hongzhi-gao opened a new pull request, #18638:
URL: https://github.com/apache/iotdb/pull/18638
## Summary
- Decode RPC result blocks directly into buffered Python tuples for row
iteration.
- Avoid eager Pandas imports in the result decoding path.
- Preserve the existing `SessionDataSet`, `RowRecord`, `Field`, and
DataFrame APIs.
- Keep the vectorized DataFrame path unchanged.
## Performance
Benchmark: 50,000 rows with `INT32`, `DOUBLE`, and `TEXT` columns; median
of 7 runs.
| Operation | Before | After |
|---|---:|---:|
| `SessionDataSet.next()` | 116,698 rows/s | 594,259 rows/s |
| `todf()` | 1,262,358 rows/s | 1,265,111 rows/s |
Row iteration is approximately 5.1x faster, while DataFrame performance
remains unchanged.
--
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]