Abstract
Lexical sparse retrievers, rely on efficient searching algorithms that operate over inverted index structures, tailored specifically for CPU. This CPU-centric design poses a challenge when adapting these algorithms for highly parallel accelerators, such as GPUs, thus deterring potential performance gains. To address this, we propose to leverage the recent advances in sparse computations offered by deep learning frameworks to directly implementing sparse retrievals on these accelerators.
This paper presents the SPARe (SPArse Retrievers) Python package, which provides a high-level API to deal with sparse retrievers on (single or multi)-accelerators, by leveraging deep learning frameworks at its core.
Experimental results show that SPARe, running on an accessible GPU (RTX 2070), can calculate the BM25 scores for close to 9 million MSMARCO documents at a rate of 800 questions per second with our specialized algorithm. Notably, SPARe proves highly effective for denser LSR indexes, significantly surpassing the performance of established systems such as PISA, Pyserini and PyTerrier.
SPARe is publicly available at https://github.com/ieeta-pt/SPARe.
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
References
Bell, N., Garland, M.: Efficient sparse matrix-vector multiplication on CUDA. NVIDIA Technical Report NVR-2008-004, NVIDIA Corporation, December 2008
Broder, A., Carmel, D., Herscovici, M., Soffer, A., Zien, J.: Efficient query evaluation using a two-level retrieval process, pp. 426–434, November 2003. https://doi.org/10.1145/956863.956944
Dai, Z., Callan, J.: Context-aware term weighting for first stage passage retrieval. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2020, New York, NY, USA, pp. 1533–1536. Association for Computing Machinery (2020). https://doi.org/10.1145/3397271.3401204
Ding, S., He, J., Yan, H., Suel, T.: Using graphics processors for high performance IR query processing. In: Proceedings of the 18th International Conference on World Wide Web, WWW 2009, New York, NY, USA, pp. 421–430. Association for Computing Machinery (2009). https://doi.org/10.1145/1526709.1526766
Formal, T., Lassance, C., Piwowarski, B., Clinchant, S.: From distillation to hard negative sampling: making sparse neural IR models more effective. In: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2022, New York, NY, USA, pp. 2353–2359. Association for Computing Machinery (2022). https://doi.org/10.1145/3477495.3531857
Formal, T., Piwowarski, B., Clinchant, S.: SPLADE: sparse lexical and expansion model for first stage ranking. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2021, New York, NY, USA, pp. 2288–2292. Association for Computing Machinery (2021). https://doi.org/10.1145/3404835.3463098
Gaioso, R., Costa, V., Guardia, H., Senger, H.: Performance evaluation of single vs. batch of queries on GPUs. Concurr. Comput. Practic. Exp. 32, August 2019. https://doi.org/10.1002/cpe.5474
Gaioso, R., Gil-Costa, V., Guardia, H., Senger, H.: A parallel implementation of wand on GPUs. In: 2018 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP), pp. 10–17 (2018). https://doi.org/10.1109/PDP2018.2018.00011
Lin, J.: A proposed conceptual framework for a representational approach to information retrieval. CoRR abs/2110.01529 (2021). https://arxiv.org/abs/2110.01529
Lin, J., Ma, X., Lin, S.C., Yang, J.H., Pradeep, R., Nogueira, R.: Pyserini: a Python toolkit for reproducible information retrieval research with sparse and dense representations. In: Proceedings of the 44th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2021), pp. 2356–2362 (2021)
MacAvaney, S., Macdonald, C.: A Python interface to PISA! In: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2022, New York, NY, USA, pp. 3339–3344. Association for Computing Machinery (2022). https://doi.org/10.1145/3477495.3531656
Macdonald, C., MacAvaney, S.: PyT-SPLADE: PyTerrier plugin for SPLADE models (2023). https://github.com/cmacdonald/pyt_splade
Macdonald, C., Tonellotto, N.: Declarative experimentation information retrieval using PyTerrier. In: Proceedings of ICTIR 2020 (2020)
Mackenzie, J., Trotman, A., Lin, J.: Wacky weights in learned sparse representations and the revenge of score-at-a-time query evaluation (2021)
Mallia, A., Khattab, O., Suel, T., Tonellotto, N.: Learning passage impacts for inverted indexes. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2021, New York, NY, USA, pp. 1723–1727. Association for Computing Machinery (2021). https://doi.org/10.1145/3404835.3463030
Mallia, A., Siedlaczek, M., Mackenzie, J., Suel, T.: PISA: performant indexes and search for academia. In: Proceedings of the Open-Source IR Replicability Challenge Co-located with 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, OSIRRC@SIGIR 2019, Paris, France, 25 July 2019, pp. 50–56 (2019). http://ceur-ws.org/Vol-2409/docker08.pdf
Robertson, S., Zaragoza, H.: The probabilistic relevance framework: BM25 and beyond. Found. Trends Inf. Retr. 3(4), 333–389 (2009). https://doi.org/10.1561/1500000019. https://www.nowpublishers.com/article/Details/INR-019
Thakur, N., Reimers, N., Rücklé, A., Srivastava, A., Gurevych, I.: BEIR: a heterogeneous benchmark for zero-shot evaluation of information retrieval models. In: Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2) (2021). https://openreview.net/forum?id=wCu6T5xFjeJ
Acknowledgments
This work was funded by the Foundation for Science and Technology (FCT) in the context of project UIDB/00127/2020. The work was produced with the support of HPC Universidade de Évora with funds from the Advanced Computing Project FCT.CPCA.2022.01. Tiago Almeida is funded by the grant 2020.05784.BD.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Almeida, T., Matos, S. (2024). SPARe: Supercharged Lexical Retrievers on GPU with Sparse Kernels. In: Goharian, N., et al. Advances in Information Retrieval. ECIR 2024. Lecture Notes in Computer Science, vol 14610. Springer, Cham. https://doi.org/10.1007/978-3-031-56063-7_33
Download citation
DOI: https://doi.org/10.1007/978-3-031-56063-7_33
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56062-0
Online ISBN: 978-3-031-56063-7
eBook Packages: Computer ScienceComputer Science (R0)