Scaling out storage is all the rage these days, largely because customers have grown tired of having to pay for storage they are not using in advance.
The issue that many IT organizations don’t understand, says Scale Computing CEO Jeff Ready, is how dynamic some of the so-called scale-out storage architectures may or may not actually be.
Ready says many storage vendors are trying to embrace the term “scale out” from a marketing perspective, but from a technical perspective, the solutions they offer don’t dynamically discover new storage nodes as they are added to the environment, and more importantly, automatically stripe data across the new and existing storage nodes.
The way storage is acquired and managed in distributed computing environments is fundamentally changing in the era of the cloud. Gone are the days when IT organizations had to make multi-year commitments to buying storage. Instead, they can now just add storage capacity when they actually need it, as in the case of Scale Computing’s storage products, 1TB at a time.
To accomplish that, however, Ready says that storage architecture needs to support what is known as wide striping, which allows data to be dynamically stored in multiple disks.
Some might argue that the shift to scale-out storage is an event that is long overdue. Regardless, it’s becoming clear that way IT organizations provision storage is never going to be the same, which for most IT organizations is going to be a very good thing.