How can software-defined storage be used in a private cloud environment?
Software-defined storage can be a physical server-based appliance with storage management software installed on it, functioning very much like an array, but really a white-box type of implementation. But it can also be storage management software installed inside a virtual machine (VM) where the VM controls all the storage operations. In either case, the server is functioning like a storage controller.
The best use cases for software-defined storage are places where you need rapid deployment of new implementations of relatively simple systems. Some examples, for instance, are a test and development environment, a sandbox or a proof of concept -- relatively simple applications where storage performance is not that critical. One of the reasons stems from comparing a software-defined environment to a purpose-built array. We went through a similar battle back in the early 2000s where we had general-purpose storage systems based on Wintel computers, and the result was that there was no way they could compete with a purpose-built ASIC, and therefore we saw them fade into the background. Now they're coming back in the form of software-defined storage. But if you need high performance, there's not going to be any substitute for an array.
But in a private cloud environment, where users want rapid deployment and to be able to spin up a machine very quickly -- that's where it might have application. Again, one of the downsides is that you now have a second storage silo to manage, and IT organizations have been doing everything they can to consolidate those silos and simplify management. Therefore, they need to be cautious about that implementation and make sure the benefits are worth the costs.
About the author:
Phil Goodwin is a storage consultant and frequent TechTarget contributor.
This was first published in February 2014