Storage - Identifier and Database
Introduction
In order to track the exact location of each accession in the future, we need to mark each tape and logical volume, and those data have to be stored in a database.
This section has two parts: identifier assignment and database storage.
Identifier Assignment
Tape
Each tape already has its own identifier starting in FF0001. There is no need to change its label or add another layer of mapping.
Logical Volume
We can name the volume whatever we want, so I chose to make it similar to the tape. Each volume is presumably 2.25TB and a 4-digit long identifier could provide up to ~22PB of total space available. There is no need to make it longer. In addition, I plan to add "A" or "B" as the prefix, indicating the compression server that generates the volume.
Database Storage
Each accession can belong to only one logical volume, so we can add a column in the current database for the volume information. The volume, on the other hand, might be stored in multiple tapes, and we need an additional table to map this relationship.
