New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
dataflow: add dataflow rendering to EXPLAIN plans #7301
Comments
This is mostly ready on the dataflow side but requires adding the SQL commands and pretty printing (we probably need to discuss what we actually want in there). |
The initial work has been done, but we'll keep this open to scope out the follow-up tasks. |
@uce: can you put labels to the areas where the outstanding tasks fall? I've tentatively added |
A variant of this was done in #13137. |
If there's a view with an index on a column, a point select on that column will be significantly faster. But an EXPLAIN will not surface the use of the index. We should add this information to EXPLAIN. In this case looking at the rendered dataflow will show index instead of scan. Users will be able to self-serve improve their queries better with this information.
Update: 2022/02/08
We added a rudimentary version of this in #8515. We were holding off documenting it, because it's very verbose and we expected to follow-up sooner. We didn't follow-up, it's still verbose, but came in handy a few times already, so I'm proposing the following sub-tasks:
explain physical plan
toexplain dataflow plan
docs/sql/explain
following the existing structure (sections "Reading ...", "Operators in", and an extra warning that it's experimental)There were various discussions in the pats to take a holistic look at
explain
output, but we didn't find a proper owner (cc @JLDLaughlin @andrioni). I think it's fine to do the above minimal things even if we expect things to change in the future.The text was updated successfully, but these errors were encountered: