POM: A Parallel Observable Machine

POM is a Parallel Observable Machine featuring mechanisms for building and observing distributed applications. It comes in the form of a library built upon the many communication kernels available on current parallel architectures and a loader that provides the user with a homogeneous syntax for launching parallel applications on any parallel platform. The prior goal of POM is not to o er numerous services to the application programmer as it is done in PVM [2], MPI [9] or P4 [3]. It mostly aims at masking the speci cities of the various communication kernels of today's machines with no signi cant degradation of performances. In that sense, our approach is quite similar to that of projects PICL [5] and PARMACS [4]. Yet, one of our main priorities while designing POM was to de ne a model of virtual machine and to clearly specify the semantics of the communications in this model. We also wanted to de ne an easily portable machine |i.e. a machine that can be ported on a given platform in a short time| and whose implementation can be achieved e ciently on many parallel platforms. POM is also provided with sophisticated observation mechanisms that are incorporated at a low level so as to keep perturbations at a minimum. The observation technique fostered in POM is based on the analysis of execution traces, rather than on a direct observation of distributed applications (in which case the observation is achieved inside the application). Besides, all the observation facilities can be enabled or disabled separately.