The development of new classes of distributed applications such as telephony, remote video, and virtual reality introduces new quality requirements that make inadequate even the current best-effort service model provided by international networks. Re

We present ANCORS, an Adaptable Network Control and Reporting System that merges technology from network management and distributed simulation to provide a unified paradigm for assessing, controlling, and designing active networks. ANCORS introduces a framework to assist in managing the substantial complexities of software reuse and scalability in active network environments. Specifically, ANCORS provides an extensible approach to the dynamic integration, management, and runtime assessment of various network protocols in live network operations. We present some of the advantages that can be obtained by merging technology from network management, distributed simulation, and active networking, and describe how ANCORS leverages complementary elements of each. We also introduce an ANCORS facility called the active network daemon Anetd, which supports the deployment and system management of a large class of legacy software and newer active network applications under the ANCORS framework. Lastly, we present a prototype network engineering service that was developed to demonstrate ANCORS's capabilities.

In this paper we introduce a new classification of misses in shared-memory multiprocessors based on interprocessor communication. We identify the set of essential misses, i.e., the smallest set of misses necessary for correct execution. Essential misses include cold misses and true sharing misses. All other misses are useless misses and can be ignored without affecting the correctness of program execution. Based on the new classification we compare the effectiveness of five different protocols which delay and combine invalidations leading to useless misses. In cache-based systems the protocols are very effective and have miss rates close to the essential miss rate. In virtual shared memory systems the techniques are also effective but leave room for improvements.