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. This paper investigates new ways to build such high-performance IR systems based on Graphical Processing Units (GPUs). GPUs were originally designed to accelerate computer graphics applications through massive on-chip parallelism. Recently a number of researchers have studied how to use GPUs for other problem domains including databases and scientific computing, but they are not aware of previous attempts to use GPUs for large-scale web search.