Towards a Hybrid Design for Fast Query Processing in DB2 with BLU Acceleration Using Graphical Processing Units: A Technology Demonstration

摘要: In this paper, we show how use Nvidia GPUs and host CPU cores for faster query processing in a DB2 database using BLU Acceleration (DB2's column store technology). Moreover, the benefits problems of hardware accelerators (more specifically GPUs) real commercial Relational Database Management System(RDBMS).We investigate effect off-loading specific operations to GPU, doing so results significant performance improvement. We then demonstrate that some queries, just perform entire operation is more beneficial. While Nvidia's fast kernels like sort, have also developed our own high such as group by aggregation. Finally, dynamic design can make optimizer metadata intelligently choose GPU kernel run. For first time literature, benchmarks representative customer environments gauge prototype, which get speed increase upwards 2x, realistic set queries.