This paper presents a checkpointing scheme that was implemented in a parallel library that runs on top of CHIMP/MPI. The main goals of the checkpointing mechanism are portability and efficiency. It runs on every platform supported by MPI in a machine-independent way. The scheme allows the migration of checkpoints and offers a flexible recovery mechanism based on data-reconfiguration. Some performance results will be presented at the end of the paper together with some techniques that can be used to increase the efficiency of the checkpointing mechanism.

Abstract Message‐based MIMD computers that support arbitrary connections over a small number of communications links cannot support parallel applications with complex communication patterns unless the support software includes message routing. This paper describes design issues for such software, using a router for the Inmos T800 transputer called Tiny as an example. Performance figures show that Tiny achieves near optimal routing performance given the hardware limits. Modifications to Tiny to monitor message traffic are also described.