ClickHouse · supported engine

ClickHouse, billions of rows browsable.

HTTP-driver analytics with SQL execution, table browsing, schema introspection, batch inserts, and primary-key-aware row editing.

Included with Mongrel Try free for 7 days

HTTP driver · TLS

Live · click to inspect

The pipeline Mongrel would build for your query.

Click any node below to see rows read, parts and granules scanned, and what Mongrel would suggest. Same view you get inside the workbench.

// ReadFromMergeTree · events

Scans 96 parts, 12,480 granules, 214M rows. The primary key (event_date, user_id) only prunes by date — tenant_id can't skip granules. A projection keyed by tenant_id would.

-- events · prod-analytics-1
SELECT
  toDate(event_time) AS day,
  uniqExact(user_id) AS dau,
  countIf(event = 'purchase') AS purchases
FROM events
WHERE event_time >= '2026-06-01'
  AND tenant_id = 42
GROUP BY day
ORDER BY purchases DESC
LIMIT 20;

-- 3.8s · 214M rows read · 96 parts
-- ✓ Proposal: projection keyed by tenant_id

ALTER TABLE events
  ADD PROJECTION by_tenant
  (SELECT * ORDER BY (tenant_id, event_time));

ClickHouse surface

What Mongrel ships for ClickHouse.

SQL editor & execution

Run ClickHouse SQL with result paging, FORMAT variants, and per-query settings like max_execution_time.

EXPLAIN PIPELINE

Visualize the query pipeline — transforms, parts, granules, rows read. See exactly where a query spends its scan.

Table & parts browsing

Browse tables with engine, ORDER BY key, partition, and row counts. Drill into system.parts to see merges and part sizes.

Schema introspection

Inspect engines, codecs, projections, materialized views, TTLs, and indexes. Full CREATE TABLE DDL one click away.

Batch inserts

Stream CSV, JSONEachRow, and Parquet files straight into MergeTree tables with batched, resumable inserts.

PK-aware row editing

Edit rows in place with primary-key-aware updates. Mongrel tracks lightweight deletes and mutations for you.

AI explain & rewrite

Paste a query and its pipeline. The assistant proposes ORDER BY key changes and projections, with the predicted plan.

Users, roles & quotas

Browse users, roles, row policies, and quotas. Profiles, settings, and server activity from system tables.

Import, export, copy

CSV, JSON, and Parquet in and out. Copy data between ClickHouse and any other engine — schema-aware.

Connect ClickHouse in 30 seconds.

7-day trial. macOS, Windows, Linux. No credit card.

Download Mongrel