACO (Topologies and Collections) is a distributed object platform
 that provides various template based RMI mechanisms and efficient collective
 operations for distributed object groups.  This talks discusses the first implementation
 of TACO on the SCC and shows how collective operations can be applied effectively to implement efficient
 consistency control protocols.

 Joerg Nolte is Vice Dean and Professor in Computer Science of the BTU Cottbus, Germany.
 He previously held research positions at the Fraunhofer Gesellschaft FIRST
 (Berlin, Germany) from 1988-2003 and the Parallel and Distributed Performance
 Lab (RWCP, Tsukuba, Japan) from 1994-1996 and 1999-2001.
 He received his M.S. and Ph.D. from TU Berlin, Germany, in Computer Scince in
 1088 and 1994, respectively.
 He has published papers in the areas of embedded systems and
 parallel/distributed systems.