Back Up Or Archive?

With the increased popularity and wider usage of file-based cameras within the broadcast Industry, new responsibilities have emerged for this valuable media content. Gone are the days of being able to resort to a Digibeta tape on a shelf if the media was unfortunately lost or deleted.
Similarly, as picture quality and frame sizes have increased, so have media file sizes. Larger storage devices, typically of the shared storage variety, are required to store the ever-increasing required capacity during post production. And, while storage is readily available, online storage still carries a cost premium.
We are also in a period where even more media is expected to be kept online or at least nearline. The traditional offline-then-online workflow in post is being replaced, with users preferring to work with higher or even full resolution media from the start to the finish of projects. We are even fast approaching the point where re-capturing / re-importing isn’t acceptable!
Here’s the issue. With only a single file copy of the raw rushes from a file-based camera and your shared storage rapidly filling with media you also need to access, what should you do? Backup and archive? Just archive? Just backup? And what’s the difference between them? In essence they are both very similar, the key difference is the lifespan of each.
Typically backups have a short retention period aimed at providing a copy of the original file that can be restored after possible data loss. The secondary purpose of backups is to recover data from an earlier time; this would tend to work in conjunction with an incremental/ snapshot policy. For example, to complete a full backup of all data on a Saturday night, then on each subsequent night, look for changes and backup these each day. This would allow data to be restored for up to seven days and reduce backup times. Backups also often represent a simple form of disaster recovery, the complete backup written on the Saturday night might be written to duplicate LTO tapes, one of which is taken offsite.
Unlike backup, an archive typically has a long retention period and they are employed for a variety of reasons, both commercial and operational.
A heritage archive, sometimes referred to as a deep archive, keeps the content for reuse at a later date. The BBC has a huge heritage archive with its own preservation department in control of the transformation from a physical archive of film and cassette to a file-based archive. Another reason for archiving is compliance, where copies are retained in accordance with established guidelines or specifications. Again, this would tend to be for a long retention period but after time, it may possibly be removed. A further and commonly used form of archiving is for economic reasons; moving content into an archive to free up online storage capacity, while keeping it available for fast and easy retrieval.
Archiving also provides an important opportunity to add additional metadata to media, for example by populating customised metadata tags with extracted information. This metadata will help when searching the archive to locate a single asset or group of assets meeting the search criteria. More often than not, the archive works in conjunction with an asset management system, providing decisions surrounding the annotation, cataloguing, storage, and archiving of the assets.
Archiving and backup systems can have two distinct and complementary functions within a media organisation. These two principles can be applied together to create an effective workflow addressing the original concern to protect valuable content, along with the need to maintain appropriate levels of online shared storage while maintaining accessibility to the current media and legacy material.
In such a workflow the camera files would be backed up on arrival with one file copy for the backup, written to two LTO tapes, one being taken offsite for disaster recovery. A second independent file copy would be made to the shared storage ready for editing. It’s important to make independent copies so any corruption wouldn’t be transferred to the shared storage from the backup. The project would be edited on the online shared storage as normal, but once completed it would then be archived to less expensive nearline storage and also to LTO tape, where again, a duplicate could be used to provide disaster recovery.
The completed project could remain on nearline storage for a fixed period to insure that if it was required within that period it could be restored quickly; after that time it would then be restored from LTO tape. With the archive being searchable, historic projects and media would be straightforward to locate and quickly resorted.
Once the project is archived, the LTO tapes which were used to provide the original backup for editing can be recycled.
www.root6.com
viewer.zmags.com/publication/381c52e3#/381c52e3/1