Design for Same-Table-JOINs #263
Conversation
Signed-off-by: Andrew Carbonetto <[email protected]>
Codecov Report
@@ Coverage Diff @@
## integ-design-same-table-joins #263 +/- ##
================================================================
Coverage 97.16% 97.16%
Complexity 4120 4120
================================================================
Files 371 371
Lines 10373 10373
Branches 704 704
================================================================
Hits 10079 10079
Misses 287 287
Partials 7 7
Flags with carried forward coverage won't be shown. Click here to find out more. 📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
Signed-off-by: Andrew Carbonetto <[email protected]>
Signed-off-by: Andrew Carbonetto <[email protected]>
|
|
||
| _Phase `*`_ denotes dependencies that must be completed prior to starting this work. | ||
|
|
||
| ### 5.1 Release Schedule |
There was a problem hiding this comment.
Maybe a release schedule and priorities may be better stored in a github issue, and linked to in this document.
There was a problem hiding this comment.
I'm going to keep this here... it's really to talk about what's a must have for the release.
|
|
||
| The following diagram explains the proposed sequence to create JOIN request to OpenSearch | ||
|
|
||
| ### 6.1 Logical Plan Changes - Presentation |
There was a problem hiding this comment.
Maybe we can change the push down OP state change to execute query or something like this. Push down happens during logical plan optimization:
LogicalPlanOptimizer.java#L67-L73
Starting from the deepest part of the Logical Plan Tree each LogicalPlan is optimized/pushed down. After push down most LogicalPlans are removed from the tree, except generally LogicalProject, a logical plan to instantiate an operator for post-processing, and the LogicalRelation will be replaced with the OpenSearchIndexScanBuilder. So we could remove LogicalFilter and LogicalSort from the middle composite state.
There was a problem hiding this comment.
I see what you're saying. The "push down" terminology is a little ambiguous. I think what we're really talking about are the transitions between states. I've changes this to Build Query which I hope makes sense?
There was a problem hiding this comment.
I think the optimize state also doesn't make sense. I've changed this to Create IndexScanBuilder to follow what's in query-optimizer-improvements.md.
Signed-off-by: Andrew Carbonetto <[email protected]>
Signed-off-by: Andrew Carbonetto <[email protected]>
* Add opensearch-same-table-join-query.md Signed-off-by: Andrew Carbonetto <[email protected]> * Update opensearch-same-table-join-query.md for first-pass review Signed-off-by: Andrew Carbonetto <[email protected]> * Add same-table-joins ER relations Signed-off-by: Andrew Carbonetto <[email protected]> * Update document for review comments Signed-off-by: Andrew Carbonetto <[email protected]> --------- Signed-off-by: Andrew Carbonetto <[email protected]>
Description
Added documentation for a same-table join proposal.
related: https://opensearch.org/docs/latest/field-types/join/
Issues Resolved
Check List
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.