As IT Pros, we serve the business; the job is to provide the systems and services that let the business operate and grow. So we have to grow those systems as fast as the business grows. If we cannot, then the business slows to run at our pace, which is definitely not good.
Although our data backup systems seem like a pretty basic part of our IT infrastructure, they are connected to -- and critical -- for nearly every production system. A new production system probably cannot go online without the backup system in place. Normally, this is no problem. Adding a new server, or VM, or application, or even cloud service to your backup should be a simple task. But what happens when the backup system is out of capacity? We just add more, right?
It turns out that many backup systems depend on hardware appliances for efficient storage of the back-end data, and those appliances typically cannot be expanded. When it is time to grow, we have to throw away the old appliance and buy a newer, bigger one. We call this a 'forklift event.' This is often expensive, and becomes a time-consuming project with meetings, decisions, justifications and technical risk.
Of course, in IT, we upgrade systems all the time, so this is not a completely shocking experience. But this is not a case of upgrading a single server to support more capacity of a single application. Your backup system touches everything, and new production projects cannot be released without it. So you can be faced with a relatively small production project that forces a very large, very expensive upgrade, which is only partially related to the production project. Imagine a small web application being delayed while you justify and manage a several month, $500,000 project to upgrade the backup system! This is proverbially the straw that breaks the camel's back.
A particularly bad way to solve this problem is to just build another or multiple, isolated backup solution (i.e. one for physical, one for virtual, one for databases). Not only will you end up with capacity stranded in different silos, but you will also have trouble finding your data when you need it. The right solution is to be sure your backup system has the native capacity to cover all your systems and to scale-out, without depending on an appliance that can only scale-up.
Commvault's solution to this problem has four elements:
- We use commodity hardware, so you don't get locked into an expensive hardware vendor with limited configurations.
- Our deduplication capability works across multiple servers, so you can grow without a 'forklift event.'
- Multiple servers require load-balancing, but doing that manually is error prone and time consuming, so Commvault software does that automatically.
- At scale, with many servers and many disk drives, something will always be broken. Commvault software can self-heal to work around failures, so a single component outage doesn't cause a system outage.
If you have a story about when you found it hard to scale your backup system, leave a comment - we'd love to hear it. And if you want help designing a system that grows the way your business does, ask your reseller about Commvault.