Self-Adaptation of Parallel Applications in Heterogeneous and Dynamic Architectures

In this paper a mechanism for adaptation of parallel computation is defined for data flow computations in dynamic and heterogeneous environments. Our mechanism is especially useful in massively parallel multi-threaded computations as found in cluster or grid computing. By basing the state of executions on a data flow graph, this approch shows extreme flexibility with respect to adaptation of parallel computation induced by application. This adaptation reflects needs for changing runtime behavior due to time observable parameters. Specifically, it allows an on-line adaptation of parallel execution in dynamic heterogeneous systems. We have implemented this mechnism in KAAPI (Kernel for Adaptative and Asynchronous Parallel Interface) and experimental results show the overhead induced is small.