High performance CPU/GPU multiresolution Poisson solver

被引：2

作者：

Van Rees, Wim M. ^{[1
]}

Rossinelli, Diego ^{[1
]}

Hadjidoukas, Panagiotis ^{[1
]}

Koumoutsakos, Petros ^{[1
]}

机构：

[1] Swiss Fed Inst Technol, Chair Computat Sci, Zurich, Switzerland

来源：

PARALLEL COMPUTING: ACCELERATING COMPUTATIONAL SCIENCE AND ENGINEERING (CSE) | 2014年 / 25卷

关键词：

tree-code; multipole method; GPU; vortex method; multiresolution; SIMULATIONS; FLOW;

D O I：

10.3233/978-1-61499-381-0-481

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

We present a multipole-based N-body solver for 3D multiresolution, block-structured grids. The solver is designed for a single heterogeneous CPU/GPU compute node, and evaluates the multipole expansions on the CPU while offloading the compute-heavy particle-particle interactions to the GPU. The regular structure of the destination points is exploited for data parallelism on the CPU, to reduce data transfer to the GPU and to minimize memory accesses during evaluation of the direct and indirect interactions. The algorithmic improvements together with HPC techniques lead to 81% and 96% of the upper bound performance for the CPU and GPU parts, respectively.

引用

页码：481 / 490

页数：10

共 50 条

[31] Taking Advantage of GPU/CPU Architectures for Sparse Conjugate Gradient Solver Computation
Kasmi, Najlae
Zbakh, Mostapha
Mahmoudi, Sidi Ahmed
Manneback, Pierre
PROCEEDINGS OF 2015 THIRD IEEE WORLD CONFERENCE ON COMPLEX SYSTEMS (WCCS), 2015,
[32] A hybrid CPU/GPU approach for the parallel algebraic recursive multilevel solver pARMS
Jamal, Aygul
Baboulin, Marc
Khabou, Amal
Sosonkina, Masha
PROCEEDINGS OF 2016 18TH INTERNATIONAL SYMPOSIUM ON SYMBOLIC AND NUMERIC ALGORITHMS FOR SCIENTIFIC COMPUTING (SYNASC), 2016, : 411 - 416
[33] Performance of CFD Codes on CPU/GPU Clusters
Horvath, Zoltan
Liebmann, Manfred
NUMERICAL ANALYSIS AND APPLIED MATHEMATICS, VOLS I-III, 2010, 1281 : 1789 - +
[34] A high-performance matrix–matrix multiplication methodology for CPU and GPU architectures
Vasilios Kelefouras
A. Kritikakou
Iosif Mporas
Vasilios Kolonias
The Journal of Supercomputing, 2016, 72 : 804 - 844
[35] Using high performance algorithms for the hybrid simulation of disease dynamics on CPU and GPU
Leonenko, Vasiliy N.
Pertsev, Nikolai V.
Artzrouni, Marc
INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE, ICCS 2015 COMPUTATIONAL SCIENCE AT THE GATES OF NATURE, 2015, 51 : 150 - 159
[36] High Performance Graph Analytics with Productivity on Hybrid CPU-GPU Platforms
Yang, Haoduo
Su, Huayou
Lan, Qiang
Wen, Mei
Zhang, Chunyuan
2018 2ND INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPILATION, COMPUTING AND COMMUNICATIONS (HP3C 2018), 2018, : 17 - 21
[37] Performance Analysis of PointPillars on CPU and GPU Platforms
Choi, Yuho
Kim, Byungguk
Kim, Seon Wook
2021 36TH INTERNATIONAL TECHNICAL CONFERENCE ON CIRCUITS/SYSTEMS, COMPUTERS AND COMMUNICATIONS (ITC-CSCC), 2021,
[38] Improving CPU Performance through Dynamic GPU Access Throttling in CPU-GPU Heterogeneous Processors
Rai, Siddharth
Chaudhuri, Mainak
2017 IEEE INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2017, : 18 - 29
[39] A CPU-GPU Parallel Ant Colony Optimization Solver for the Vehicle Routing Problem
Rey, Anton
Prieto, Manuel
Gomez, J., I
Tenllado, Christian
Ignacio Hidalgo, J.
APPLICATIONS OF EVOLUTIONARY COMPUTATION, EVOAPPLICATIONS 2018, 2018, 10784 : 653 - 667
[40] 2D Fast Poisson Solver for High-Performance Computing
Kalinkin, Alexander
Laevsky, Yuri M.
Gololobov, Sergey
PARALLEL COMPUTING TECHNOLOGIES, PROCEEDINGS, 2009, 5698 : 112 - +

← 1 2 3 4 5 →