# 9× faster than ClickHouse

Source: https://www.hotdata.dev/compare/clickhouse
Site index: https://www.hotdata.dev/llms.txt

On the same hardware, with your ClickHouse SQL as written.

## At a glance

| | Hotdata | ClickHouse |
| --- | --- | --- |
| What it is | A Rust query engine built for low latency at high concurrency. | A column-oriented analytics database. |
| Where data lives | Queries your sources in place, or Parquet in an instant database. | Loaded into ClickHouse tables. |
| Deployment | In your VPC, run by us. Or fully managed. | Self-hosted or ClickHouse Cloud. |
| Pricing | Compute runs only while queries run. | Compute by the hour, plus storage. |
| Concurrency | 9× faster than ClickHouse with 10 concurrent clients. | Scales up with bigger services. |
| Combining sources | Join Postgres, warehouses, lakes, and files in one query. | Built around its own tables. |
| Search | Vector, full-text, and geospatial in one SQL query. | Vector and full-text indexes. |
| SQL | Postgres-familiar HotSQL. PostgreSQL, DuckDB, Snowflake, and ClickHouse SQL run as written. | ClickHouse SQL. |

## Performance

TPC-H SF1 · 22 queries · 10 concurrent clients.

| System | Hardware | p50 | p95 | Queries/s |
| --- | --- | --- | --- | --- |
| Hotdata | 4 vCPU · 16 GiB | 0.17 s | 0.38 s | 51.1 |
| ClickHouse | 4 vCPU · 16 GiB | 1.40 s | 3.39 s | 6.4 |

## When to choose Hotdata

- You want 9× lower tail latency on the same hardware.
- Your data lives in Postgres, a warehouse, and object storage, and you don't want to load it all first.
- Your ClickHouse SQL should run as written, next to Postgres and Snowflake SQL.
- You want it run for you, in your own VPC.

## When ClickHouse still fits

- You ingest large event streams into one database and serve dashboards from it.
- You want to self-host and tune the database yourself.

## Get started

The [quick start](/docs/quick-start) gets a table into Hotdata in a few minutes.