We advocate the Loop-of-stencil-reduce pattern as a means of simplifying the implementation of data-parallel programs on heterogeneous multi-core platforms. Loop-of-stencil-reduce is general enough to subsume map, reduce, map-reduce, stencil, stencil-reduce, and, crucially, their usage in a loop in both data-parallel and streaming applications, or a combination of both. The pattern makes it possible to deploy a single stencil computation kernel on different GPUs. We discuss the implementation of Loop-of-stencil-reduce in FastFlow, a framework for the implementation of applications based on the parallel patterns. Experiments are presented to illustrate the use of Loop-of-stencil-reduce in developing data-parallel kernels running on heterogeneous systems

In this paper, a highly effective parallel filter for visual data restoration is presented. The filter is designed following a skeletal approach, using a newly proposed stencil-reduce, and has been implemented by way of the FastFlow parallel programming library. As a result of its high-level design, it is possible to run the filter seamlessly on a multicore machine, on multi-GPGPUs, or on both. The design and implementation of the filter are discussed, and an experimental evaluation is presented.

Abstract-In this paper we advocate the Loop-of-stencilreduce pattern as a way to simplify the parallel programming of heterogeneous platforms (multicore+GPUs). Loop-of-Stencilreduce is general enough to subsume map, reduce, map-reduce, stencil, stencil-reduce, and, crucially, their usage in a loop. It transparently targets (by using OpenCL) combinations of CPU cores and GPUs, and it makes it possible to simplify the deployment of a single stencil computation kernel on different GPUs. The paper discusses the implementation of Loop-of-stencilreduce within the FastFlow parallel framework, considering a simple iterative data-parallel application as running example (Game of Life) and a highly effective parallel filter for visual data restoration to assess performance. Thanks to the high-level design of the Loop-of-stencil-reduce, it was possible to run the filter seamlessly on a multicore machine, on multi-GPUs, and on both.

Abstract-This paper presents the optimisation efforts on the creation of a graph-based mapping representation of gene adjacency. The method is based on the Hi-C process, starting from Next Generation Sequencing data, and it analyses a huge amount of static data in order to produce maps for one or more genes. Straightforward parallelisation of this scheme does not yield acceptable performance on multicore architectures since the scalability is rather limited due to the memory bound nature of the problem. This work focuses on the memory optimisations that can be applied to the graph construction algorithm and its (complex) data structures to derive a cache-oblivious algorithm and eventually to improve the memory bandwidth utilisation. We used as running example NuChart-II, a tool for annotation and statistic analysis of Hi-C data that creates a gene-centric neighborhood graph. The proposed approach, which is exemplified for Hi-C, addresses several common issue in the parallelisation of memory bound algorithms for multicore. Results show that the proposed approach is able to increase the parallel speedup from 7x to 22x (on a 32-core platform). Finally, the proposed C++ implementation outperforms the first R NuChart prototype, by which it was not possible to complete the graph generation because of strong memory-saturation problems.

Abstract. The MPI-2 standard has been implemented for a few years in most of the MPI distributions. As MPI-1.2, it leaves it up to the user to decide when and where the processes must be run. Yet, the dynamic creation of processes, enabled by MPI-2, turns it harder to handle their scheduling manually. This paper presents a scheduler module, that has been implemented with MPI-2, that determines, on-line (i.e. during the execution), on which processor a newly spawned process should be run. The scheduler can apply a basic Round-Robin mechanism or use load information to apply a list scheduling policy, for MPI-2 programs with dynamic creation of processes. A rapid presentation of the scheduler is given, followed by experimental evaluations on three test programs: the Fibonacci computation, the N -Queens benchmark and a computation of prime numbers. Even with the basic mechanisms that have been implemented, a clear gain is obtained regarding the run-time, the load balance, and consequently regarding the number of processes that can be run by the MPI program.