hubgeter opened a new pull request, #64611:
URL: https://github.com/apache/doris/pull/64611
### What problem does this PR solve?
Problem Summary:
The Storage-API writer sized each block by calling
estimateSingleRowPayloadBytes for every row, which built a one-row Arrow batch
and ZSTD-serialized it just to count bytes. On large/wide data this caused two
problems:
- CPU: per-row Arrow allocation + ZSTD serialization dominated the write
path, making big-table writes take hours.
- Accuracy: a one-row batch is dominated by fixed Arrow IPC framing/padding,
so the size was over-estimated ~25x and blocks rotated far too often.
Replace it with estimateBatchPayloadBytes(), which sums
FieldVector.getBufferSize() over the already-built batch (O(columns), no
serialization). Rewrite the per-row loop as writeBatch(), which writes the
whole input batch and lazily rotates the block once the accumulated size
reaches maxBlockBytes.
### Release note
None
### Check List (For Author)
- Test <!-- At least one of them must be included. -->
- [ ] Regression test
- [ ] Unit Test
- [ ] Manual test (add detailed scripts or steps below)
- [ ] No need to test or manual test. Explain why:
- [ ] This is a refactor/code format and no logic has been changed.
- [ ] Previous test can cover this change.
- [ ] No code files have been changed.
- [ ] Other reason <!-- Add your reason? -->
- Behavior changed:
- [ ] No.
- [ ] Yes. <!-- Explain the behavior change -->
- Does this need documentation?
- [ ] No.
- [ ] Yes. <!-- Add document PR link here. eg:
https://github.com/apache/doris-website/pull/1214 -->
### Check List (For Reviewer who merge this PR)
- [ ] Confirm the release note
- [ ] Confirm test cases
- [ ] Confirm document
- [ ] Add branch pick label <!-- Add branch pick label that this PR should
merge into -->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
