High-performance dataflow computing in hybrid memory systems with UPC++ DepSpawn
![Thumbnail](/dspace/bitstream/handle/2183/28158/Fraguela_Basilio_2021_High-performance_dataflow_computing.pdf.jpg?sequence=5&isAllowed=y)
Use this link to cite
http://hdl.handle.net/2183/28158Collections
- GI-GAC - Artigos [192]
Metadata
Show full item recordTitle
High-performance dataflow computing in hybrid memory systems with UPC++ DepSpawnDate
2021Citation
Fraguela, B.B., Andrade, D. High-performance dataflow computing in hybrid memory systems with UPC++ DepSpawn. J Supercomput 77, 7676–7689 (2021). https://doi.org/10.1007/s11227-020-03607-1
Abstract
[Abstract]: Dataflow computing is a very attractive paradigm for high-performance computing, given its ability to trigger computations as soon as their inputs are available. UPC++ DepSpawn is a novel task-based library that supports this model in hybrid shared/distributed memory systems on top of a Partitioned Global Address Space environment. While the initial version of the library provided good results, it suffered from a key restriction that heavily limited its performance and scalability. Namely, each process had to consider all the tasks in the application rather than only those of interest to it, an overhead that naturally grows with both the number of processes and tasks in the system. In this paper, this restriction is lifted, enabling our library to provide higher levels of performance. This way, in experiments using 768 cores the performance improved up to 40.1%, the average improvement being 16.1%.
Keywords
Dataflow computing
Hybrid parallelism
PGAS
Runtimes
Programability
High-performance computing
Hybrid parallelism
PGAS
Runtimes
Programability
High-performance computing