NEW · CLICKHOUSE 25

The fastest open-source database for real-time analytics.

Query terabytes in under a second. Run SQL over events, logs, and metrics — without an ETL army.

47k+
GitHub stars
2.8k+
Contributors
779+
Releases
events.sql 847ms · 12.4B rows
123456789
-- top countries by event volume SELECT country, count() AS events, uniqExact(user_id) AS users FROM events WHERE ts >= now() - INTERVAL 1 DAY GROUP BY country ORDER BY events DESC LIMIT 10;
CLICKHOUSE read · 18.2 GB · scanned in 0.83s throughput · 21.9 GB/s
Anthropic Cloudflare eBay Lyft Sentry Tinybird Disney+ GitLab
BUILT FOR SCALE

Engineered for the hardest analytics workloads.

Columnar storage, vectorized execution, and per-partition parallelism — ClickHouse turns single-node hardware into a distributed analytics cluster.

Sub-second on petabytes.

Vectorized SQL engine reads only the columns you query. Compress 10× without losing speed.

Real-time ingest.

Stream from Kafka, Kinesis, or HTTP. Query rows the moment they land — no batch window.

One node, or a fleet.

Run on a laptop, scale to thousands of replicas — the same SQL, the same storage format.

SQL you already know.

ANSI-SQL with JSON, geo, vector, and time-series functions bolted on without ceremony.

Open-source, always.

Apache 2.0. 47k stars, 2.8k contributors. Run it self-hosted or use ClickHouse Cloud.

Multi-tenant ready.

Quotas, row-level security, and resource pools per workload — designed for shared infra.

CLICKHOUSE CLOUD

Your data warehouse,
without the warehouse.

A fully-managed ClickHouse service that auto-scales storage and compute independently. No clusters, no babysitting — just SQL.

console.clickhouse.cloud / analytics_prod healthy · us-east-1
Rows scanned · 1h
12.4B
Avg latency
23ms
QPS
8,142
Storage
4.7 TB
SELECT count() FROM events WHERE ts >= now() - INTERVAL 1 HOUR 847ms
PRICING

Pay for queries.
Not idle clusters.

Development

$0 /month

For prototyping and dev environments. Spin up a service in seconds.

  • 16 GB shared compute
  • Auto-suspend
  • 1 service
Try free

Enterprise

Custom · annual

BYOC, private networking, dedicated capacity. Talk to our team.

  • Single-tenant deployments
  • Private link & VPC peering
  • Compliance attestations
Talk to sales

Deploy your way.

Self-hosted on your hardware, fully-managed in ClickHouse Cloud, or BYOC inside your AWS account.

Start free Read docs
DESIGN SYSTEM REFERENCE

Tokens, by ClickHouse.

Near-pure black canvas, electric-yellow voltage, Inter 700 with tight tracking, SQL code blocks. The yellow is the brand.

→ READ THE FULL DESIGN.MD

Color

grouped & named

Brand

primary#faff69
primary-active#e6eb52
primary-disabled#3a3a1f

Surface

canvas#0a0a0a
surface-soft#121212
surface-card#1a1a1a
surface-elevated#242424
hairline#2a2a2a

Text

ink / on-dark#ffffff
body#cccccc
muted#888888
muted-soft#5a5a5a

Semantic

success#22c55e
warning#f59e0b
error#ef4444
accent-blue#3b82f6

Type

Inter 700 · tight tracking
display-xl72 / -2.5 The fastest database.
display-lg56 / -2 Section opener.
display-md40 / -1.5 Sub-section head.
stat-display56 / -1.5 47k+
title-lg24 / -0.3 Pricing tier title.
body-md16 / 0 Default running text on the black canvas — Inter weight 400.
code14 / 0 SELECT count() FROM events;
caption-uppercase12 / 1.5 SECTION LABEL · 12 / 600

Radius

8px buttons · 12px cards
XS · 4
SM · 6
MD · 8
LG · 12
PILL

Components

working matrix
BADGES & PILLS
NEW healthy v25.3
CATEGORY TABS
Overview Queries Metrics
TEXT INPUT
CODE WINDOW · SQL
query.sql 23ms
123
SELECT country, count() AS n FROM events GROUP BY country;
View DESIGN.md