When analyzing relational time-series data, it is essential to have the ability to join multiple datasets using time dimension by looking up at the value being closest in time.
Without proper query engine support, users are forced to implement inefficient work-around which is both slow & hard to read.
Description & use-case
When analyzing relational time-series data, it is essential to have the ability to join multiple datasets using time dimension by looking up at the value being closest in time.
Without proper query engine support, users are forced to implement inefficient work-around which is both slow & hard to read.
DuckDB has a great explanation here.
Many other popular databases & query engines support it:
Related issues
The most interesting is probably in DataFusion:
Code of Conduct