Generating Custom Code for Efficient Query Execution on Heterogeneous Processors

Processor manufacturers build increasingly specialized processors to mitigate the effects of the power wall to deliver improved performance. Currently, database engines are manually optimized for each processor: A costly and error prone process. In this paper, we propose concepts to enable the database engine to perform per-processor optimization automatically. Our core idea is to create variants of generated code and to learn a fast variant for each processor. We create variants by modifying parallelization strategies, specializing data structures, and applying different code transformations. Our experimental results show that the performance of variants may diverge up to two orders of magnitude. Therefore, we need to generate custom code for each processor to achieve peak performance. We show that our approach finds a fast custom variant for multi-core CPUs, GPUs, and MICs. Sebastian Breß DFKI GmbH and TU Berlin E-mail: [email protected] Bastian Köcher TU Berlin E-mail: [email protected] Henning Funke TU Dortmund E-mail: [email protected] Tilmann Rabl TU Berlin and DFKI GmbH E-mail: [email protected] Volker Markl TU Berlin and DFKI GmbH E-mail: [email protected] CPU Cores GPU Cores Heterogeneous Processor Chip Dedicated Accelerator (GPU, MIC) Interconnect Fig. 1 Modern processors expose heterogeneity in the form of heterogeneous cores located on the same processor chip or specialized accelerator cards.