This repo presents benchmarks(LeetCode, Calcite) used by SlabCity. SlabCity is a synthesis-based query rewriting technique capable of whole-query optimization without relying on any rewrite rules. SlabCity directly searches the space of SQL queries using a novel query synthesis algorithm that leverages a new concept called query dataflows.
Due to patent copyright issues, we are not open sourcing our code. Any local copies of code obtained during evaluation phase and before 9/8/2023 should be treated condidential and should not be shared or distributed without explicit permission from the authors.
This repo contains following supplemental materials:
- A new benchmark for query rewriting, which has 1131 real-life queries from LeetCode participants
- Calcite benchmark used in the paper
A brief structure is shown below:
├── benchmarks
│ ├── LeetCode # leetcode benchmark used in paper
│ └── CalCite # calcite benchmark used in paper
| └── data # generation scripts used in paper
└── README.mdEach benchmark under benchmarks folder contains two sub-folders: queries and schema. Queries are stored in csv files in queries folder, where each line has a unique query id in each problem and the query text.
Queries are stored in csv files in queries folder, where each line has a unique query id in each problem and the query text. For example, benchmarks/LeetCode/queries/1308.csv has a line
2,"select distinct gender, day, sum(score_points) over (partition by gender order by day) as total from scores order by gender, day"
which means this query is crawled from LeetCode problem 1308, and a unique id 2. Each query is syntactically different.
schema folder contains schemas stored in json file. The schema has following format:
{
"Problem Number": ..
"Tables":[
{
"TableName": ..,
"PKeys":[..]
"FKeys":[..]
"Others":[..]
}
]
}
Primary keys are recorded in PKeys array, Foreign keys are recorded in FKeys array, and other columns are recorded in Others array,
Besides these integrity constraints, LeetCode problems often have more constraints. A complete set of constraints can be can be found in contraints folder under LeetCode benchmark folder.
benchmarks/data folder contains all scripts used to generate databases for each benchmark. Detailed instructions can be found in README.md under that folder.
If you use this benchmark, please cite our paper: https://dl.acm.org/doi/abs/10.14778/3611479.3611515