How to Tune Your Database Management System for Peak Performance

Table of Links

Abstract and 1 Introduction

1.1 Configuration Parameter Tuning Challenges and 1.2 Contributions

2 Tuning Objectives

3 Overview of Tuning Framework

4 Workload Characterization and 4.1 Query-level Characterization

4.2 Runtime-based Characterization

5 Feature Pruning and 5.1 Workload-level Pruning

5.2 Configuration-level Pruning

5.3 Summary

6 Knowledge from Experience

7 Configuration Recommendation and 7.1 Bayesian Optimization

7.2 Neural Network

7.3 Reinforcement Learning

7.4 Search-based Solutions

8 Experimental Setting

9 Related Work

10 Discussion and Conclusion, and References

3 OVERVIEW OF TUNING FRAMEWORK

In this study, we break down the entire knob tuning pipeline into five essential components: Workload Characterization, Feature Pruning, Knowledge from Experience, Configuration Recommendation as depicted in Fig. 1. The initial phase, Workload Characterization, is crafted to comprehensively model a workload, utilizing either logical-level queries or embedding runtime metrics for a dynamic understanding of performance indicators Feature Pruning, applied both at the workload-level to minimize execution time and at the configuration-level to streamline the search space, is strategically employed to enhance efficiency through the application of pruning techniques. Knowledge from Experience draws upon historical insights, enabling the tuning algorithm to efficiently converge towards optimal configurations. Finally, Configuration Recommendation involves utilizing diverse techniques, including Bayesian Optimization, Neural networks, Reinforcement learning, and Search-based methods, to generate configurations finely tailored to the specific characteristics of the workload.