ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
MergeTree performance for time-series data depends on partitioning and compression choices made at CREATE TABLE. This post walks through a practical schema for HTTP logs, shows how to tier ZSTD and LZ4 codecs, and explains how to verify partition pruning with EXPLAIN.
Align ClickHouse MergeTree PARTITION BY and ORDER BY with query patterns and retention to enable partition pruning and index skipping for time-series analytics.
ClickHouse MergeTree TTL automates data expiration and tiered storage without external jobs. Define TTL clauses at table creation to delete or move partitions based on a DateTime column, align partitioning for efficient drops, and monitor merges via system tables.
Guide to pick the right MergeTree engine (plain, replicated, aggregating, collapsing) based on replication, aggregation, and update‑handling needs for time‑series data.
In ClickHouse MergeTree tables, ORDER BY is the physical sort order of your data, not a query clause. Here's how the sorting key drives sparse-index pruning, with a worked multi-tenant example.
Learn how to diagnose and fix the 'Too many parts' (Exception 252) error in ClickHouse. This guide covers identifying merge backlogs, handling stuck mutations, and optimizing insert patterns.
How to configure ClickHouse TTL expressions on MergeTree tables to delete or relocate aged data, verify expiration through system tables, and recover from a misfiring policy.
Guide to configuring ClickHouse TTL on MergeTree tables for automatic row expiration, including steps, verification, and recovery.
Learn how to pick a ClickHouse MergeTree primary key that enables partition pruning and skip indexes, turning slow range scans into fast lookups.
clickhouse