Opening The Black-Box: Explaining Learned Cost Models For Databases

Download PDF

Abstract

Learned Cost Model s ( LCM s) have shown superior results over traditional database cost models as they can significantly improve the accuracy of cost predictions. However, LCM s still fail for some query plans, as prediction errors can be large in the tail. Unfortunately, recent LCM s are based on complex deep ne ural models, and thus, there is no easy way to understand where this accuracy drop is rooted, which critically prevents systematic troubleshooting. In this demo paper, we present the very first approach for opening the black box by bringing AI explainability approaches to LCM s. As a core contribution, we developed new explanation techniques that extend existing methods that are available for the general explainability of AI models and adapt them significantly to be usable for LCM s. In our demo, we provide an interactive tool to showcase how explainability for LCM s works. We believe this is a first step for making LCM s debuggable and thus paving the road for new approaches for systematically fixing problems in LCMs.