DPP-PMRF: Rethinking Optimization for a Probabilistic Graphical Model Using Data-Parallel Primitives
Conference
·
OSTI ID:1471048
- Univ. of Oregon, Eugene, OR (United States)
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
We present a new parallel algorithm for probabilistic graphical model optimization. The algorithm relies on data-parallel primitives (DPPs), which provide portable performance over hardware architecture. We evaluate results on CPUs and GPUs for an image segmentation problem. Compared to a serial baseline, we observe runtime speedups of up to 13X (CPU) and 44X (GPU). We also compare our performance to a reference, OpenMP-based algorithm, and find speedups of up to 7X (CPU).
- Research Organization:
- Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
- DOE Contract Number:
- AC02-05CH11231
- OSTI ID:
- 1471048
- Country of Publication:
- United States
- Language:
- English
Similar Records
Maximal clique enumeration with data-parallel primitives
A Massively Parallel Adaptive Fast Multipole Method on Heterogeneous Architectures
The Metropolis Monte Carlo method with CUDA enabled Graphic Processing Units
Conference
·
Sun Oct 01 00:00:00 EDT 2017
·
OSTI ID:1440003
A Massively Parallel Adaptive Fast Multipole Method on Heterogeneous Architectures
Journal Article
·
Sat Dec 31 23:00:00 EST 2011
· Communications of the ACM
·
OSTI ID:1039644
The Metropolis Monte Carlo method with CUDA enabled Graphic Processing Units
Journal Article
·
Fri Jan 31 23:00:00 EST 2014
· Journal of Computational Physics
·
OSTI ID:22230865