Learned Indexes and Database Reliability: Performance with Explicit Failure Bounds

Wesley Freeman

Abstract

This review examines learned indexes in reliable database systems. The organizing question is when learned data structures improve performance without weakening correctness, predictability, and maintainability. Ten related scholarly sources are synthesized through a decision-centered framework spanning problem definition, mechanism, measurement, evaluation, implementation, and governance. The review does not invent experiments, pooled estimates, or unreported quantitative results. It instead evaluates the strength and transferability of the available evidence, with particular attention to generalizing speedups from stationary key distributions to evolving production data. The resulting framework links technical or empirical performance to explicit use conditions and identifies tests that should precede wider adoption in data-management systems with strict correctness requirements.

Keywords

learned indexesdatabasesreliabilityperformancedata structures

References

Antol, M., Ol’ha, J., Slanináková, T., & Dohnal, V. (2021). Learned Metric Index — Proposition of learned indexing for unstructured data. Information Systems, 100, 101774. https://doi.org/10.1016/j.is.2021.101774 DOI
Ge, J., Zhang, H., Shi, B., Luo, Y., Guo, Y., Chai, Y., Chen, Y., & Pan, A. (2023). SALI: A Scalable Adaptive Learned Index Framework based on Probability Models. Proceedings of the ACM on Management of Data, 1(4), 1-25. https://doi.org/10.1145/3626752 DOI
Hébert, J. R., Shivappa, N., Wirth, M. D., Hussey, J. R., & Hurley, T. G. (2018). Perspective: The Dietary Inflammatory Index (DII)—Lessons Learned, Improvements Made, and Future Directions. Advances in Nutrition, 10(2), 185-195. https://doi.org/10.1093/advances/nmy071 DOI
Kipf, A., Marcus, R., van Renen, A., Stoian, M., Kemper, A., Kraska, T., & Neumann, T. (2020). RadixSpline: A Single-Pass Learned Index. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2004.14541 DOI
Li, P., Lu, H., Zhu, R., Ding, B., Yang, L., & Pan, G. (2023). DILI: A Distribution-Driven Learned Index. Proceedings of the VLDB Endowment, 16(9), 2212-2224. https://doi.org/10.14778/3598581.3598593 DOI
Shi, J., Cong, G., & Li, X. L. (2022). Learned Index Benefits. Proceedings of the VLDB Endowment, 15(13), 3950-3962. https://doi.org/10.14778/3565838.3565848 DOI
Sun, Z., Zhou, X., & Li, G. (2023). Learned Index: A Comprehensive Experimental Evaluation. Proceedings of the VLDB Endowment, 16(8), 1992-2004. https://doi.org/10.14778/3594512.3594528 DOI
Tian, Y., Yan, T., Zhao, X., Huang, K., & Zhou, X. (2022). A Learned Index for Exact Similarity Search in Metric Spaces. IEEE Transactions on Knowledge and Data Engineering, 35(8), 7624-7638. https://doi.org/10.1109/tkde.2022.3206441 DOI
Wongkham, C., Lu, B., Liu, C., Zhong, Z., Lo, E., & Wang, T. (2022). Are updatable learned indexes ready?. Proceedings of the VLDB Endowment, 15(11), 3004-3017. https://doi.org/10.14778/3551793.3551848 DOI
Wu, J., Zhang, Y., Chen, S., Wang, J., Chen, Y., & Xing, C. (2021). Updatable learned index with precise positions. Proceedings of the VLDB Endowment, 14(8), 1276-1288. https://doi.org/10.14778/3457390.3457393 DOI

Publication details

Journal: Systems, Networks and Secure Computing
Volume: 1 (2026)
Article number: snsc20260004
License: CC BY 4.0

[1] Antol, M., Ol’ha, J., Slanináková, T., & Dohnal, V. (2021). Learned Metric Index — Proposition of learned indexing for unstructured data. Information Systems, 100, 101774. https://doi.org/10.1016/j.is.2021.101774 DOI

[2] Ge, J., Zhang, H., Shi, B., Luo, Y., Guo, Y., Chai, Y., Chen, Y., & Pan, A. (2023). SALI: A Scalable Adaptive Learned Index Framework based on Probability Models. Proceedings of the ACM on Management of Data, 1(4), 1-25. https://doi.org/10.1145/3626752 DOI

[3] Hébert, J. R., Shivappa, N., Wirth, M. D., Hussey, J. R., & Hurley, T. G. (2018). Perspective: The Dietary Inflammatory Index (DII)—Lessons Learned, Improvements Made, and Future Directions. Advances in Nutrition, 10(2), 185-195. https://doi.org/10.1093/advances/nmy071 DOI

[4] Kipf, A., Marcus, R., van Renen, A., Stoian, M., Kemper, A., Kraska, T., & Neumann, T. (2020). RadixSpline: A Single-Pass Learned Index. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2004.14541 DOI

[5] Li, P., Lu, H., Zhu, R., Ding, B., Yang, L., & Pan, G. (2023). DILI: A Distribution-Driven Learned Index. Proceedings of the VLDB Endowment, 16(9), 2212-2224. https://doi.org/10.14778/3598581.3598593 DOI

[6] Shi, J., Cong, G., & Li, X. L. (2022). Learned Index Benefits. Proceedings of the VLDB Endowment, 15(13), 3950-3962. https://doi.org/10.14778/3565838.3565848 DOI

[7] Sun, Z., Zhou, X., & Li, G. (2023). Learned Index: A Comprehensive Experimental Evaluation. Proceedings of the VLDB Endowment, 16(8), 1992-2004. https://doi.org/10.14778/3594512.3594528 DOI

[8] Tian, Y., Yan, T., Zhao, X., Huang, K., & Zhou, X. (2022). A Learned Index for Exact Similarity Search in Metric Spaces. IEEE Transactions on Knowledge and Data Engineering, 35(8), 7624-7638. https://doi.org/10.1109/tkde.2022.3206441 DOI

[9] Wongkham, C., Lu, B., Liu, C., Zhong, Z., Lo, E., & Wang, T. (2022). Are updatable learned indexes ready?. Proceedings of the VLDB Endowment, 15(11), 3004-3017. https://doi.org/10.14778/3551793.3551848 DOI

[10] Wu, J., Zhang, Y., Chen, S., Wang, J., Chen, Y., & Xing, C. (2021). Updatable learned index with precise positions. Proceedings of the VLDB Endowment, 14(8), 1276-1288. https://doi.org/10.14778/3457390.3457393 DOI