Altinity Blog

Most consoles run on a separate server beside your database. Altinity's SQL Browser's architecture starts from a different premise: what if the console was just a file?

Scarf provides usage statistics for open source software. When their data volume grew by more than 200X, they worked with Altinity to get their costs in line.

MindWise's o_rabbit project writes ingested data as Parquet files in S3, registered as Iceberg tables, and queried with Project Antalya swarm clusters.

AI coding agents can act autonomously. That's powerful. And terrifying. See how to use OAuth, Altinity's MCP server, and the RBAC tools inClickHouse® make that safe.

The Altinity.Stable 26.3 Build has a lot of new features and improvements, some of which will prevent you from reverting back to 25.8 if needed. See how to upgrade safely with a possible downgrade path.

Saving money is one of our favorite ways to help ClickHouse® users at Altinity. Robert Hodges walks through six best practices that delight your accountants while keeping ClickHouse fast, from open source and schema tuning to TTLs, tiered storage, compute...

The story of how ClickHouse Keeper caught up to ZooKeeper performance with a 2x speedup in 26.6, strengthening its case for ClickHouse replication.

The BYOC model lets you run analytics databases in your cloud account. You control where your data and compute resources live, you control backups and storage, but Altinity handles the infrastructure for you.

Can a single ClickHouse BYOC node stay responsive across 3 billion rows? Our ClickHouse Star Schema Benchmark measured selective query latency as low as 34 ms.

Nirvana Labs (Nirvana), a high performance storage cloud, is announcing a strategic partnership between Nirvana and Altinity, the largest independent managed service provider for ClickHouse®. The integration is now generally available as a Bring Your Own Cloud (BYOC) deployment on...

Using materialized views to process data from Kafka into ClickHouse® is a common architecture. See how changing where a materialized view wrote its rows reduced ingestion lag by more than 50%.

Every ClickHouse® schema works fine at 10GB. Most fall apart somewhere between 1TB and 100TB, and rarely for the reason you would guess. Altinity's Josh Lee and Last9's Prathamesh Sontpatki walk through the real schema and architecture decisions behind a...