Support parallel execution for aggregation with distinct
Open
sig/execution
type/enhancement
- Dominant language
- Go
- Stars
- 40.5k
- Forks
- 6.2k
- PR merge metrics
- PR metrics pending
Description
## Enhancement
Currently, if an aggregation contains at least 1 agg function with distinct, TiDB will execute the aggregation using 1 threads, it is clearly not optimal for big data set, we should support parallel exectuion for aggregation with distinct, at least for `count(distinct)` which is widely used.
Contributor guide
Assessment
This issue has not been assessed yet.