Selective array and map column reader #10448
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
✅ Deploy Preview for meta-velox canceled.
|
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
7d880e9 to
c15cbed
Compare
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 14, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Differential Revision: D58937281
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 14, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Differential Revision: D58937281
c15cbed to
06df407
Compare
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
06df407 to
65ae727
Compare
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 22, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Differential Revision: D58937281
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 22, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Differential Revision: D58937281
65ae727 to
b8d779a
Compare
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
Yuhta
approved these changes
Jul 22, 2024
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 22, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Differential Revision: D58937281
b8d779a to
042f656
Compare
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 23, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Reviewed By: Yuhta Differential Revision: D58937281
042f656 to
85d2b57
Compare
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
HuamengJiang
pushed a commit
to HuamengJiang/velox-1
that referenced
this pull request
Jul 23, 2024
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Reviewed By: Yuhta Differential Revision: D58937281
85d2b57 to
26e885b
Compare
Summary: Pull Request resolved: facebookincubator#10448 Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding. There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options 1) a materializeNonNull api for encodings 2) a materialize materializeNullable api for encodings for combined nulls 3) a way to have selective reader not having to materialize combined nulls without compromising efficiency. For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods. Reviewed By: Yuhta Differential Revision: D58937281
26e885b to
f64b60a
Compare
|
This pull request was exported from Phabricator. Differential Revision: D58937281 |
|
This pull request has been merged in afd6753. |
|
Conbench analyzed the 1 benchmark run on commit There were no benchmark performance regressions. 🎉 The full Conbench report has more details. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Implement selective array and map column reader. This is another type of top level column without independent null streams, hence requiring some new functionalities for loading nullable encoding.
There is another nuance in the diff where selective reader currently always loads the nulls first and then the values, and passes the combined nulls into readLengths methods instead of just the top level incoming nulls for scattering. We have 3 more ideal options
For now we have added a hack in NimbleData to load values along with the nulls for nullable encodings and return the cached value when calling readLengths later. In order to fit this access pattern, we also override the skip methods.
Differential Revision: D58937281