Main definition
Data tiering is the practice of automatically moving data between storage types — such as high-performance flash, standard disk, and low-cost archive or cloud storage based on how frequently it’s accessed and how critical performance is to its use. Tiering ensures active data stays fast and accessible while inactive data moves to cheaper storage, without users needing to manually manage file locations. Tiering frees up expensive primary storage for ongoing work, but also allows data that has been moved to a cooler storage to be recalled easily.
A large majority of stored data is rarely or never accessed after a short window, making data tiering an attractive solution for reducing storage spend.
Starfish’s tiering capabilities are policy-driven and storage-neutral, meaning tiering rules can move data from any connected primary storage — NAS, Lustre, or otherwise — to a wide range of destination tiers, including object storage and cloud archive targets. Because Starfish maintains metadata about where files have moved, users and applications can continue to find and, where needed, retrieve tiered data without IT having to manually track file locations.
Related Links
- Modern Data Storage Infrastructure Software for AI & HPC | Vdura
- Tiered Data Storage Guide | Pure Storage
- Why Companies Should Use Data Tiering | NetApp
- Storage Management 101: Tiering is Necessary | DATAVERSITY
- Starfish Storage Solutions | Starfish
