Implementing High-Performance Geometric Multigrid Solver with Naturally Grained Messages
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Structured-grid linear solvers often require manually packing and unpacking of communication data to achieve high performance.Orchestrating this process efficiently is challenging, labor-intensive, and potentially error-prone.In this paper, we explore an alternative approach that communicates the data with naturally grained messagesizes without manual packing and unpacking. This approach is the distributed analogue of shared-memory programming, taking advantage of the global addressspace in PGAS languages to provide substantial programming ease. However, its performance may suffer from the large number of small messages. We investigate theruntime support required in the UPC ++ library for this naturally grained version to close the performance gap between the two approaches and attain comparable performance at scale using the High-Performance Geometric Multgrid (HPGMG-FV) benchmark as a driver.
- 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:
- 1378602
- Journal Information:
- Proceedings - 2015 9th International Conference on Partitioned Global Address Space Programming Models, PGAS 2015, Journal Name: Proceedings - 2015 9th International Conference on Partitioned Global Address Space Programming Models, PGAS 2015
- Country of Publication:
- United States
- Language:
- English
Similar Records
Global Arrays Parallel Programming Toolkit
Tuning collective communication for Partitioned Global Address Space programming models