29117

pSTL-Bench: A Micro-Benchmark Suite for Assessing Scalability of C++ Parallel STL Implementations

Ruben Laso, Diego Krupitza, Sascha Hunold
Faculty of Informatics, TU Wien, Vienna, Austria
arXiv:2402.06384 [cs.DC], (9 Feb 2024)

@misc{laso2024pstlbench,

   title={pSTL-Bench: A Micro-Benchmark Suite for Assessing Scalability of C++ Parallel STL Implementations},

   author={Ruben Laso and Diego Krupitza and Sascha Hunold},

   year={2024},

   eprint={2402.06384},

   archivePrefix={arXiv},

   primaryClass={cs.DC}

}

Download Download (PDF)   View View   Source Source   

521

views

Since the advent of parallel algorithms in the C++17 Standard Template Library (STL), the STL has become a viable framework for creating performance-portable applications. Given multiple existing implementations of the parallel algorithms, a systematic, quantitative performance comparison is essential for choosing the appropriate implementation for a particular hardware configuration. In this work, we introduce a specialized set of micro-benchmarks to assess the scalability of the parallel algorithms in the STL. By selecting different backends, our micro-benchmarks can be used on multi-core systems and GPUs. Using the suite, in a case study on AMD and Intel CPUs and NVIDIA GPUs, we were able to identify substantial performance disparities among different implementations, including GCC+TBB, GCC+HPX, Intel’s compiler with TBB, or NVIDIA’s compiler with OpenMP and CUDA.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: