DPP-PMRF: Rethinking Optimization for a Probabilistic Graphical Model Using Data-Parallel Primitives

Conference · Sun Oct 21 00:00:00 EDT 2018

OSTI ID:1471048

Lessley, Brenton ^[1]; Perciano, Talia ^[2]; Heinemann, Colleen ^[2]; Camp, David ^[2]; Childs, Hank ^[1]; Bethel, E. Wes ^[2]

---

1. Univ. of Oregon, Eugene, OR (United States)
2. 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 Lab. (LBNL), Berkeley, CA (United States)
• Sponsoring Organization: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
• DOE Contract Number: AC02-05CH11231
• OSTI ID: 1471048
• Resource Relation: Conference: 8. IEEE Symposium on Large Data Analysis and Visualization (LDAV 2018), Berlin (Germany), 21 Oct 2018
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING