Thread affinity mapping for irregular data access on shared cache GPGPU

Hsien Kai Kuo*, Kuan Ting Chen, Bo-Cheng Lai, Jing Yang Jou

*Corresponding author for this work

研究成果: Conference contribution同行評審

10 引文 斯高帕斯(Scopus)

摘要

Memory Coalescing and on-chip shared Cache are two effective techniques to alleviate the memory bottleneck in modern GPGPUs. These two techniques are very useful on applications with regular memory accesses. However, they become ineffective on concurrent threads with large numbers of uncoordinated accesses and the potential performance benefit could be significantly degraded. This paper proposes a thread affinity mapping methodology to coordinate the irregular data accesses on shared cache GPGPUs. Based on the proposed affinity metrics, threads are congregated into execution groups which are able to fully exploit the memory coalescing and data sharing within an application. An average of 3.5x runtime speedup is achieved on a Fermi GPGPU. The speedup scales with the sizes of test cases, which makes the proposed methodology an effective and promising solution for the continually increasing complexities of applications in the future many-core systems.

原文English
主出版物標題ASP-DAC 2012 - 17th Asia and South Pacific Design Automation Conference
頁面659-664
頁數6
DOIs
出版狀態Published - 26 四月 2012
事件17th Asia and South Pacific Design Automation Conference, ASP-DAC 2012 - Sydney, NSW, Australia
持續時間: 30 一月 20122 二月 2012

出版系列

名字Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC

Conference

Conference17th Asia and South Pacific Design Automation Conference, ASP-DAC 2012
國家Australia
城市Sydney, NSW
期間30/01/122/02/12

指紋 深入研究「Thread affinity mapping for irregular data access on shared cache GPGPU」主題。共同形成了獨特的指紋。

引用此