Evaluating OpenMP 4.0's Effectiveness as a Heterogeneous Parallel Programming Model

Martineau, Matt; McIntosh–Smith, Simon; Gaudin, Wayne

doi:10.1109/ipdpsw.2016.70

Cited by 34 publications

(30 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, future work will focus on understanding this difference and investigating solutions to improving task performance on this platform. This work builds upon the success of previous mini-app work within the HPC group at the University of Bristol [16], demonstrating that mini-apps are powerful tools to both compare and test different programming models, as well investigate different language features that can lead to increased performance for a more general set of applications.…”

Section: Resultsmentioning

confidence: 93%

On the Performance of Parallel Tasking Runtimes for an Irregular Fast Multipole Method Application

Atkinson¹,

McIntosh–Smith²

2017

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

This paper will present our work on optimising and comparing the performance of an irregular algorithm for the increasingly important fast multipole method with the use of tasks. Our aim is to provide insight into how different methods of synchronisation can affect the performance of tree-based particle methods, finding that performance can be improved by 21% on some platforms. We also compare the performance of the chosen application between different OpenMP implementations and to other task-parallel programming models, finding that significant performance differences can be observed on both NUMA and Many Integrated Core architectures.

show abstract

Section: Resultsmentioning

confidence: 93%

On the Performance of Parallel Tasking Runtimes for an Irregular Fast Multipole Method Application

Atkinson¹,

McIntosh–Smith²

2017

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…The concept of attempting to represent a cross section of applications or application characteristics is becoming well recognised (e.g. Martineau et al, 2016), and others are investigating applicability to environmental models (e.g Stone et al, 2012), and noting the importance of representativeness in the selection of the mini-apps.…”

Section: Related Workmentioning

confidence: 99%

Crossing the Chasm: How to develop weather and climate models for next generation computers?

Budich¹,

Lawrence²,

Rezny³

2017

Preprint

View full text Add to dashboard Cite

Abstract.Weather and climate models are complex pieces of software which include many individual components, each of which is evolving under the pressure to exploit advances in computing to enhance some combination of a range of possible improvements (higher spatio/temporal resolution, increased fidelity in terms of resolved processes, more quantification of uncertainty etc). However, after many years of a relatively stable computing environment with little choice in processing architecture or 5 programming paradigm (basically X86 processors using MPI for parallelism), the existing menu of processor choices includes significant diversity, and more is on the horizon. This computational diversity, coupled with ever increasing software complexity, leads to the very real possibility that weather and climate modelling will arrive at a chasm which will separate scientific aspiration from our ability to develop and/or rapidly adapt codes to the available hardware.In this paper we review the hardware and software trends which are leading us towards this chasm, before describing current 10 progress in addressing some of the tools which we may be able to use to bridge the chasm. This brief introduction to current tools and plans is followed by a discussion outlining the scientific requirements for quality model codes which have satisfactory performance and portability, while simultaneously supporting productive scientific evolution. We assert that the existing method of incremental model improvements employing small steps which adjust to the changing hardware environment is likely to be inadequate for crossing the chasm between aspiration and hardware at a satisfactory pace, in part because insti-15 tutions cannot have all the relevant expertise in house. Instead, we outline a methodology based on large community efforts in engineering and standardisation, one which will depend on identifying a taxonomy of key activities -perhaps based on 1 Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2017-186 Manuscript under review for journal Geosci. Model Dev. Discussion started: 5 September 2017 c Author(s) 2017. CC BY 4.0 License. existing efforts to develop domain specific languages, identify common patterns in weather and climate codes, and develop community approaches to commonly needed tools, libraries etc -and then collaboratively building up those key components. Such a collaborative approach will depend on institutions, projects and individuals adopting new interdependencies and ways of working.

show abstract

“…Maintaining data resident on a device is generally one of the most important considerations for offloading to accelerators. We previously discussed the difficulties that are encountered when attempting to copy data to and from the device using the structured target enter data directive [2]. With OpenMP 4.0, the initial copying of resident data into the device data environment would be approached as shown in Listing 1.1.…”

Section: Structured and Unstructured Data Regionsmentioning

confidence: 99%

“…We have previously shown that it is not yet possible to write a single homogeneous line of directives to achieve performance portability with OpenMP [6] [2]. Standardisation of the compiler implementations is important for future performance portability, for instance, the newest Clang implementation automatically chooses optimal team and thread counts, so that the developer does not have to list architecture-specific values.…”

Section: Homogeneous Directivesmentioning

confidence: 99%

The Productivity, Portability and Performance of OpenMP 4.5 for Scientific Applications Targeting Intel CPUs, IBM CPUs, and NVIDIA GPUs

Martineau

McIntosh–Smith

2017

Scaling OpenMP for Exascale Performance and Portability

Self Cite

View full text Add to dashboard Cite

This research considers the productivity, portability, and performance offered by the OpenMP parallel programming model, from the perspective of scientific applications. We discuss important considerations for scientific application developers tackling large software projects with OpenMP, including straightforward code mechanisms to improve productivity and portability. Performance results are presented across multiple modern HPC devices, including Intel Xeon, and Xeon Phi CPUs, POWER8 CPUs, and NVIDIA GPUs. The results are collected for three exemplar applications: hydrodynamics, heat conduction and neutral particle transport, using modern compilers with OpenMP support. The results show that while current OpenMP implementations are able to achieve good performance on the breadth of modern hardware for memory bandwidth bound applications, our memory latency bound application performs less consistently.

show abstract

Evaluating OpenMP 4.0's Effectiveness as a Heterogeneous Parallel Programming Model

Cited by 34 publications

References 13 publications

On the Performance of Parallel Tasking Runtimes for an Irregular Fast Multipole Method Application

On the Performance of Parallel Tasking Runtimes for an Irregular Fast Multipole Method Application

Crossing the Chasm: How to develop weather and climate models for next generation computers?

The Productivity, Portability and Performance of OpenMP 4.5 for Scientific Applications Targeting Intel CPUs, IBM CPUs, and NVIDIA GPUs

Contact Info

Product

Resources

About