DASHMM

The Dynamic Adaptive System for Hierarchical Multipole Methods (DASHMM) is a flexible, extensible library implementing multipole methods using a dynamic and adaptive computational approach. DASHMM is designed to easily allow users to create tools usable in production-level end science that display significantly improved strong scaling and efficiency for these applications.

More details about the project can be found under Details. DASHMM can be applied to a number of application domains. To start learning how to use DASHMM, please visit User Guides and Tutorials. To download the latest version of DASHMM, please visit Get DASHMM, or for prior versions please visit Resources. To learn more about the DASHMM team, visit About DASHMM Team.

News

  • June 2nd, 2017 - DASHMM performance evaluation paper published in Proceedings of the 18th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC2017) Paper (377.25 KiB, 0 downloads) Presentation (2.77 MiB, 0 downloads)
  • April 11th, 2017 - DASHMM 1.2.0 is released.
  • January 24th, 2017 - 2017 Summer REU annouced.
  • January 20th, 2017 - DASHMM 1.1.0 is released.
  • November 2nd, 2016 - DASHMM 1.0.0 is released.
  • September 30th, 2016 - First DASHMM Code paper published in Commun. Comput. Phys., 2016, 20, pp 1106-1126. Code paper
  • April 25th, 2016 - DASHMM version 0.6 is released.
  • January 29th, 2016 - DASHMM version 0.5 is released.
  • REU program announced for Summer 2015
  • April 12th, 2015 - Presentation (652.83 KiB, 0 downloads) about DASHMM project at APS April Meeting 2015 in Baltimore, MD.
  • October 1st, 2014 - The DASHMM project begins.

Acknowledgements

Funding for the DASHMM project is provided by the National Science Foundation.