Back to papers
AXE: A Task Decomposition Approach to Learned LSM Tuning
Summary: AXE decomposes LSM tuning into (1) training a learned surrogate cost model from logs or existing performance models and (2) synthesizing many training samples to train a learned tuner that optimizes the surrogate, avoiding costly online executions. Outperforms BO 71% of the time with 100x lower tuning overhead, handles categorical knobs, scales across instances without retraining, and reduces reliance on expert cost models/solvers.
(summarized by gpt-5-mini on Feb 09 2026)
- Paper ID
- 14206
- Venue
- VLDB
- Year
- 2025
- Pagerank
- 4.1905499e-05
- Overall Rank
- 10,853 | 24.58%
- DOI
-
10.14778/3773731.3773735
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank |
Citing Paper |
Year |
Venue |
Pagerank |
Outgoing Citations (Sorted by Pagerank)
Showing 30 of 30 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank |
Cited Paper |
Year |
Venue |
Pagerank |
| 160 |
Automated Selection of Materialized Views and Indexes for SQL Databases |
2000 |
VLDB |
0.00040053897 |
| 183 |
Automatic Database Management System Tuning Through Large-scale Machine Learning |
2017 |
SIGMOD |
0.00036859633 |
| 237 |
An Efficient, Cost-Driven Index Selection Tool for Microsoft SQL Server |
1997 |
VLDB |
0.00031727601 |
| 407 |
Database Cracking |
2007 |
CIDR |
0.00023941779 |
| 510 |
An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning |
2019 |
SIGMOD |
0.00021420477 |
| 608 |
Monkey: Optimal Navigable Key-Value Store |
2017 |
SIGMOD |
0.00019233548 |
| 779 |
QTune: A Query-Aware Database Tuning System with Deep Reinforcement Learning |
2019 |
VLDB |
0.00016719473 |
| 804 |
An End-to-End Learning-based Cost Estimator |
2020 |
VLDB |
0.0001643674 |
| 1,309 |
Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging |
2018 |
SIGMOD |
0.00012655712 |
| 1,437 |
AsterixDB: A Scalable, Open Source BDMS |
2014 |
VLDB |
0.00011973401 |
| 1,614 |
MyRocks: LSM-Tree Database Storage Engine Serving Facebook's Social Graph |
2020 |
VLDB |
0.00011137963 |
| 2,153 |
The Data Calculator*: Data Structure Design and Cost Synthesis from First Principles and Learned Cost Models |
2018 |
SIGMOD |
9.418541e-05 |
| 2,606 |
Design Continuums and the Path Toward Self-Designing Key-Value Stores that Know and Learn |
2019 |
CIDR |
8.4621503e-05 |
| 3,623 |
Cost Models for Big Data Query Processing: Learning, Retrofitting, and Our Findings |
2020 |
SIGMOD |
6.9017341e-05 |
| 3,655 |
Facilitating Database Tuning with Hyper-Parameter Optimization: A Comprehensive Experimental Evaluation |
2022 |
VLDB |
6.8723042e-05 |
| 3,995 |
ResTune: Resource Oriented Tuning Boosted by Meta-Learning for Cloud Databases |
2021 |
SIGMOD |
6.5475871e-05 |
| 4,180 |
LlamaTune: Sample-Efficient DBMS Configuration Tuning |
2022 |
VLDB |
6.3725334e-05 |
| 4,227 |
Cosine: A Cloud-Cost Optimized Self-Designing Key-Value Storage Engine |
2022 |
VLDB |
6.3381409e-05 |
| 4,800 |
Efficient Deep Learning Pipelines for Accurate Cost Estimations Over Large Scale Query Workload |
2021 |
SIGMOD |
5.9077188e-05 |
| 4,863 |
COLT: Continuous On-Line Database Tuning |
2006 |
SIGMOD |
5.8646122e-05 |
| 5,250 |
One Model to Rule them All: Towards Zero-Shot Learning for Databases |
2022 |
CIDR |
5.6007779e-05 |
| 6,117 |
Compactionary: A Dictionary for LSM Compactions |
2022 |
SIGMOD |
5.1992629e-05 |
| 6,394 |
Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty |
2022 |
VLDB |
5.0770427e-05 |
| 6,440 |
From Auto-tuning One Size Fits All to Self-designed and Learned Data-intensive Systems |
2019 |
SIGMOD |
5.0546781e-05 |
| 6,494 |
Foundations of Automated Database Tuning |
2006 |
VLDB |
5.0334983e-05 |
| 7,623 |
Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads |
2023 |
SIGMOD |
4.6890662e-05 |
| 7,742 |
Rethinking Learned Cost Models: Why Start from Scratch? |
2023 |
SIGMOD |
4.6585812e-05 |
| 8,011 |
CAMAL: Optimizing LSM-trees via Active Learning |
2024 |
SIGMOD |
4.6022693e-05 |
| 9,069 |
Structural Designs Meet Optimality: Exploring Optimized LSM-tree Structures in A Colossal Configuration Space |
2024 |
SIGMOD |
4.3983078e-05 |
| 9,194 |
MLOS in Action: Bridging the Gap Between Experimentation and Auto-Tuning in the Cloud |
2024 |
VLDB |
4.3726269e-05 |
Semantically Similar Papers
| Overall Rank |
Paper |
Year |
Venue |
Pagerank |
| 8,490 |
SA-LSM: Optimize Data Layout for LSM-tree Based Storage using Survival Analysis |
2022 |
VLDB |
4.494994e-05 |
| 9,390 |
Rethinking The Compaction Policies in LSM-trees |
2025 |
SIGMOD |
4.341433e-05 |
| 4,180 |
LlamaTune: Sample-Efficient DBMS Configuration Tuning |
2022 |
VLDB |
6.3725334e-05 |
| 3,567 |
Accordion: Better Memory Organization for LSM Key-Value Stores |
2018 |
VLDB |
6.9606988e-05 |
| 5,801 |
Dissecting, Designing, and Optimizing LSM-based Data Stores |
2022 |
SIGMOD |
5.3217858e-05 |
| 8,011 |
CAMAL: Optimizing LSM-trees via Active Learning |
2024 |
SIGMOD |
4.6022693e-05 |
| 1,368 |
SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data |
2017 |
VLDB |
0.0001235708 |
| 6,394 |
Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty |
2022 |
VLDB |
5.0770427e-05 |
| 5,920 |
Breaking Down Memory Walls: Adaptive Memory Management in LSM-based Storage Systems |
2021 |
VLDB |
5.2686888e-05 |
| 7,743 |
Efficient Data Ingestion and Query Processing for LSM-Based Storage Systems |
2019 |
VLDB |
4.6581858e-05 |