feat: Parallel Arrow file format reading #8897
Merged
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
github-actions
bot
added
core
my-vegetable-has-exploded
force-pushed
the
arrow-repartition
branch
from
8e4d9f2 to
9348b09
Compare
alamb
approved these changes
Jan 25, 2024
There was a problem hiding this comment.
The code looks good to me @my-vegetable-has-exploded -- thank you very much.
I think it is likely to not work well on remote object store given how many requests are made but I also think that could be handled by a follow on PR
My only concern with this PR as written is if the tests actually exercise the multi-batch reading code given how small the input files in repartition.slt are
| @@ -61,6 +61,7 @@ postgres = ["bytes", "chrono", "tokio-postgres", "postgres-types", "postgres-pro | |||
| [dev-dependencies] | |||
| env_logger = { workspace = true } | |||
| num_cpus = { workspace = true } | |||
| tokio = { version = "1.0", features = ["rt-multi-thread"] } | |||
| @@ -253,7 +253,16 @@ query TT | |||
| EXPLAIN SELECT * FROM arrow_table | |||
| ---- | |||
| logical_plan TableScan: arrow_table projection=[f0, f1, f2] | |||
| physical_plan ArrowExec: file_groups={1 group: [[WORKSPACE_ROOT/datafusion/core/tests/data/example.arrow]]}, projection=[f0, f1, f2] | |||
| physical_plan ArrowExec: file_groups={4 groups: [[WORKSPACE_ROOT/datafusion/core/tests/data/example.arrow:0..461], [WORKSPACE_ROOT/datafusion/core/tests/data/example.arrow:461..922], [WORKSPACE_ROOT/datafusion/core/tests/data/example.arrow:922..1383], [WORKSPACE_ROOT/datafusion/core/tests/data/example.arrow:1383..1842]]}, projection=[f0, f1, f2] | |||
There was a problem hiding this comment.
👍 looks good to me -- though I I wonder will this actually read in parallel (or do these ranges all end up in the same reader)?
9 tasks
# for free
to join this conversation on GitHub.
Already have an account?
# 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.
Which issue does this PR close?
Closes #8503
Rationale for this change
What changes are included in this PR?
If file_meta.range is some, filter recordbatches according to range, then scan recordbatches.
Are these changes tested?
physical plan of scaning arrow files changes in
repartition_scan.slt.Are there any user-facing changes?