ClickHouse
Open-source column-oriented OLAP database for real-time analytical queries over large datasets.
Praised for very fast analytical queries, often after moving off Postgres,123 and for cheap, low-maintenance log and event storage.456 Commenters stress that it complements rather than replaces Postgres,78 and some find it heavier to operate than DuckDB.910
Why people recommend it
Query speed · 3
Users report large speedups on analytics queries, such as 60s to 0.5s and a year of data in about 200ms.
“truly amazing software that just works without any tuning for this kind of exploratory analytics workload.”
“Since moving to Clickhouse I think I can grab a full year's data in around 200ms”
Show 1 more
“before we switched to clickhouse and it was night and day.”
Logging backend · 3
A common choice for log storage with Vector, Grafana or OTel, outperforming Loki on the same hardware.
“substantially better performance on the same hardware”
“We use Vector and Clickhouse for logging and works really well.”
Show 1 more
“we use Grafana + Loki + ClickHouse and it's been mostly maintenance-free for years. Costs under $100/month.”
Low maintenance · 3
Described as working without tuning and needing far less upkeep than multi-system stacks.
“we switched from Druid+postgres+trino to just one big clickhouse node and I’ve never looked back. Much more performant, practical, and a lot less maintenance.”
“we use Grafana + Loki + ClickHouse and it's been mostly maintenance-free for years. Costs under $100/month.”
Show 1 more
“truly amazing software that just works without any tuning for this kind of exploratory analytics workload.”
Data connectors · 3
Reads directly from S3, Postgres and other sources, and offers a single-binary local mode.
“ClickHouse also supports a lot of data sources and has a local mode where you just use a single binary with local-only access.”
“we switched from Druid+postgres+trino to just one big clickhouse node and I’ve never looked back. Much more performant, practical, and a lot less maintenance.”
Show 1 more
“we use regular CH in production. I'm pretty familiar with it and I can always run large queries on a production server with 1TB of RAM.”
Scalability · 2
Handles billions of events and can be deployed on clusters of many machines.
“that's why we used clickhouse db here which scales to billions of events easily”
“When a single server is not enough, you deploy ClickHouse on a cluster, up to thousands of machines”
Resource efficiency · 2
Noted for efficient resource use and storage compressed 4-5x better than the previous setup.
“substantially better performance on the same hardware”
Caveats and dissent
Not transactional · 2
Commenters say it is not a Postgres replacement and is best paired with one.
“I use clickhouse AND postgres. Clickhouse is not a replacement for postgres at all.”
“something like clickhouse or redis combined with postgresql might be more than enough for many things”
Operational complexity · 2
Some find DuckDB simpler for single-node work despite ClickHouse being faster.
“I use ClickHouse for production analytics”
“I found that clickhouse was the fastest”
Specialized alternatives · 2
Some teams moved to purpose-built tools, such as VictoriaMetrics for metrics or Parquet plus DuckDB.
“We tried using ClickHouse as a database for metrics in 2017, but then decided implementing a specialized database for metrics.”
“At Cronitor we use ClickHouse, but we're leaving it behind for our next product”
Development pace · 1
One commenter distrusts the very high commit rate and the AI-generated issues in the repo.
“I wouldn't touch it with a 10ft pole considering how quickly they're merging code into main.”
Often compared with
3 moved from PostgreSQL to ClickHouse
1 moved from ClickHouse to DuckDB
1 moved from Loki to ClickHouseDatadog2 comments mention both
1 moved from Datadog to ClickHouse
1 moved from ClickHouse to VictoriaMetrics
Who recommends it in their own words
From the makers not counted in the ranking
“building ClickHouse with PGO does not give any meaningful improvement”
“I host a publicly open database with Hacker News data at”
“Sai from ClickHouse”
“another amazing database technology which is equally cooler and useful to the world and for a different use-case (analytics)”
Threads where it came up
- Ten years of ClickHouse in open source 6 mentions
- I built Vector. Now I'm answering the question your observability vendor won't 2 mentions
- Remove pgrust: the entry is cheating (per-query kernels and result caching) 1 mention
- Show HN: Pug – open-source product analytics 1 mention
- PGO or not PGO this is the dilemma. Step 2 1 mention
- I've operated petabyte-scale ClickHouse clusters for 5 years 1 mention
Data: clickhouse.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 37
“Turning off dictionary-related optimizations in pgrust immediately made it 5x slower (from 3.8s to 20.4s on my machine), while ClickHouse takes a total of 15.0s.”
“built using go, nats, clickhouse, postgres”
“I wouldn't touch it with a 10ft pole considering how quickly they're merging code into main.”
“we use regular CH in production. I'm pretty familiar with it and I can always run large queries on a production server with 1TB of RAM.”
“which exports to ClickHouse (formerly Datadog)”
“clickhouse is resource efficient”
“truly amazing software that just works without any tuning for this kind of exploratory analytics workload.”
“have even experimented with Clickhouse as a complete alternative”
“At Cronitor we use ClickHouse, but we're leaving it behind for our next product”
“It includes a ClickHouse data lake for ingesting data”
“We use ClickHouse PG behind it, and haven't had any issues so far”
“that's why we used clickhouse db here which scales to billions of events easily”
“We use Clickhouse and Python or Go.”
“something like clickhouse or redis combined with postgresql might be more than enough for many things”
“we used ClickHouse SQL as a starting point as that's what our underlying DB is”
“substantially better performance on the same hardware”
“we switched from Druid+postgres+trino to just one big clickhouse node and I’ve never looked back. Much more performant, practical, and a lot less maintenance.”
“Since moving to Clickhouse I think I can grab a full year's data in around 200ms”
“The web admin panel they built is great and we’ve had solid performance.”
“I use clickhouse AND postgres. Clickhouse is not a replacement for postgres at all.”
“ClickHouse queries are generated on the fly”
“ClickHouse also supports a lot of data sources and has a local mode where you just use a single binary with local-only access.”
“rather than ClickHouse-class scans over cold history”
“We tried using ClickHouse as a database for metrics in 2017, but then decided implementing a specialized database for metrics.”
“maybe some kind of OTel, wide log, ClickHouse, Grafana thing”
“built around markdown and clickhouse”
“I built an MCP over our backend and Clickhouse”
“We use ClickHouse for a trend detection pipeline”
“I use ClickHouse for production analytics”
“Our tech stack is React, Typescript, AWS, Pulumi, Postgres, Clickhouse and Turbopuffer.”
“before we switched to clickhouse and it was night and day.”
“When a single server is not enough, you deploy ClickHouse on a cluster, up to thousands of machines”
“we use Grafana + Loki + ClickHouse and it's been mostly maintenance-free for years. Costs under $100/month.”
“routing to different ClickHouse tables in one place”
“Our stack is Go, NextJS (TS), Postgres and ClickHouse.”
“We use Vector and Clickhouse for logging and works really well.”
“I found that clickhouse was the fastest”