Abstract | ||
---|---|---|
We introduce a task-parallel framework for non-intrusive Bayesian Uncertainty Quantification and Propagation of complex and computationally demanding physical models on massively parallel computing architectures. The framework incorporates Laplace asymptotic approximations and stochastic algorithms along with distributed numerical differentiation. Sampling is based on the Transitional Markov Chain Monte Carlo algorithm and its variants while the optimization tasks associated with the asymptotic approximations are treated via the Covariance Matrix Adaptation Evolution Strategy. Exploitation of task-based parallelism is based on a platform-agnostic adaptive load balancing library that orchestrates scheduling of multiple physical model evaluations on computing platforms that range from multicore systems to hybrid GPU clusters. Experimental results using representative applications demonstrate the flexibility and excellent scalability of the proposed framework. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1007/978-3-662-48096-0_41 | Lecture Notes in Computer Science |
Keywords | Field | DocType |
Task-based parallelism,Bayesian uncertainty quantification | Uncertainty quantification,Computer science,Scheduling (computing),Massively parallel,Load balancing (computing),Parallel computing,Evolution strategy,CMA-ES,Bayesian probability,Scalability | Conference |
Volume | ISSN | Citations |
9233 | 0302-9743 | 2 |
PageRank | References | Authors |
0.42 | 5 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Panagiotis E. Hadjidoukas | 1 | 109 | 18.24 |
Panagiotis Angelikopoulos | 2 | 3 | 2.15 |
Lina Kulakova | 3 | 2 | 0.42 |
C. Papadimitriou | 4 | 15 | 2.90 |
Petros Koumoutsakos | 5 | 1065 | 84.99 |