Using Graphics Processors for High-Performance IR Query Processing
Web search engines are facing formidable performance challenges as they need to process thousands of queries per second over billions of documents. To deal with this heavy workload, current engines use massively parallel architectures of thousands of machines that require large hardware investments. We investigate new ways to build such high-performance systems based on Graphical Processing Units (GPUs). GPUs were originally designed to accelerate computer graphics applications through massive on-chip parallelism. Recently number of researchers have studied how to use GPUs for other problem domains including databases and scienti.c computing 2, 3, 5, but we are not aware of previous attempts use GPUs for large-scale web search. Our contribution here is to design a basic system architecture for GPU-based highperformance IR, and to describe how to perform highly e cient query processing within such an architecture. Preliminary experimental results based on a prototype implementation suggest that signi.cant gains in query processing performance might be obtainable with such an approach.
Web search query processing GPU
Shuai Ding Jinru He Hao Yan Torsten Suel
CIS Department, Polytechnic University Brooklyn, NY, 11201, USA
国际会议
第十七届国际万维网大会(the 17th International World Wide Web Conference)(WWW08)
北京
英文
2008-04-21（万方平台首次上网日期，不代表论文的发表时间）