ClickHouse Blog

In financial services, market data systems typically split into two tiers: real-time and historical.

We've released @clickhouse/rowbinary, a Node.js reader and writer for ClickHouse's RowBinary, RowBinaryWithNames, and RowBinaryWithNamesAndTypes formats. You can…

What are huge pages, and why do they matter for Postgres? The OS hands out memory in 4KB pages, and the CPU keeps a small cache of virtual-to-physical translations, the TLB. A huge page is the same memory in much bigger units, 2MB or 1GB, so one page-table entry covers up to…

Summary

TL;DR

ClickHouse released pg_re2, a Postgres extension that replaces POSIX regex with RE2-powered functions, delivering 1.8× to 8.6× performance gains on regex operations and 1.1× to 1.8× speedup on indexed queries. The extension also solves syntax incompatibilities between Postgres POSIX regex and ClickHouse RE2, enabling correct pushdown of regex expressions in the pg_clickhouse foreign data wrapper.

ClickHouse describes why Postgres and ClickHouse are becoming the default combination for AI applications, which demand real-time transactions and analytics on massive data volumes. The post covers AI's data-growth demands, the collapse of traditional OLTP/OLAP boundaries, and ClickHouse's investments in tools like PeerDB/ClickPipes for CDC, pg_clickhouse for cross-database querying, and pg_stat_ch for observability.

Article explores how chDB (ClickHouse embedded) can serve as a local data engine for AI agents, reducing latency and instability from network calls by embedding a full query engine in the agent process. Covers three use cases: agent memory using append-only MergeTree with vector search, chDB as a federation hub, and token/cost savings from eliminating network failures and retry loops.

ClickStack now generates dashboards and connected investigative workflows from natural language prompts using an AI Notebook that explores telemetry, validates assumptions, and explains its reasoning. The same MCP toolchain powers both ClickStack's built-in AI agent and external agents like Claude, Cursor, and Codex, allowing teams to create and refine dashboards through their preferred tools.

TLDR;

TL;DR:

Another month goes by, which means it's time for another release!

PgBouncer is single-threaded. A single process uses one CPU core, no matter how many the machine has. On a 16-vCPU box that means one core does all the connection pooling while the other fifteen sit idle, and the pooler starts capping throughput long before Postgres runs out of…

LogHouse, our internal logging platform for ClickHouse Cloud, now stores 431 PiB of uncompressed data across 1.59 quadrillion rows. It spans 30+ regions across three cloud providers, and a typical query against any of them comes back in a few hundred milliseconds.

At ClickHouse, we love testing how our products perform across a variety of datasets and benchmarks. We strongly believe that benchmarking products in a transparent and reproducible manner is key to providing quality information to end users, and that it fosters a fair and…

We've had decades to learn what it means to design software and hardware for humans. Did you know the first prototype of a mouse was made of wood, and had a single little red button in one corner? It was terrible. But since then (1964!), we've come a long, long way. The "mouse"…

This is a guest post from Tsvetan Stoychev, ex-Principal Software Engineer at Akamai Technologies.

Postgres backups are one of those pieces of infrastructure that should be boring. They sit in the background, continuously archiving WAL files, uploading backups, and making sure that when something goes wrong, recovery is possible.

Summary

TL;DR