Skip to main content

Custom Data Mapping for Composable Data Management

  • Conference paper
  • First Online:
Scaling OpenMP for Exascale Performance and Portability (IWOMP 2017)

Part of the book series: Lecture Notes in Computer Science ((LNPSE,volume 10468))

Included in the following conference series:

  • 979 Accesses

Abstract

Early experiences with OpenMP 4.0, as well as other directive-based offload models, have shown that deep copy is a key challenge to porting complex applications to offload directives. Without a flexible deep-copy mechanism, pointer-based data structures are at best difficult to manage, particularly when shared memory between the host and device cannot be assumed. Despite the importance of the issue, and the considerable effort expended by vendors, standards bodies and users, no solution has emerged as the clear choice. We propose an approach that combines a restricted compiler-assisted (sometimes called “true”) deep copy with a mechanism for users to register their own custom mapping implementations that we call packers. This combination offers the flexibility to address complex cases when necessary while keeping the complexity out of the directives, a balance that serves all cases.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 (LLNL-CONF-733464).

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

References

  1. Beyer, J., Oehmke, D., Sandoval, J.: Transferring user-defined types in OpenACC. In: Proceedings of Cray User Group (CUG14) (2014)

    Google Scholar 

  2. Hoshino, T., Maruyama, N., Matsuoka, S.: An OpenACC extension for data layout transformation. In: 2014 First Workshop on Accelerator Programming Using Directives (WACCPD), pp. 12–18. IEEE (2014)

    Google Scholar 

  3. Karlin, I., et al.: Early Experiences Porting Three Applications to OpenMP 4.5. In: Maruyama, N., Supinski, B.R., Wahib, M. (eds.) IWOMP 2016. LNCS, vol. 9903, pp. 281–292. Springer, Cham (2016). doi:10.1007/978-3-319-45550-1_20

    Chapter  Google Scholar 

  4. OpenACC Standards Committee: Deep copy attach and detach. Technical report TR-16-1 (2016)

    Google Scholar 

  5. OpenACC Working Group and Others: The OpenACC Application Programming Interface, Version 2.5 (2015)

    Google Scholar 

  6. OpenMP Architecture Review Board: OpenMP Application Programming Interface, Version 4.5 (2015)

    Google Scholar 

  7. OpenMP Architecture Review Board: OpenMP Technical Report 4: Version 5.0 Preview 1. Technical report TR-4 (2016)

    Google Scholar 

  8. OpenMP Architecture Review Board: OpenMP Technical Report 5: Memory Management Support for OpenMP 5.0. Technical report TR-5 (2017)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Tom Scogland .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2017 Springer International Publishing AG

About this paper

Cite this paper

Scogland, T., Earl, C., de Supinski, B. (2017). Custom Data Mapping for Composable Data Management. In: de Supinski, B., Olivier, S., Terboven, C., Chapman, B., Müller, M. (eds) Scaling OpenMP for Exascale Performance and Portability. IWOMP 2017. Lecture Notes in Computer Science(), vol 10468. Springer, Cham. https://doi.org/10.1007/978-3-319-65578-9_23

Download citation

  • DOI: https://doi.org/10.1007/978-3-319-65578-9_23

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-319-65577-2

  • Online ISBN: 978-3-319-65578-9

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics