Javelin++: scalability issues in global computing

Javelin is a Java-based infrastructure for global computing. This paper pre sents Javelin++, an extension of Javelin, intended to support a much larger set of computational hosts. Contributions to scalability and fault toleran ce are presented. This is the focus of the paper, Two scheduling schemes ar e presented: probabilistic work stealing and deterministic work stealing. T he distributed deterministic work stealing is integrated with a distributed deterministic eager scheduler, which is one of the paper's primary origina l contributions, An additional fault tolerance mechanism is implemented for replacing hosts that have failed or retreated. A Javelin++ API is sketched , then illustrated on a raytracing application. Performance results for the two schedulers are reported, indicating that Javelin++, with its broker ne twork, scales better than the original Javelin, Copyright (C) 2000 John Wil ey & Sons, Ltd.