Storage Durability Dilemmas: Replacing EBS, Ephemeral vs Durable, and Practical Backup Strategies

Storage durability just got debated in public threads. The headline grabber: replace EBS and rethink Postgres storage from first principles, embracing fluid, forkable storage that separates ephemeral from durable. ^[1]

Fluid storage and durable fundamentals A lively thread argues for moving beyond EBS-based Postgres storage and toward forkable, portable storage that survives across clouds and data centers. The idea is to trade lock-in for flexibility, aiming for durability without paying for idle capacity. That means letting ephemeral components come and go while the core data stays protected. Advocates say portability lets you test disaster recovery without separate silos. They also cite examples where swapping storage backends doesn’t force a Postgres rewrite. The debate weighs cost against risk, asking teams how far they’ll go to prevent vendor-specific drift. ^[1]

Practical backups: dump files and restic On the practical side, POST 2 points to backup approaches that you can actually move around: dump files and the backup tool restic.

• dump files - restic That approach aims to deliver reproducible backups usable across on-prem, cloud, or edge deployments. ^[2]

The takeaway: durability is a balance between cost, portability, and disaster-ready backups you can trust anywhere. Watch how teams test disaster recovery across environments next.

References