Performance evaluation of Unified Memory with prefetching and oversubscription for selected parallel CUDA applications on NVIDIA Pascal and Volta GPUs-Reference-Cited by-同舟云学术

Performance evaluation of Unified Memory with prefetching and oversubscription for selected parallel CUDA applications on NVIDIA Pascal and Volta GPUs

Published:2019-08-20 Issue:11 Volume:75 Page:7625-7645
ISSN:0920-8542
Container-title:The Journal of Supercomputing
language:en
Short-container-title:J Supercomput

Author:

Knap Marcin,Czarnul Paweł^ORCID

Publisher

Springer Science and Business Media LLC

Subject

Hardware and Architecture,Information Systems,Theoretical Computer Science,Software

Link

http://link.springer.com/content/pdf/10.1007/s11227-019-02966-8.pdf

Reference33 articles.

1. Ashcraft MB, Lemon A, Penry DA, Snell Q (2017) Compiler optimization of accelerator data transfers. Int J Parallel Prog. https://doi.org/10.1007/s10766-017-0549-3

2. Czarnul P (2018) Benchmarking overlapping communication and computations with multiple streams for modern gpus. In: Ganzha M, Maciaszek LA, Paprzycki M (eds) Communication Papers of the 2018 Federated Conference on Computer Science and Information Systems, FedCSIS 2018, Poznań, Poland, September 9–12, 2018, pp 105–110

3. Czarnul P (2018) Parallel programming for modern high performance computing systems, 1st edn. Chapman and Hall/CRC, Taylor&Francis, Boca Raton

4. Czarnul P (2018) Parallelization of large vector similarity computations in a hybrid cpu+gpu environment. J Supercomput 74(2):768–786. https://doi.org/10.1007/s11227-017-2159-7

5. Finkel H, Sharif H (2017) Openmp, unified memory, and prefetching. PADAL17: 2017-08-03, Exascale Computing Project. https://www.bnl.gov/compsci/docs/Hal-Finkel-padal_2017.pdf

Cited by 33 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Descend: A Safe GPU Systems Programming Language;Proceedings of the ACM on Programming Languages;2024-06-20

2. Shared Virtual Memory: Its Design and Performance Implications for Diverse Applications;Proceedings of the 38th ACM International Conference on Supercomputing;2024-05-30

3. Performance Portability of Generated Cardiac Simulation Kernels Through Automatic Dimensioning and Load Balancing on Heterogeneous Nodes;2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW);2024-05-27

4. GrOUT: Transparent Scale-Out to Overcome UVM's Oversubscription Slowdowns;2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW);2024-05-27

5. GPU Memory Reallocation Techniques in Fully Homomorphic Encryption Workloads;Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing;2024-04-08