These days it seems that all someone has to do is use the word “deduplication” in conjunction with a data protection product and that data protection product magically looks “better”. But what organizations have to be careful to do is not allow deduplication to color their view of what they hope to accomplish with the implementation of disk-based data protection. Rather organizations need to look at data protection from a different viewpoint that it is not tainted by deduplication and allows them to fully leverage the flexibility that disk-based backup provides.
I agree for the most part that deduplication makes current data protection processes better. Deduplication facilitates the introduction of disk into their current backup process by making disk economical to deploy as data reduction ratios of 20:1 or greater can be achieved.
This 20:1 data reduction ratio is critical since this allows disk to cross the cost of tape threshold so organizations can now cost justify the deployment of disk. As a result, the other advantages of disk come into play: faster backups and restores; backups and restores that complete with near 100% success rates; and, significant decreases in the amount of time that organizations spend troubleshooting failed backup jobs.
But it is at this point that organizations need to take off their deduplication colored glasses and ask the following question regarding deduplication: “What does deduplication change about current backup processes?”
Not much else aside from what has already been mentioned above. Backup is still a point in time process that generates server, storage, and network resource spikes when it is run. Depending on where it is run (source or target), deduplication may make things better than they are today, but as data sets continue to grow, you’ll eventually run into the same problems with backup windows, RPOs, and resource spikes.
Many vendors are already talking about the growing need for more “continuous” data protection approaches. One of the big benefits to these types of technologies, which include replication and continuous data protection (CDP), is that they significantly reduce resource spikes by spreading the capture of daily changes out across the entire day.
Data is expected to grow at rates of 50% – 60% a year or more for most enterprises, so this is a salient point to consider when developing backup strategies. Data growth rates make the use of deduplication technology a tactical, not a strategic, play. It is for reasons like this that it behooves organizations to look at disk-based data protection from a totally different viewpoint than one filled with deduplication’s glow.
But you do still have to consider storage hardware costs when deploying disk as a backup target, so here is a perspective to consider. Suppose one did not have to deduplicate data but could get 90 – 100% of the data reduction benefits and storage hardware cost savings that data deduplication provides, the same near-100% backup and recovery success rates, the same decreases in time managing backups and the elimination of backup windows.
In addition, one would also get a number of other benefits of critical importance in data protection:
- Near real time application recoveries performed locally or remotely
- Create consistent copies of application data that can be used as source files for file or folder restores or backup to tape or the cloud for long term data retention without imposing any overhead on production applications
- Non-disruptive testing software upgrades to first see how they might impact production application servers
- Verify that disaster recoveries and application fail overs work as intended
- Recover production applications to either virtual or physical machines
Disk-based data protection software from providers like InMage makes this change in viewpoint possible. Not only does it deliver all of these new benefits, it consumes approximately the same storage capacity as data deduplication.
A previous blog that I wrote on this subject illustrates this point. When the same set of production data is protected over a 30 day period is in one case protected using traditional backup software with data deduplication and in another case protected using InMage’s CDP software, data deduplication is at best 10 – 20% more efficient than InMage CDP. Considering the declining price of disk and the powerful new benefits that InMage CDP affords, that is a small price to pay.
Data deduplication is a powerful new technology but many organizations are being blinded by the relatively few recovery benefits that it provides. This is causing them to miss out on the real opportunities that disk-based data protection technologies like InMage can provide.
InMage CDP puts the world of disk-based data protection in a totally different light as it changes more than just how organizations should view the protection and recovery of their production data. It should open their eyes to the entirely new possibilities that storing their backup copies of their production data on disk create.