Bigger buffer k-d trees on multi-many-core systems

Publikation: Bidrag til bog/antologi/rapportBidrag til bog/antologi

A buffer k -d tree is a k -d tree variant for massively-parallel nearest neighbor search. While providing valuable speed-ups on modern many-core devices in case both a large number of reference and query points are given, buffer k -d trees are limited by the amount of points that can fit on a single device. In this work, we show how to modify the original data structure and the associated workflow to make the overall approach capable of dealing with massive data sets. We further provide a simple yet efficient way of using multiple devices given in a single workstation. The applicability of the modified framework is demonstrated in the context of astronomy, a field that is faced with huge amounts of data.
OriginalsprogEngelsk
TitelHigh Performance Computing for Computational Science – VECPAR 2018 : 13th International Conference, São Pedro, Brazil, September 17–19, 2018, Revised Selected Papers
ForlagSpringer
Publikationsdato2019
Sider202-214
ISBN (Trykt)978-3-030-15995-5
ISBN (Elektronisk)978-3-030-15996-2
DOI
StatusUdgivet - 2019
Begivenhed13th International Conference on High Performance Computing in Computational Science - São Pedro, Brasilien
Varighed: 17 sep. 201819 sep. 2018

Konference

Konference13th International Conference on High Performance Computing in Computational Science
LandBrasilien
BySão Pedro
Periode17/09/201819/09/2018
NavnLecture Notes in Computer Science
Vol/bind11333
ISSN0302-9743

Links

ID: 166494866