Brief Announcement: Faster Stencil Computations using Gaussian Approximations

被引：3

作者：

Ahmad, Zafar ^{[1
]}

Chowdhury, Rezaul ^{[1
]}

Das, Rathish ^{[2
]}

Ganapathi, Pramod ^{[1
]}

Gregory, Aaron ^{[1
]}

Zhu, Yimin ^{[1
]}

机构：

[1] SUNY Stony Brook, Stony Brook, NY 11794 USA

[2] Univ Waterloo, Waterloo, ON N2L 3G1, Canada

来源：

PROCEEDINGS OF THE 34TH ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES, SPAA 2022 | 2022年

基金：

加拿大自然科学与工程研究理事会;

关键词：

Linear Stencil; Gaussian Approximation; Fast Gauss Transform; FINITE-DIFFERENCE;

D O I：

10.1145/3490148.3538558

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Stencil computations are widely used to simulate the change of state of physical systems. The current best algorithm for performing aperiodic linear stencil computations on a d(>= 1)-dimensional grid of size N for T timesteps does (Theta) over tilde (TN1-1/d + N log N) work. We introduce novel techniques based on random walks and Gaussian approximations for an asymptotic improvement of this work bound for a class of linear stencils. We also improve the span (i.e., parallel running time on an unbounded number of processors) asymptotically from the current state of the art.

引用

页码：291 / 293

页数：3

共 50 条

[31] Scalable communication for high-order stencil computations using CUDA-aware MPI
Pekkila, Johannes
Vaisala, Miikka S.
Kapyla, Maarit J.
Rheinhardt, Matthias
Lappi, Oskar
PARALLEL COMPUTING, 2022, 111
[32] Brief Announcement: Node Sampling Using Centrifugal Random Walks
Sevillal, Andres
Mozo, Alberto
Fernandez Anta, Antonio
DISTRIBUTED COMPUTING, DISC 2012, 2012, 7611 : 419 - +
[33] Brief Announcement: COP Composition Using Transaction Suspension in the Compiler
Avni, Hillel
Suissa-Peleg, Adi
DISTRIBUTED COMPUTING (DISC 2014), 2014, 8784 : 550 - 552
[34] Toward Faster Reinforcement Learning for Robotics: Using Gaussian Processes
Younes, Ali
Panov, Aleksandr I.
ARTIFICIAL INTELLIGENCE, 2019, 11866 : 160 - 174
[35] High-Level Programming of Stencil Computations on Multi-GPU Systems Using the SkelCL Library
Steuwer, Michel
Haidl, Michael
Breuer, Stefan
Gorlatch, Sergei
PARALLEL PROCESSING LETTERS, 2014, 24 (03)
[36] Scalable Bayesian Optimization Using Vecchia Approximations of Gaussian Processes
Jimenez, Felix
Katzfuss, Matthias
INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 206, 2023, 206
[37] Brief Announcement: Sublinear-Space Distance Labeling Using Hubs
Gawrychowski, Pawel
Kosowski, Adrian
Uznanski, Przemyslaw
PROCEEDINGS OF THE 2016 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING (PODC'16), 2016, : 43 - 45
[38] Brief Announcement: Optimal Bit-Reversal Using Vector Permutations
Lokhmotov, Anton
Mycroft, Alan
SPAA'07: PROCEEDINGS OF THE NINETEENTH ANNUAL SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES, 2007, : 198 - 199
[39] Toward Faster Reinforcement Learning for Robotics Applications by Using Gaussian Processes
Younes, A.
Yushchenko, A. S.
XLIII ACADEMIC SPACE CONFERENCE, DEDICATED TO THE MEMORY OF ACADEMICIAN S P KOROLEV AND OTHER OUTSTANDING RUSSIAN SCIENTISTS - PIONEERS OF SPACE EXPLORATION, 2019, 2171
[40] Energy-efficient Stencil Computations on Distributed GPUs using Dynamic Parallelism and GPU-controlled Communication
Oden, Lena
Klenk, Benjamin
Froening, Holger
2014 ENERGY EFFICIENT SUPERCOMPUTING WORKSHOP (E2SC), 2014, : 31 - 40

← 1 2 3 4 5 →