In this paper, we present a design methodology management system, which assists designers in selecting a suitable design process and invoking the selected sequence of tools on the correct versions of design data.We i n troduce a formal graph representation of design methodologies in which nodes represent either tasks or design data.Using a graph grammar, nodes representing abstract tasks are replaced by graphs of less abstract tasks and intermediate speci cations.Graph grammars enable us to concisely and exibly describe a large class of methodologies.Often there are several alternative methodologies or tools available for some subtasks, leading to dierent results.Our system utilizes automated control agents, in combination with user interaction, to select among methodologies and invoke tools.Multiple alternatives can easily be explored simultaneously.

A new upproach to cache coherent networks is proposed and analyzed.This approuch is highly scalable, so it is appropriate for massively parallel multiprocessors.The network switches of a MIN are enhanced such that they are capable of handling the majority of the cache coherence burden.The latency for both reads and writes is O(log N) independent of the number of copies.The memory overhead of this approach is O(Clog2 N).Permis.sica to copy without fee all or pan of this material is granted, pmwded that the copies are not made or diarlwed for dired -tial sdvantage, tbe ACM copyright nctice and the title of the cubbcation and its date appear, and rmke ~given that copying is by pekksion d the As.?ociation for Computing Machinefy.T. copy otherwise, w to qublish, quires a fw andkm spscitic pemniskm.

Techniques are proposed for computing a global virtual time (GVT), which is the minimum local virtual time of processes in Time Warp. The algorithm computes a conservative estimate of GVT using an approach which is considerably simpler than previous algorithms for computing GVT. This algorithm does not require a global synchronization of processors. An inherent problem is GVT computation relates to handling messages in transit. Several alternatives are proposed for solving the transient message problem. The algorithm is suitable for distributed shared memory machines such as the BBN Butterfly and message passing machines with a variety of interconnection networks.>