Maintaining backup health amid constant change

27.06.2006
Backup is degrading. No, I'm not trying to imply that there is something ignoble or shameful about the activity -- some of my best friends are backup administrators! I mean that in an active environment, backup is in a continual state of change, and a system that may have been performing optimally three to six months ago is likely to be doing substantially less so today. As servers and applications are added to an environment and data on primary storage grows, backup is affected exponentially, and without focused attention, the backup environment will inevitably degrade.

The decay can be gradual, and therefore may not be noticed when administrators are focused on the day-to-day operational aspects of backup, particularly when lacking comprehensive backup reporting to measure trending. As a result, small problems can go unnoticed until they become large. For this reason, it is important to develop a regular routine -- a backup health calendar -- to check various aspects of the backup environment's health at appropriate intervals -- daily, weekly, monthly, quarterly and annually.

Here are a few examples of each:

Daily

-- Review successful and unsuccessful backups.

-- Resolve failures before the next backup window.