Quantifying Overheads in Charm++ and HPX using Task Bench

Wu, Nanmiao; Gonidelis, Ioannis; Liu, Simeng; Fink, Zane; Gupta, Nidhi; Mohammadiporshokooh, Karame; Diehl, Patrick; Kaiser, Hartmut; Kalé, Laxmikant V.

doi:10.48550/arxiv.2207.12127

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Like HPX, Charm++ also provides facilities for distributed programming (for which, at present, the C++ provides no standard). For a comparison of Charm++ and HPX with OpenMP and MPI (a widely accepted standard for distributed parallel programming) using Task Bench, we refer to [10]. Other notable AMTS are: Chapel [1], X10 [2], and UPC++ [12].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Shared memory parallelism in Modern C++ and HPX

Diehl¹,

Brandt²,

Kaiser³

2023

Preprint

View full text Add to dashboard Cite

Parallel programming remains a daunting challenge, from the struggle to express a parallel algorithm without cluttering the underlying synchronous logic, to describing which devices to employ in a calculation, to correctness. Over the years, numerous solutions have arisen, many of them requiring new programming languages, extensions to programming languages, or the addition of pragmas. Support for these various tools and extensions is available to a varying degree. In recent years, the C++ standards committee has worked to refine the language features and libraries needed to support parallel programming on a single computational node. Eventually, all major vendors and compilers will provide robust and performant implementations of these standards. Until then, the HPX library and runtime provides cutting edge implementations of the standards, as well as proposed standards and extensions. Because of these advances, it is now possible to write high performance parallel code without custom extensions to C++. We provide an overview of modern parallel programming in C++, describing the language and library features, and providing brief examples of how to use them.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Here, the overhead of using HPX is negligible. For more details on the overheads of HPX and Charm++, we refer to [10]. For HPX's parallel algorithms using hpx::for_each (b), AMD performed better as Intel and Arm is around one order of magnitude less.…”

Section: Performance Comparisonmentioning

confidence: 99%