PAGANI: a parallel adaptive GPU algorithm for numerical integration

Conference · Thu Dec 31 23:00:00 EST 2020 · No journal information

DOI:https://doi.org/10.1145/3458817.3476198· OSTI ID:1781076

Sakiotis, Ioannis ^[1]; Arumugam, Kamesh ^[2]; Paterno, Marc ^[3]; Ranjan, Desh ^[1]; Terzić, Balša ^[1]; Zubair, Mohammad ^[1]

---

1. Old Dominion U.
2. NVIDIA, Santa Clara
3. Fermilab

We present a new adaptive parallel algorithm for the challenging problem of multi-dimensional numerical integration on massively parallel architectures. Adaptive algorithms have demonstrated the best performance, but efficient many-core utilization is difficult to achieve because the adaptive work-load can vary greatly across the integration space and is impossible to predict a priori. Existing parallel algorithms utilize sequential computations on independent processors, which results in bottlenecks due to the need for data redistribution and processor synchronization. Our algorithm employs a high-throughput approach in which all existing sub-regions are processed and sub-divided in parallel. Repeated sub-region classification and filtering improves upon a brute-force approach and allows the algorithm to make efficient use of computation and memory resources. A CUDA implementation shows orders of magnitude speedup over the fastest open-source CPU method and extends the achievable accuracy for difficult integrands. Our algorithm typically outperforms other existing deterministic parallel methods.

• Research Organization: Old Dominion U.; Fermi National Accelerator Laboratory (FNAL), Batavia, IL (United States); NVIDIA, Santa Clara
• Sponsoring Organization: US Department of Energy
• DOE Contract Number: AC02-07CH11359
• OSTI ID: 1781076
• Report Number(s): FERMILAB-CONF-21-081-SCD; oai:inspirehep.net:1861820; arXiv:2104.06494
• Journal Information: No journal information, Journal Name: No journal information
• Country of Publication: United States
• Language: English