How do you define Software-defined Storage?

(Originally posted May 1, 2013 on my domain blog at Reposted here for completeness as I move to

If you are like most of the clients I deal with, you are starting to recognize the storage part of your infrastructure represents a BIG opportunity for improvement in 2013 – in agility, in efficiency, and in cost. When demand (data growth) outpaces supply (ability of hardware vendors to increase areal density driving down costs) as dramatically as it has begun to do, something has to change in the way storage infrastructure is approached in order to help balance the equation again. That ‘change’ creates a perfect economic environment for vendor innovation resulting in creative new solutions for clients. If you have been paying attention to the storage space, you’ve noticed an increased investment pace as vendors explore technical innovations and try to explain these innovations to potential clients. One of my biggest frustrations though is when the industry can’t settle on terminology for describing a solution approach leaving clients thoroughly confused and paralyzed.

Think about how long it took us to settle down on the term ‘cloud’. Most folks felt like ‘cloud’ was going to help them, but it has taken quite a while for the industry at large to understand what exactly ‘cloud’ means and how to get there. Software-defined Storage (SDS) is another of those terms that holds great promise for IT managers, but is suffering from lack of definition. ESG analyst Mark Peters recently noted in an InformationWeek article  that “When new things emerge we love to give them names, but pretty soon the generic name can become an impediment to understanding the specific products and values that it’s meant to describe. Worse still, we all quickly reach a point where it’s even deemed embarrassing to profess to not understand something, so we genuinely get the blind leading the blind – or at minimum some level of blandness – or a mistaken idea that everything sporting the same name is necessarily similar”. SDS is such an important shift for clients in their quest to regain balance in the storage infrastructure that we all need to settle on terminology so clients can get on with serious evaluation of the available offerings.

Recently Ashish Nadkarni lead work by the insightful storage team at IDC to create a taxonomy for what they called Software-Based Storage. The report, IDC’s Worldwide Software-Based (Software-Defined) Storage Taxonomy, 2013 is a must read for clients who are serious about understanding the innovations taking place and how they can make use of it in their datacenters. It’s also a must-read for vendors and marketers who are trying to describe their offerings to potential clients. The report notes that “IDC believes that software-based storage will slowly but surely become a dominant part of every datacenter, either as a component of a software-defined datacenter or simply as a means to store data more efficiently and cost-effectively compared with traditional storage”. I agree – but am hopeful that settling on some definitions can help us accelerate the ‘slowly but surely’. Let’s take a quick look at IDC’s view.

Core component of a Software-defined Data Center (SDDC):  I like IDC’s definition of an SDDC, “a loosely coupled set of software components that seek to virtualize and federate datacenter-wide hardware resources such as storage, compute, and network resources and eventually virtualize facilities-centric resources as well. The goal for a software-defined datacenter is to tie together these various disparate resources in the datacenter and make the datacenter available in the form of an integrated service…” Most IT managers I work with have understood the economic and agility value of shifting to virtualized compute resources and treating their server hardware as a commodity that simply provides horsepower. This is software-based compute using tools like VMware vSphere, Microsoft Hyper-V or KVM. SDS expands the thought (and benefit) to the storage part of the infrastructure treating storage hardware as a commodity that provides capacity. In a conversation I had with Ashish, he noted that “Software-based data centers will fundamentally alter IT – and how it is run… Software-based storage is a core component of this metamorphosis! The time to start planning is now.”

IDC Software-defined Data Center
IDC Software-defined Data Center

Defining Software-defined Storage: According to IDC’s taxonomy, there are a set of key attributes that IT managers can look for in identifying software-based storage solutions.

First, a software-based storage solution is, well, software. It is designed to run on commodity hardware and leverage commodity persistent data resources. This is in contrast to most traditional storage systems that, while they may have software microcode at their core, depend on some custom ASIC, a specialized CPU, or a controller to perform some or all of their storage functions.

Second, in IDC’s view, a software-based storage solution offers a full suite of storage services – equivalent to traditional hardware systems. I have had clients ask “if all I get is equivalent function, what’s the point?” It’s a good question. Aside from the obvious cost benefit of being able to treat your storage hardware as a commodity, I might suggest that a ‘good’ software-based storage solution not only meets the equivalent functions of traditional physical systems, but also enables key capabilities across an entire datacenter worth of infrastructure that are simply not possible inside the confines of a physical system – things like data mobility, replication consistency, and tier management.

Third, a software-based storage solution federates physical storage capacity from multiple locations like internal disks, flash systems, other external storage systems, and soon from the cloud and cloud object platforms. In my next post, I plan to discussed one of the more powerful outcomes of this type of single software-based view on storage infrastructure.

IDC’s taxonomy goes further to flesh out distinguishing aspects of software-based storage solutions such as how data is organized (block, file, object), where data is persistently stored (internal, DAS, SAN, NAS, cloud, object), what sort of services are offered, and how the solution is packaged and delivered. In the area of packaging, I think IDC did a good job in observing that while software-based storage is software, clients don’t always have to consume it as a download or off a stack of DVD’s. Consuming SDS in an appliance form factor or as a cloud service is perfectly reasonable.

IDC Software-based Storage
IDC Software-based Storage

As I said before, I think IDC’s report is a well-done piece of work and might just be what the industry needs to settle on definitions so we can get on with transforming the way IT managers deal with storage infrastructure.

7 Replies to “How do you define Software-defined Storage?”

Join the conversation!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s