DASH: Data Structures and Algorithms with Support for Hierarchical Locality

Research output: Contribution to book/conference proceedings/anthology/reportConference contributionContributedpeer-review

Contributors

Abstract

DASH is a realization of the PGAS (partitioned global address space) model in the form of a C++ template library. Operator overloading is used to provide global-view PGAS semantics without the need for a custom PGAS (pre-) compiler. The DASH library is implemented on top of our runtime system DART, which provides an abstraction layer on top of existing one-sided communication substrates. DART contains methods to allocate memory in the global address space as well as collective and one-sided communication primitives. To support the development of applications that exploit a hierarchical organization, either on the algorithmic or on the hardware level, DASH features the notion of teams that are arranged in a hierarchy. Based on a team hierarchy, the DASH data structures support locality iterators as a generalization of the conventional local/global distinction found in many PGAS approaches.

Details

Original languageEnglish
Title of host publicationEuro-Par 2014: Parallel Processing Workshops
PublisherSpringer, Cham
Pages542-552
Number of pages11
ISBN (print)978-3-319-14312-5
Publication statusPublished - 2014
Peer-reviewedYes

Publication series

SeriesLecture Notes in Computer Science, Volume 8806
ISSN0302-9743

Conference

Title20th Euro-Par International Workshops
Duration25 - 26 August 2014
CityPorto
CountryPortugal

External IDs

WOS 000354785000046
Scopus 84920100746

Keywords

Keywords

  • High Performance Computing, Template Library, Distribute Memory Machine, Data Container, Global Address Space

Library keywords