Dynamic Node Recon guration in a Parallel-Distributed Environment

Idle workstations in a network represent a signi cant computing potential. In particular, their processing power can be used by parallel-distributed programs that treat the network as a loosely-coupled multiprocessor. But the set of machines free to participate in load sharing changes over time as users come and go from their workstations. To make full use of the available resources, parallel-distributed applications in the network must recon gure to adapt to these changes as they run. This paper describes a node recon guration facility for Amber, an object-based parallel programming system for networks of multiprocessors. We describe system support that allows a parallel Amber application to adapt to changing network conditions by expanding to make use of new nodes as they become idle, and by contracting as nodes become busy. A key characteristic of Amber's node recon guration is that it is handled at the user level in the Amber runtime system; it does not depend on a kernel-level process migration facility. Our experiments with Amber show that node recon guration can be implemented easily and e ciently in a runtime library.