Data Retention Policy
A data retention policy is a rule that defines how long data is stored before it is deleted or moved to cheaper storage. It balances the cost of keeping data against its usefulness and any compliance requirements.
Balancing cost, compliance, and usefulness
Every system that collects data eventually faces the retention question. Keep everything forever and storage costs climb. Delete too aggressively and you lose the history you need for trends, audits, or debugging an incident that started weeks ago.
Retention policies make that tradeoff explicit. A common pattern is to keep recent data at full resolution, downsample older data, and eventually delete or archive the oldest. Compliance often forces a floor: financial and healthcare data may legally have to be kept for years.
The hidden trap is when cost, not policy, drives retention. Many teams quietly cut retention just to control their database bill, then get caught without the history they needed.
How Arc handles Data Retention Policy
Arc lets you set retention by data type and keeps data cheap enough that you rarely have to choose between cost and history. Because storage is compressed Parquet on object storage, keeping more history costs object-storage prices, not premium-database prices.
Arc is a high-performance columnar database. Open Parquet on storage you own, single Go binary, production-ready in 30 seconds.