Parquet file compaction #621
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
thorfour
force-pushed
the
parquet-file-part
branch
3 times, most recently
from
b20f7fc to
bd69917
Compare
asubiotto
reviewed
Dec 12, 2023
thorfour
force-pushed
the
parquet-file-part
branch
4 times, most recently
from
13cc7fd to
6508cab
Compare
Adds a Release() requirement to the Part interface this allows the file parts to mark their sections as released to be able to safely truncate the underylying file.
Instead of checking if it's a arrow Part allow the Part to determine how to release
Adds Parquet file compaction that will append the Parquet files into a os temporary file. During compaction of the file part it will truncate the file after the parts are released so that new parts can be appended to the file.
Instead of having it be responsible for each level just have it hande a single level, and make multiple of them when configuring the levels
Instead of creating a new file part we can just reuse the parquet part with the underlying parquet.File in the buffer being on disk
thorfour
force-pushed
the
parquet-file-part
branch
from
6508cab to
a1ac233
Compare
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Outdated suggestions cannot be applied.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.
This adds an optional compaction function to FrostDB to allow for compaction to write to disk instead of an in-memory buffer.
It appends compacted parts into an os temporary file, and will truncate the file when the level is compacted into the next level.
I plan to run benchmarks on this but initially ran with these settings on Parca and observed the expected reduction in memory usage, but would expect the query latency to increase as well as compaction times.
Closes #605