You are spending too much money on your backup infrastructure. You've built this massive infrastructure specifically to handle one task...The Full Backup. Most enterprises do their entire full backup job over the weekend so they have enough backup-window time to get the job done. We've been doing backups this way for the last 20 years, is it time for change? Could changing it rein in the costs of the backup process?Why do full backups in the first place? Historically, full backups were a direct result of living in a tape-based world. You did full backups as often as you could to lessen the amount of tape mounts that were required to recover a server or a data center. Now, however, most data centers don't exist solely in this tape based world for data protection. Disk has augmented the processes and there should be less need to do full backups as often. There are no mount time concerns with disk and with disk targets that have deduplication capabilities like those from Data Domain or Exagrid and it is cost effective to store weeks, if not months worth, of backups on those.

The real issue with full backups is the infrastructure that you must build to handle a huge data load that only happens once a week. The rest of the week that infrastructure typically goes heavily underutilized. Especially in the cost consciousness of the next few years, continuing to invest in an infrastructure to handle a once a week event isn't as popular as one server to one application hardware deployments. Just as virtualization changed the server world, something has to change in the backup world.

So what can you do about full backups? Obviously, an inexpensive step is to leverage your existing backup applications scheduler to do less full backups. Maybe instead of once a week, complete it once a month or once a quarter and then count on the backup disk to do its job. That said, many customers are uncomfortable going that long without a full backup or don't have the time to plan a sophisticated backup schedule like this. Backup applications like Zmanda's Amanda can analyze the backup schedule to spread the full backup load across the week automatically without constant fine tuning from the backup administrator.

Beyond modifying the existing schedule you can change the way the backup is done in its entirety by using technologies that reduce the amount of data going across the network all the time, like the deduplication technique used by Avamar or the snapshot-style backup approach that SyncSort's Backup Express uses.

While both of these techniques would require the purchase of additional backup software, thinning data at its source can reduce the investment in the backup infrastructure, which may offset those costs.

Whether it is taking advantage of disk-based technologies and advanced scheduling or changing the backup process itself, the ability to reduce the size of a full backup can significantly reduce the ongoing investment in the backup infrastructure.

Track us on Twitter: http://twitter.com/storageswiss.

Subscribe to our RSS feed.

George Crump is founder of Storage Switzerland, an analyst firm focused on the virtualization and storage marketplaces. It provides strategic consulting and analysis to storage users, suppliers, and integrators. An industry veteran of more than 25 years, Crump has held engineering and sales positions at various IT industry manufacturers and integrators. Prior to Storage Switzerland, he was CTO at one of the nation's largest integrators.