After reading the blog “Is RAID/Replication/Erasure Coding Backup?” by Curtis Preston, it got me thinking about the benefits of backup over RAID, erasure coding and replication. Preston discusses the concept of history and how backup allows you to ‘back up time.’ RAID, erasure coding and replication alone allow you to replicate data to other locations, or provide redundancy, in case you need to access the data during a hardware failure. It doesn’t provide you with historical information that helps you recover to a specific time in the past. It can be of benefit, but it does not have the intelligence to provide an automated restore or recovery process. A good backup solution should catalog or index each backup, making recovery intelligent, easy and useful. You want to be able to recover the point in time you want, not just access what might be available.
Similar to Marty McFly accidently sending the DeLorean time machine to some random time and date in “Back to the Future,” imagine you were stuck racing against the clock, trying to figure out where you were and how to get back – versus simply having a cool car with the ability to set the dial and travel to a set point in time of your choosing to prevent disaster or rewrite history. To draw an analogy, with RAID, erasure coding and replication, you are at the mercy of the DeLorean, crossing your fingers that the car’s controls don’t break down with only the DeLorean knowing where and when you went.
Some backup solutions can automatically index the data as it is being backed up, bringing some intelligence to the process and making it easier to search for the right point in time. There are other solutions that catalog or index the data but house the catalog in one location, making it a single point of failure. If that server goes down, where does my intelligent, quick and easy recovery go?
The purpose of using RAID, erasure coding and replication is to build in redundancy so you don’t have a single point of failure. If those three aren’t providing you the benefits of backup, but your backup solution has a single point of failure, you aren’t truly better off. You want a backup solution that distributes the index across the architecture, leaving no single points of failure. If the one server goes down, you still have the ability to quickly search and find your specific point in time for recovery, without blindly restoring multiple backups to find the right one.
Commvault provides an intelligent backup process leveraging a distributed index, which removes any single point of failure, and allows you to easily find which recovery point you need. Commvault also gives you control of your backup and recovery process through policy management. Create backup policies by data type or recovery profile, which allows you to customize your solution to fit your business needs and data sources. With Commvault, you can customize your DeLorean’s console, providing easy-to-use controls to efficiently search and automate your recovery, and make sure you arrive at the right time in the past to prevent future disruptions.
So Preston, imagine if we could take you back in time just before your first big restore – imagine how much easier and less nerve-wracking it would be to leverage today’s modern backup system instead of SM-Arch. That nervous energy that you felt as the rest of the IT guys watched you restoring a few gigs of data would be replaced by simple confidence! And then your Wayne’s World dissolve could go to a Madonna moment, instead of an IT moment … Party On!