# Accelerating String-Key Learned Index Structures via Memoization-based Incremental Training

arXiv: https://arxiv.org/abs/2403.11472 · HTML: https://mlsanity.com/paper/2403.11472/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 79 references that are in the map.
- First posted on arXiv 18 Mar 2024.

## Builds on (most cited first, 24 of 24)

- [The Case for Learned Index Structures](https://mlsanity.com/paper/1712.01208/) (2017)
- [Learning to Optimize Tensor Programs](https://mlsanity.com/paper/1805.08166/) (2018)
- [ALEX: An Updatable Adaptive Learned Index](https://mlsanity.com/paper/1905.08898/) (2019)
- [RadixSpline: A Single-Pass Learned Index](https://mlsanity.com/paper/2004.14541/) (2020)
- [FITing-Tree: A Data-aware Index Structure](https://mlsanity.com/paper/1801.10207/) (2018)
- [Learning Multi-dimensional Indexes](https://mlsanity.com/paper/1912.01668/) (2019)
- [Benchmarking Learned Indexes](https://mlsanity.com/paper/2006.12804/) (2020)
- [Tsunami: A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads](https://mlsanity.com/paper/2006.13282/) (2020)
- [The PGM-index: a multicriteria, compressed and learned approach to data indexing](https://mlsanity.com/paper/1910.06169/) (2019)
- [Updatable Learned Index with Precise Positions](https://mlsanity.com/paper/2104.05520/) (2021)
- [Learned Indexes for a Google-scale Disk-based Database](https://mlsanity.com/paper/2012.12501/) (2020)
- [In-RDBMS Hardware Acceleration of Advanced Analytics](https://mlsanity.com/paper/1801.06027/) (2018)
- [CARMI: A Cache-Aware Learned Index with a Cost-based Construction Algorithm](https://mlsanity.com/paper/2103.00858/) (2021)
- [LIDER: An Efficient High-dimensional Learned Index for Large-scale Dense Passage Retrieval](https://mlsanity.com/paper/2205.00970/) (2022)
- [APEX: A High-Performance Learned Index on Persistent Memory](https://mlsanity.com/paper/2105.00683/) (2021)
- [Bounding the Last Mile: Efficient Learned String Indexing](https://mlsanity.com/paper/2111.14905/) (2021)
- [WISK: A Workload-aware Learned Index for Spatial Keyword Queries](https://mlsanity.com/paper/2302.14287/) (2023)
- [Updatable Learned Indexes Meet Disk-Resident DBMS -- From Evaluations to Design Choices](https://mlsanity.com/paper/2305.01237/) (2023)
- [Towards Practical Learned Indexing](https://mlsanity.com/paper/2108.05117/) (2021)
- [NFL: Robust Learned Index via Distribution Transformation](https://mlsanity.com/paper/2205.11807/) (2022)
- [LMSFC: A Novel Multidimensional Index based on Learned Monotonic Space Filling Curves](https://mlsanity.com/paper/2304.12635/) (2023)
- [Micro-architectural Analysis of a Learned Index](https://mlsanity.com/paper/2109.08495/) (2021)
- [RECIPE : Converting Concurrent DRAM Indexes to Persistent-Memory Indexes](https://mlsanity.com/paper/1909.13670/) (2019)
- [Cuckoo Trie: Exploiting Memory-Level Parallelism for Efficient DRAM Indexing](https://mlsanity.com/paper/2201.09331/) (2022)

## Built on by (most cited first, 4 of 4)

- [A New Paradigm in Tuning Learned Indexes: A Reinforcement Learning Enhanced Approach](https://mlsanity.com/paper/2502.05001/) (2025)
- [AI-Driven Research for Databases](https://mlsanity.com/paper/2604.06566/) (2026)
- [Mathematical Foundations of Poisoning Attacks on Linear Regression over Cumulative Distribution Functions](https://mlsanity.com/paper/2603.00537/) (2026)
- [Sig2Model: A Boosting-Driven Model for Updatable Learned Indexes](https://mlsanity.com/paper/2509.20781/) (2025)

## Abstract (arXiv)

Learned indexes use machine learning models to learn the mappings between keys and their corresponding positions in key-value indexes. These indexes use the mapping information as training data. Learned indexes require frequent retrainings of their models to incorporate the changes introduced by update queries. To efficiently retrain the models, existing learned index systems often harness a linear algebraic QR factorization technique that performs matrix decomposition. This factorization approach processes all key-position pairs during each retraining, resulting in compute operations that grow linearly with the total number of keys and their lengths. Consequently, the retrainings create a severe performance bottleneck, especially for variable-length string keys, while the retrainings are crucial for maintaining high prediction accuracy and in turn, ensuring low query service latency. To address this performance problem, we develop an algorithm-hardware co-designed string-key learned index system, dubbed SIA. In designing SIA, we leverage a unique algorithmic property of the matrix decomposition-based training method. Exploiting the property, we develop a memoization-based incremental training scheme, which only requires computation over updated keys, while decomposition results of non-updated keys from previous computations can be reused. We further enhance SIA to offload a portion of this training process to an FPGA accelerator to not only relieve CPU resources for serving index queries (i.e., inference), but also accelerate the training itself. Our evaluation shows that compared to ALEX, LIPP, and SIndex, a state-of-the-art learned index systems, SIA-accelerated learned indexes offer 2.6x and 3.4x higher throughput on the two real-world benchmark suites, YCSB and Twitter cache trace, respectively.