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
View all SubscribersAdd a CodeTriage badge to clickhouse
Help out
- Issues
- Bound each docker-library test run in the Docker image jobs
- Join reordering costs a many-to-many join as a key lookup when NDV statistics are missing: TPC-H Q5 times out on repeat runs
- Run the bare-binary clickhouse-local calls of tests 03134, 03267 and 03536 with a per-test HOME
- Join reordering: without NDV, assume the smallest table joined on a key holds the unique key
- Let the adaptive aggregator thaw under the top-K pruning of `ORDER BY count() DESC LIMIT n`
- MergeTree reads build the same PREWHERE steps once per stream: build them once per read (ClickBench Q42 -5.7%, Q41 on 20 parts -5.8% at max_threads=16)
- Park three macOS test hosts in runner-init and log its output to a file the CloudWatch agent tails
- Lazy materialization reads the deferred columns with only as many threads as the main read has streams: let its readers use max_threads (ClickBench Q23 −23%, SELECT * ... LIMIT 10 −28%)
- Fix missing rows from JOIN runtime filter index analysis when the probe key nests a CAST
- Bring 04666_run_query_in_background under 10 s by overlapping its CREATE AS SELECT populates with the other cases
- Docs
- C++ not yet supported