MergeTree Archives | Page 2 of 2

ClickHouse IOPS Troubleshooting: A Complete Guide to Diagnosing and Fixing Disk I/O Bottlenecks

NEEDS FEATURED IMAGE. ClickHouse IOPS troubleshooting guide for DBAs and SREs: diagnose disk I/O bottlenecks with ClickHouse system tables, verify at the OS layer with iostat and pidstat, fix merge storms and part explosion, tune caches, tier storage, and monitor IOPS continuously with Prometheus and Grafana.

[…]

ClickHouse Real-Time POS Analytics: Sub-Second Decisioning for QSR Enterprises

ChistaDATA Inc.

If you’ve spent any time working in QSR (Quick Service Restaurant) data infrastructure, you already know the problem. The need for ClickHouse real-time POS analytics has never been more urgent — millions of POS transactions […]

Data Reliability Engineering for ClickHouse: Principles and Practices

ChistaDATA Inc.

Data Reliability Engineering ClickHouse (DRE) is the discipline that keeps analytical data trustworthy at every stage of its lifecycle — ingestion, transformation, storage, and serving. For teams running ClickHouse as their analytical backbone, DRE is […]

ClickHouse Query Optimization: 8 Indexing Strategies That Actually Work

ChistaDATA Inc.

The ClickHouse indexing strategies that actually move query latency — sorting key, sparse primary index, data-skipping indexes, projections and materialized views — with runnable SQL. Need it faster now? ChistaDATA delivers expert ClickHouse performance tuning, consulting and 24×7 support.

[…]

ClickHouse Primary Index Design: Parts, Partitions and Granules

ChistaDATA Inc.

A practitioner’s deep dive into MergeTree internals — parts, partitions, granules, and primary index design — from the ChistaDATA engineering team.

[…]

How MergeTree Internals Drive Query Latency

ChistaDATA Inc.

How MergeTree Internals Drive Query Latency: Parts, Granules, and Mark Files ClickHouse’s MergeTree engine stands as one of the most sophisticated columnar database architectures, designed specifically for high-performance analytical workloads. Understanding how its internal components—parts, […]

Why ClickHouse is so fast — MergeTree read path and query execution design

Why ClickHouse Is So Fast: 7 Design Choices, Measured on 1 Billion Rows

ChistaDATA Inc.

ChistaDATA · ClickHouse internals, measured on 1,000,000,000 rows · re-checked on 26.8 LTS Why ClickHouse is so fast is best answered with a number: on a two-core machine with 7 GB of memory, it averaged […]

Columnar databases vs row-based databases: bytes read for a one-billion-row aggregate, 80.3 GB vs 1.03 GB

Columnar Databases vs Row-Based Databases: Measured on 1 Billion Rows

Shiv Iyer

Columnar databases · Row-based databases · Lab-measured at one billion rows Columnar databases answered a one-billion-row aggregate by reading 1.03 GB. A row store would have read 80 GB. Columnar databases store each column in […]