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
- MergeTreeReaderWide recomputes stream file names (getStreamNameForColumn) on every read: 7-15% of time with small blocks
- Per-block overhead in MergeTreeReadersChain: getColumnsForPatches clones a Block without patch parts, NameSet and shrinkToFit per block
- MergeTreeSource::tryGenerate formats an OpenTelemetry span name per block even when tracing is disabled
- ExecutingGraph::updateNode costs 9-21% with small blocks: per-call heap allocations, map lookup and locks
- High per-block cost of function execution: default-implementation layers and type dispatch cost ~8 µs per Filter block
- Per-chunk bookkeeping: prepareAggregateInstructions, memory tracker check and progress reporting on every block
- Views and MVs ignore CTE column alias list, inlining body unaliased
- Column alias list after a table, CTE or table function is silently dropped
- Query condition cache for TopK reads of Parquet files in object storage
- IN subquery with coarser type than primary key prunes matching rows
- Docs
- C++ not yet supported