2018-06-11
Make the most out of your SIMD investments: Counter control flow divergence in compiled query pipelines
Publication
Publication
Presented at the
International Workshop on Data Management on New Hardware (June 2018), Houston, TX, USA
Increasing single instruction multiple data (SIMD) capabilities in modern hardware allows for compiling efficient data-parallel query pipelines. This means GPU-alike challenges arise: control flow divergence causes underutilization of vector-processing units. In this paper, we present efficient algorithms for the AVX-512 architecture to address this issue. These algorithms allow for fine-grained assignment of new tuples to idle SIMD lanes. Furthermore, we present strategies for their integration with compiled query pipelines without introducing inefficient memory materializations. We evaluate our approach with a high-performance geospatial join query, which shows performance improvements of up to 35%.
Additional Metadata | |
---|---|
doi.org/10.1145/3211922.3211928 | |
International Workshop on Data Management on New Hardware | |
Organisation | Centrum Wiskunde & Informatica, Amsterdam (CWI), The Netherlands |
Lang, H, Kipf, A, Passing, L.K, Boncz, P.A, Neumann, T, & Kemper, A. (2018). Make the most out of your SIMD investments: Counter control flow divergence in compiled query pipelines. In 14th International Workshop on Data Management on New Hardware, DaMoN 2018. doi:10.1145/3211922.3211928
|