Operationalizing Property-Based Testing for Data-Intensive Scalable Computing Systems
Read the paper · arXiv:2606.11132
What this paper does with SQLancer
SQLancer is the demonstration the paper builds on -- credited with showing the effectiveness of query oracles including TLP -- and is explicitly excluded from the evaluation: the authors say they do not compare against it because using it for Spark would require an additional adapter.
DiscPBT is a property-based testing engine for Apache Spark. The authors argue that fuzzing's shallow oracles catch crashes but miss semantic drift and optimisation errors in data-intensive scalable computing frameworks, and that operationalising property-based testing there needs reusable property definitions and a way to instantiate them into valid workloads. DiscPBT supplies eight reusable meta-properties and generators for workload skeletons and input data, reaching 1.2x the branch coverage and 1153x the plan diversity of CometFuzz.
How it was classified
uses infrastructure — no
SQLancer is cited, not used; nothing in the mentions describes reusing its code.
extends technique — no
No technique is extended; the citation is background.
compares with — no
M3 states the opposite of a comparison: SQLancer was deliberately not run, because applying it to Spark would need an adapter. A pattern fired on that sentence, which is a comparison only in the sense of explaining its absence.
describes as state of the art — yes
M4 credits SQLancer with demonstrating the effectiveness of query oracles in relational DBMS testing, which is a claim about its standing.
In relational DBMS testing, SQLancer [ 32] demonstrated the effectiveness of query oracles, including TLP [ 2], which checks result preservation under predicate-based partitioning, and NoREC [ 3], which validates queries through equivalence-preserving rewrites that disable optimizer-dependent execution paths.
SQLancer publications it cites (7)
Bibliography entries that resolved to a SQLancer publication, or to a paper by one of the project's authors. A sentence citing one of these numbers is a reference to SQLancer even when it never writes the name.
| # | Entry | Matched as |
|---|---|---|
| 2 | M. Rigger and Z. Su, “Finding bugs in database systems via query partitioning,” Proceedings of the ACM on Programming Languages, vol. 4, no. OOPSLA, pp. 1–30, 2020. | sqlancer publication · TLP |
| 3 | ——, “Detecting optimization bugs in database engines via non-optimizing reference engine construction,” inProceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foun... | sqlancer publication · NOREC |
| 4 | ——, “Testing database engines via pivoted query synthesis,” in14th USENIX Symposium on Operating Systems Design and Implementation (OSDI), 2020, pp. 667–682. Yaoxuan Wu, Ingrid Lee, Ahmad Humayun, Muhammad Ali Gulzar,... | sqlancer publication · PQS |
| 5 | C. Zhang, L. Wang, and M. Rigger, “Finding cross-rule optimization bugs in datalog engines,”Proceedings of the ACM on Programming Languages, vol. 8, no. OOPSLA1, pp. 110–136, 2024. | project authored |
| 32 | “Sqlancer: Automatic testing of database systems,” https://github.com/sqlancer/ sqlancer, accessed 2026-01-28. | sqlancer publication |
| 33 | C. Zhang and M. Rigger, “Constant optimization driven database system testing,” Proceedings of the ACM on Management of Data, vol. 3, no. 1, pp. 1–24, 2025. | sqlancer publication · CODDTEST |
| 34 | Q. Mang, J. Ba, P. He, and M. Rigger, “Finding logic bugs in graph-processing systems via graph-cutting,” inProceedings of the ACM on Management of Data(SIGMOD), 2025. | project authored |
Every place it refers to SQLancer (7)
7 sentences, each stored verbatim from the extracted text with where it was found and how. “Citation marker” means the sentence names no tool at all and was reached through a reference number that resolved to a SQLancer publication.
| Id | Sentence | Found by | Where |
|---|---|---|---|
| M1 | Prior work has demonstrated the value of semantic oracles based on rewrites and decompositions [ 2–5]. |
citation marker |
1 Introduction page 1 |
| M2 | Work SUT Workload Gen Data Gen Semantic Oracles DB Fuzzing and Testing SQLsmith [1] DB engine✓ ✗None (failure/crash) SQLancer [2–5] DB engine Templ. |
name |
1 Introduction page 2 |
| M3 | We also do not compare against SQLancer because using it for Spark would require an additional adapter. |
name |
5.2 RQ2: Workflow Complexity and Coverage page 10 |
| M4 | In relational DBMS testing, SQLancer [ 32] demonstrated the effectiveness of query oracles, including TLP [ 2], which checks result preservation under predicate-based partitioning, and NoREC [ 3], which validates queries through equivalence-preserving rewrites that disable optimizer-dependent execution paths. |
name |
6 Related Work page 11 |
| M5 | CODDTest [ 33] explores constant-based subquery rewrites as a semantic test oracle. |
technique |
6 Related Work page 11 |
| M6 | In Datalog testing, IRE checks cross-rule optimizations through computation composition [ 5]. |
citation marker project authored |
6 Related Work page 11 |
| M7 | In graph-system testing, GSlicer [ 34] derives metamorphic checks by partitioning graph inputs to validate graph libraries and graph databases. |
citation marker project authored |
6 Related Work page 11 |