Traditionally, query engines are optimized for CPUs, but nowadays modern servers are becoming increasingly heterogeneous and equipped with multiple hardware accelerators, like GPUs. In this line of work, we investigate how different accelerators can be used by the query engine to increase its performance as well as provide isolation between queries. We design new hardware-conscious algorithms, study how existing ones perform across different micro-architectures and investigate multi-device query execution. Lastly, we provide engine designs that generalize device-specific approaches to achieve efficient heterogeneous-device execution through just-in-time code generation.
P. Chrysogelos; M. Karpathiotakis; R. Appuswamy; A. Ailamaki : HetExchange: Encapsulating heterogeneous CPU-GPU parallelism in JIT compiled engines. 2019. 45th International Conference on Very Large Data Bases, Los Angeles, California, USA, August 26-30, 2019.
P. Sioulas; P. Chrysogelos; M. Karpathiotakis; R. Appuswamy; A. Ailamaki : Hardware-conscious Hash-Joins on GPUs. 2019. IEEE International Conference on Data Engineering, Macau SAR, China, April 8-12, 2019.
P. Chrysogelos; P. Sioulas; A. Ailamaki : Hardware-conscious Query Processing in GPU-accelerated Analytical Engines. 2019. 9th Biennial Conference on Innovative Data Systems Research, Asilomar, California, USA, January 13-16, 2019.
E. Tzirita Zacharatou; H. Doraiswamy; A. Ailamaki; C. Silva; J. Freire : GPU Rasterization for Real-Time Spatial Aggregation over Arbitrary Polygons. 2017. The 44th International Conference on Very Large Data Bases, Rio de Janeiro, Brazil, August 27-31, 2018. DOI : 10.14778/3157794.3157803.
R. Appuswamy; M. Karpathiotakis; D. Porobic; A. Ailamaki : The Case For Heterogeneous HTAP. 2017. 8th Biennial Conference on Innovative Data Systems Research, Chaminade, California, January 8-11,2017.