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

Journal Article · Tue Nov 01 04:00:00 EDT 2011 · IEEE Transactions on Visualization and Computer Graphics

DOI:https://doi.org/10.1109/TVCG.2010.259· OSTI ID:1016364

Camp, David ^[1]; Garth, Christoph ^[2]; Childs, Hank ^[3]; Pugmire, David ^[4]; Joy, Kenneth I. ^[2]

---

1. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Univ. of California, Davis, CA (United States)
2. Univ. of California, Davis, CA (United States)
3. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
4. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)

Streamline computation in a very large vector field data set represents a significant challenge due to the non-local and datadependentnature of streamline integration. In this paper, we conduct a study of the performance characteristics of hybrid parallel programmingand execution as applied to streamline integration on a large, multicore platform. With multi-core processors now prevalent in clustersand 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-overblocks,and present a novel MPI-hybrid algorithm for each approach to compute streamlines. Our findings indicate that the work sharing betweencores in the proposed MPI-hybrid parallel implementation results in much improved performance and consumes less communication andI/O bandwidth than a traditional, non-hybrid distributed implementation.

• Research Organization: Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, CA (US)
• Sponsoring Organization: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21); NSF
• DOE Contract Number: AC02-05CH11231
• OSTI ID: 1016364
• Report Number(s): LBNL--4563E
• Journal Information: IEEE Transactions on Visualization and Computer Graphics, Journal Name: IEEE Transactions on Visualization and Computer Graphics Journal Issue: 11 Vol. 17; ISSN 1077-2626
• Publisher: IEEE
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING
ALGORITHMS
COMMUNICATIONS
COMPUTER ARCHITECTURE
DISTRIBUTION
HYBRID SYSTEMS
IMPLEMENTATION
MPI-hybrid
Multi-core
Streamline
PERFORMANCE
PROGRAMMING
SUPERCOMPUTERS
VECTOR FIELDS