Database Systems on GPUs

In recent years, the growth of the amount of data required to be analyzed by computing systems has outstripped the performance of the hardware CPUs, resulting in the hardware becoming a potential bottleneck. The Graphical Processing Units (GPU), however, has become a powerful, programmable, many-core processor in the past decade that turns out to be an effective means to improve the performance of main memory databases. In this monograph, the authors provide background on GPU as a parallel architecture and the traditional parallel query processing in main memory databases. GPU-optimized system designs are then covered before surveying a series of commercial and research systems. Database Systems on GPUs provides a thorough introduction to GPU optimized database systems especially in online analytical processing (OLAP). It can be used as a short text for graduate level students or a survey for researchers. The authors emphasis is on the breadth of the subject and cover as many publications as possible, with necessary details in some key GPU-optimized designs.