News & Events

  • Professor Dr. Andrew Lumsdaine will deliver Monday's Keynote

    Date: 05/18/2014

    HIPS 2014-19th International Workshop on High-Level Parallel Programming Models and Supportive Environments

    Held in Conjunction With IPDPS 2014 Phoenix

    Overview
    The 19th HIPS workshop, to be held as a full-day meeting on May 19, 2014 at the IPDPS 2014 conference in Phoenix, focuses on high-level programming of multiprocessors, compute clusters, and massively parallel machines. Like previous workshops in the series, which was established in 1996, this event serves as a forum for research in the areas of parallel applications, language design, compilers, runtime systems, and programming tools. It provides a timely and lightweight forum for scientists and engineers to present the latest ideas and findings in these rapidly changing fields. In our call for papers, we especially encouraged innovative approaches in the areas of emerging programming models for large-scale parallel systems and many-core architectures.

    Topics of interest to the HIPS workshop include but are not limited to:

    •New programming languages and constructs for exploiting parallelism and locality
    •Experience with and improvements for existing parallel languages and run-time environments such as MPI, OpenMP, Cilk, UPC, Co-array Fortran, X10, and Chapel
    •Parallel compilers, programming tools, and environments
    •(Scalable) tools for performance analysis, modeling, monitoring, and debugging
    •OS and architectural support for parallel programming and debugging
    •Software and system support for extreme scalability including fault tolerance
    •Programming environments for heterogeneous multicore systems and accelerators such as GPUs, FPGAs, Cells, and MICs

    Monday, May 19, 2014

    8:15 -- 8:30 Opening Remarks
    8:30 -- 9:30 Title: The (Parallel) BGL: A High-Performance Parallel Graph Algorithms Library

    Andrew Lumsdaine
    Professor, Computer Science Department, Indiana University
    Director, Center for Research in Extreme Scale Technologies (CREST)

    Abstract: Graphs and graph algorithms have long been a fundamental abstraction in computer science and many types of data-driven applications - the emerging "fourth pillar" of science - depend on graph computations. The resource requirements for graph computations can be quite large, however, running graph algorithms on today's HPC systems presents a number of challenges - the paradigms, software, and hardware that have worked well for mainstream scientific applications are not well matched to large-scale graph problems.

    In this talk we present the design and implementation of multiple generations of the Boost Graph Library, a library of reusable high-performance software components for graph computation. The original (sequential) BGL applies the paradigm of generic programming to the domain of graph computations. Subsequent (parallel) generations of the BGL were able to be built from the sequntial BGL lifting away the implicit requirements of sequential execution and a single shared address space. This process allows us to create generic algorithms having sequential expression and requiring only the introduction of parallel data structures for parallel execution. By characterizing these extensions as well as the extension process, we develop general principles and patterns for using (and reusing) generic parallel software libraries. We demonstrate that the resulting algorithm implementations are both efficient and scalable with performance results for several algorithms implemented in the open-source Parallel Boost Graph Library. We conclude by discussing on-going and future work, most notably the new active-message system being incorporated into PBGL to enable efficient execution on multi-core, hybrid, and exascale architectures.