A Hybrid Circular Queue Method for Iterative Stencil Computations on GPUs

In this paper, we present a hybrid circular queue method that can significantly boost the performance of stencil computations on GPU by carefully balancing usage of registers and shared-memory. Unlike earlier methods that rely on circular queues predominantly implemented using indirectly addressable...

Full description

Saved in:
Bibliographic Details
Published inJournal of computer science and technology Vol. 27; no. 1; pp. 57 - 74
Main Author 杨杨 崔慧敏 冯晓兵 薛京灵
Format Journal Article
LanguageEnglish
Published Boston Springer US 2012
Springer Nature B.V
Subjects
Algorithms
Analysis
Artificial Intelligence
Balancing
Central processing units
Communication
Computation
Computer memory
Computer Science
Computers
CPUs
Data Structures and Information Theory
Decomposition
Exploitation
Graphics boards
Information Systems Applications (incl.Internet)
Iterative methods
Placement
Platforms
Queues
Queuing
R&D
Registers
Regular Paper
Research & development
Reuse
Software Engineering
Stencils
Studies
Theory of Computation
共享内存
内存共享
图形模板
循环队列
数据寄存器
混合方法
计算
迭代方法
circular queue
occupancy
GPU
stencil computation
register
Online AccessGet full text

Cover

More Information
Summary:In this paper, we present a hybrid circular queue method that can significantly boost the performance of stencil computations on GPU by carefully balancing usage of registers and shared-memory. Unlike earlier methods that rely on circular queues predominantly implemented using indirectly addressable shared memory, our hybrid method exploits a new reuse pattern spanning across the multiple time steps in stencil computations so that circular queues can be implemented by both shared memory and registers effectively in a balanced manner. We describe a framework that automatically finds the best placement of data in registers and shared memory in order to maximize the performance of stencil computations. Validation using four different types of stencils on three different GPU platforms shows that our hybrid method achieves speedups up to 2.93X over methods that use circular queues implemented with shared-memory only.
Bibliography:stencil computation, circular queue, GPU, occupancy, register
In this paper, we present a hybrid circular queue method that can significantly boost the performance of stencil computations on GPU by carefully balancing usage of registers and shared-memory. Unlike earlier methods that rely on circular queues predominantly implemented using indirectly addressable shared memory, our hybrid method exploits a new reuse pattern spanning across the multiple time steps in stencil computations so that circular queues can be implemented by both shared memory and registers effectively in a balanced manner. We describe a framework that automatically finds the best placement of data in registers and shared memory in order to maximize the performance of stencil computations. Validation using four different types of stencils on three different GPU platforms shows that our hybrid method achieves speedups up to 2.93X over methods that use circular queues implemented with shared-memory only.
11-2296/TP
Yang Yang, Hui-Min Cui, Xiao-Bing Feng, Jing-Ling Xue（ 1.State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences Beijing 100190, China ;2.Graduate University of Chinese Academy of Sciences, Beijing 100190, China; 3.programming Languages and Compilers Group, School of Computer Science and Engineering University of New South Wales, Sydney, NSW 2052, Australia）
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ISSN:1000-9000
1860-4749
DOI:10.1007/s11390-012-1206-3