DBScholar

Back to papers

On the Relative Cost of Sampling for Join Selectivity Estimation

Summary: Quantifies when sampling (t_cross) is cheaper than computing the exact star-join for selectivity estimation, deriving bounds and approximations for relative cost as functions of input relation sizes, arity, and the estimator's precision criterion. Identifies dangling tuples as a major negative factor and characterizes mixed effects of data skew, yielding concrete regimes and thresholds that indicate when sampling is or isn't cost-effective for join selectivity estimation. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID
1011
Venue
PODS
Year
1994
Pagerank
0.00010778889
Overall Rank
1,440 | 90.13%
DOI
10.1145/182591.182594

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@inproceedings{haas_pods94,
        address = {New York, NY, USA},
        series = {{PODS} '94},
        title = {{On the Relative Cost of Sampling for Join Selectivity Estimation}},
        url = {https://dl.acm.org/doi/10.1145/182591.182594},
        doi = {10.1145/182591.182594},
        booktitle = {Proceedings of the {ACM} {SIGMOD} Symposium on {Principles} of {Database} {Systems}},
        publisher = {Association for Computing Machinery},
        author = {Haas, Peter J. and Naughton, Jeffrey F. and Swami, Arun N.},
        year = {1994}
}

Incoming Citations (Sorted by Pagerank)

Showing 12 of 12 citing papers.

Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 11 of 11 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers