clickhouse
https://github.com/yandex/clickhouse
C++
ClickHouse is a free analytic DBMS for big data.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
C++ not yet supported66 Subscribers
Add a CodeTriage badge to clickhouse
Help out
- Issues
- Parquet: a file whose row groups are all skipped by statistics still builds the PREWHERE steps; skipping that makes ClickBench Q43 over 100 files about 7% faster with 14% less CPU
- `SET profile` / `SETTINGS profile` / `?profile=` bypass `CONST` constraints for users defined in users.xml, contrary to the constraints docs
- Iceberg tables are much slower than MergeTree in the performance tests: subcolumns, sort order, join planning, position deletes
- Do not look up a stream file for the size hints of every granule of a Compact part
- Parquet ORDER BY <String> LIMIT with s != '' skips no row groups because every minimum is ''; bounding them by their dictionary makes ClickBench Q26 12-16% faster
- Fix ZNONODE race when reading StorageKafka2 topic partition locks
- parseDateTime resets month and day when a weekday follows the date
- Parquet: decoding dictionary indexes under a PREWHERE filter re-scans the filter for every run; one bitmask per call makes ClickBench Q31 and Q12 7-10% faster
- `mapValues` `bloom_filter` throws `BAD_GET`: constant not cast to value type
- Parquet over S3 at default settings waits a full round trip for each separately read row group; reading the next one ahead makes ClickBench S3 queries 20-27% faster
- Docs
- C++ not yet supported