To use or not to use the SIMD gather instruction?

Abstract

The Single Instruction Multiple Data (SIMD) paradigm became a core principle for optimizing query operators. Until now, only the LOAD/STORE instructions are considered to be efficient enough to achieve the expected speedups, while avoiding GATHER/SCATTER is considered almost imperative. However, the GATHER instruction offers a very flexible way to populate SIMD registers with data elements coming from non-consecutive memory locations. As we will show within the paper, the GATHER instruction can achieve the same performance as the LOAD instruction, if applied properly. To enable the proper usage, we outline a novel access pattern which then allows fine-grained, partition-based SIMD implementations using the GATHER instruction.