Please use this identifier to cite or link to this item:
Type: Artigo
Title: Toast: Automatic Tiling For Iterative Stencil Computations On Gpus
Author: Rocha
Rodrigo C. O.; Pereira
Alyson D.; Ramos
Luiz; Goes
Luis F. W.
Abstract: The stencil pattern is important in many scientific and engineering domains, spurring great interest from researchers and industry. In recent years, various optimizations have been proposed for parallel stencil applications running on graphics processing units (GPUs). In particular, tiling is a technique that can significantly enhance application performance by improving data locality and by reducing the volume of communication between host memory and GPU. In addition, tiling enables stencil applications to process inputs that are larger than the physical GPU memory. However, implementing tiling efficiently is complex, time-consuming, and error-prone. In this paper, we propose transparently optimized automatic stencil tiling (TOAST), an automatic tiling mechanism for iterative stencil computations running on GPUs; TOAST has 3 main benefits: (1) It incorporates an optimization model that seeks to maximize data reuse within tiles while respecting the amount of dynamically available GPU memory; (2) it offers a virtualized GPU memory for stencil computations, allowing for large input data; and (3) it performs optimal tiling transparently to the developer of the parallel stencil application. The current implementation of TOAST augments the PSkel framework with an internal solver based on genetic algorithms. Our experimental results show that TOAST improves the performance of iterative stencil applications by up to 13 x compared with their multithreaded (central processing unit-based) optimized versions and up to 48 x compared with a naive tiling approach on GPU. The TOAST mechanism is able to automatically achieve a low percentual overhead of data management compared with actual stencil computation.
Subject: Autotuning
Optimization Model
Parallel Skeletons
Stencil Computation
Editor: Wiley-Blackwell
Citation: Concurrency And Computation-practice & Experience. Wiley-blackwell, v. 29, p. , 2017.
Rights: fechado
Identifier DOI: 10.1002/cpe.4053
Date Issue: 2017
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
File SizeFormat 
000398717400011.pdf2.05 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.