Abstract
In recent years, the GPU has been integrated into the CPU processor to increase the computing power. Compared with a discrete GPU, the processors on the coupled architecture share the same memory and the data transfer via the PCIe bus is not needed. The group-by and aggregation are fundamental but time-consuming operations for a DBMS. Using hash tables is a common approach to perform a grouping. In this paper, we study how to utilize the heterogeneous CPU-GPU platform to accelerate the group-by and aggregation based on the chained hashing. We analyze the behaviors of the CPU and the coupled GPU under different workloads and propose a flexible co-processing strategy to take full advantage of the hybrid architecture. A detailed experimental study is conducted and the results demonstrate that the coupled GPU could help achieve a better overall performance for the group-by and aggregation operation.
This research was partially supported by the National Key R&D Program of China (No.2020YFC1523300).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
NVIDIA GeForce RTX 3090. https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3090/
Kaldewey, T., Lohman, G., Mueller, R., Volk, P.: GPU join processing revisited. In: DaMoN, pp. 55–62 (2012)
Yuan, Y., Lee, R., Zhang, X.: The yin and yang of processing data warehousing queries on GPU devices. PVLDB 6(10), 817–828 (2013)
Shanbhag, A., Madden, S., Yu, X.: A study of the fundamental performance characteristics of GPUs and CPUs for database analytics. In: SIGMOD, pp. 1617–1632 (2020)
AMD. http://www.amd.com/en-gb/products/processors/desktop/a-series-apu
Intel. https://ark.intel.com/content/www/br/pt/ark/products/graphics/212682/intel-uhd-graphics-750
Fang, W., He, B., Luo, Q.: Database compression on graphics processors. PVLDB 3(1), 670–680 (2010)
Heimel, M., Saecker, M., Pirk, H., Manegold, S., Markl, V.: Hardware-oblivious parallelism for in-memory column-stores. PVLDB 6(9), 709–720 (2013)
Pirk, H., Manegold, S., Kersten, M.: Waste not... efficient co-processing of relational data. In: ICDE, pp. 508–519 (2014)
Wang, K., et al.: Concurrent analytical query processing with GPUs. PVLDB 7(11), 1011–1022 (2014)
Power, J., Li, Y., Hill, D.M., Patel, M.J., Wood, A.D.: Toward GPUs being mainstream in analytic processing. In: DaMoN, pp. 11:1–11:8 (2015)
Karnagel, T., Mueller, R., Lohman, M.G.: Optimizing GPU-accelerated group-by and aggregation. In: ADMS, pp. 13–24 (2015)
Karnagel, T., Habich, D., Lehner, W.: Adaptive work placement for query processing on heterogeneous computing resources. PVLDB 10(7), 733–744 (2017)
Tome, D., Gubner, T., Raasveldt, M., Rozenberg, E., Boncz, P.: Optimizing group-by and aggregation using GPU-CPU co-processing. In: ADMS, pp. 1–10 (2018)
Rosenfeld, V., Breb, S., Zeuch, S., Rabl, T., Markl, V.: Performance analysis and automatic tuning of hash aggregation on GPUs. In: DaMoN, pp. 8:1–8:11 (2019)
Sioulas, P., Chrysogelos, P., Karpathiotakis, M., Appuswamy, R., Ailamaki, A.: Hardware-conscious hash-joins on GPUs. In: ICDE, pp. 698–709 (2019)
He, J., Lu, M., He, B.: Revisiting co-processing for hash joins on the coupled CPU-GPU architecture. PVLDB 6(10), 889–900 (2013)
He, J., Zhang, S., He, B.: In-cache query co-processing on coupled CPU-GPU architectures. PVLDB 8(4), 329–340 (2014)
Luan, H., Chang, L.: An evaluation of analytical queries on CPUs and coupled GPUs. Concurrency Comput. Pract. Experience 29(5), e3982 (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Luan, H., Fu, Y. (2022). Accelerating Group-By and Aggregation on Heterogeneous CPU-GPU Platforms. In: Xie, Q., Zhao, L., Li, K., Yadav, A., Wang, L. (eds) Advances in Natural Computation, Fuzzy Systems and Knowledge Discovery. ICNC-FSKD 2021. Lecture Notes on Data Engineering and Communications Technologies, vol 89. Springer, Cham. https://doi.org/10.1007/978-3-030-89698-0_100
Download citation
DOI: https://doi.org/10.1007/978-3-030-89698-0_100
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-89697-3
Online ISBN: 978-3-030-89698-0
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)