Academic Journal

Streamline Integration Using MPI-Hybrid Parallelism on a Large Multicore Architecture

التفاصيل البيبلوغرافية
العنوان: Streamline Integration Using MPI-Hybrid Parallelism on a Large Multicore Architecture
المؤلفون: David Camp, Christoph Garth, Hank Childs, Dave Pugmire, Kenneth I. Joy
المساهمون: The Pennsylvania State University CiteSeerX Archives
المصدر: http://vis.lbl.gov/Publications/2010/LBNL-4563E.pdf.
المجموعة: CiteSeerX
مصطلحات موضوعية: Ç
الوصف: —Streamline computation in a very large vector field data set represents a significant challenge due to the nonlocal and datadependent nature of streamline integration. In this paper, we conduct a study of the performance characteristics of hybrid parallel programming and execution as applied to streamline integration on a large, multicore platform. With multicore processors now prevalent in clusters and supercomputers, there is a need to understand the impact of these hybrid systems in order to make the best implementation choice. We use two MPI-based distribution approaches based on established parallelization paradigms, parallelize over seeds and parallelize over blocks, and present a novel MPI-hybrid algorithm for each approach to compute streamlines. Our findings indicate that the work sharing between cores in the proposed MPI-hybrid parallel implementation results in much improved performance and consumes less communication and I/O bandwidth than a traditional, nonhybrid distributed implementation. Index Terms—Concurrent programming, parallel programming, modes of computation, parallelism and concurrency, picture/image generation, display algorithms.
نوع الوثيقة: text
وصف الملف: application/pdf
اللغة: English
Relation: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.299.2732; http://vis.lbl.gov/Publications/2010/LBNL-4563E.pdf
الاتاحة: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.299.2732
http://vis.lbl.gov/Publications/2010/LBNL-4563E.pdf
Rights: Metadata may be used without restrictions as long as the oai identifier remains attached to it.
رقم الانضمام: edsbas.35931761
قاعدة البيانات: BASE