Efficient Inspected Critical Sections in Data-Parallel GPU Codes
Blaß T, Philippsen M, Veldema R (2019)
Publication Language: English
Publication Type: Conference contribution, Original article
Publication year: 2019
Publisher: Springer International Publishing
Series: Lecture Notes in Computer Science (LNCS)
City/Town: Cham
Pages Range: 223-239
Conference Proceedings Title: Proceedings of the 30th International Workshop on Languages and Compilers for Parallel Computing (LCPC 2017)
Event location: College Station, TX
ISBN: 978-3-030-35224-0
URI: https://www2.cs.fau.de/publication/download/lcpc2017_blass.pdf
DOI: 10.1007/978-3-030-35225-7_15
Abstract
Optimistic concurrency control and STMs rely on the assumption of sparse conflicts. For data-parallel GPU codes with many or with dynamic data dependences, a pessimistic and lock-based approach may be faster, if only GPUs would offer hardware support for GPU-wide fine-grained synchronization. Instead, current GPUs inflict dead- and livelocks on attempts to implement such synchronization in software.
The paper demonstrates how to build GPU-wide non-hanging critical sections that are as easy to use as STMs but also get close to the performance of traditional fine-grained locks. Instead of sequentializing all threads that enter a critical section, the novel programmer-guided Inspected Critical Sections (ICS) keep the degree of parallelism up. As in optimistic approaches threads that are known not to interfere, may execute the body of the inspected critical section concurrently.
Authors with CRIS profile
Thorsten Blaß
Lehrstuhl für Informatik 2 (Programmiersysteme)
Michael Philippsen
Lehrstuhl für Informatik 2 (Programmiersysteme)
Ronald Veldema
Lehrstuhl für Informatik 2 (Programmiersysteme)
Related research project(s)
How to cite
APA:
Blaß, T., Philippsen, M., & Veldema, R. (2019). Efficient Inspected Critical Sections in Data-Parallel GPU Codes. In Rauchwerger, Lawrence (Eds.), Proceedings of the 30th International Workshop on Languages and Compilers for Parallel Computing (LCPC 2017) (pp. 223-239). College Station, TX, US: Cham: Springer International Publishing.
MLA:
Blaß, Thorsten, Michael Philippsen, and Ronald Veldema. "Efficient Inspected Critical Sections in Data-Parallel GPU Codes." Proceedings of the 30th International Workshop on Languages and Compilers for Parallel Computing (LCPC 2017), College Station, TX Ed. Rauchwerger, Lawrence, Cham: Springer International Publishing, 2019. 223-239.
BibTeX: Download