Computer Science > Databases
[Submitted on 3 Mar 2024]
Title:TreeTracker Join: Turning the Tide When a Tuple Fails to Join
View PDFAbstract:Many important query processing methods proactively use semijoins or semijoin-like filters to delete dangling tuples, i.e., tuples that do not appear in the final query result. Semijoin methods can achieve formal optimality but have high upfront cost in practice. Filter methods reduce the cost but lose the optimality guarantee.
We propose a new join algorithm, TreeTracker Join ($\mathsf{TTJ}$), that achieves the data complexity optimality for acyclic conjunctive queries (ACQs) without semijoins or semijoin-like filters. $\mathsf{TTJ}$ leverages join failure events, where a tuple from one of the relations of a binary join operator fails to match any tuples from the other relation. $\mathsf{TTJ}$ starts join evaluation immediately and when join fails, $\mathsf{TTJ}$ identifies the tuple as dangling and prevents it from further consideration in the execution of the query. The design of $\mathsf{TTJ}$ exploits the connection between query evaluation and constraint satisfaction problem (CSP) by treating a join tree of an ACQ as a constraint network and the query evaluation as a CSP search problem. $\mathsf{TTJ}$ is a direct extension of a CSP algorithm, TreeTracker, that embodies two search techniques backjumping and no-good. We establish that join tree and plan can be constructed from each other in order to incorporate the search techniques into physical operators in the iterator form. We compare $\mathsf{TTJ}$ with hash-join, a classic semijoin method: Yannakakis's algorithm, and two contemporary filter methods: Predicate Transfer and Lookahead Information Passing. Favorable empirical results are developed using standard query benchmarks: JOB, TPC-H, and SSB.
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.