There is an increasing need for IT to be able to support both traditional applications and well as emerging ones. This phenomenon is sometimes referred to as “bimodal IT” which is just a way of saying that you need to be able to support those business-critical legacy applications while also making it really easy to build new, innovative systems. Whereas the public cloud is often considered to satisfy the second part of the bimodal equation, what if you could bring the necessary cloud-like characteristics to your local data center instead?
That’s part of what you get with composable infrastructure but you also get so much more. The primary characteristics and benefits that define composable infrastructure are defined in the following sections.
Click here to download our free ebook on What’s Next for Enterprise IT Architectures and Solutions
Fluid Resource Pools
The real secret sauce, and probably the most important characteristics, behind composable infrastructure is its support for what is often referred to as disaggregated resource control. This term basically means that you have the ability to independently scale resources at will. If you need more storage capacity, you simply add more disks to the system. The system is architected in such a way to as to make it dead simple to scale in this way without having to incur downtime or complex expansion procedures. Sure, you’ve been able to scale resources independently in the traditional world forever, but those resources were always part of the same management plane. When considering composable infrastructure, you’re also looking at ways to far more seamlessly deploy and manage all of the various resource areas.
This fluidity makes it very easy to granularly target resource deployment where you need it most. You’re not forced to live within predefined limits and resource ratios established by a vendor as you are with hyperconverged infrastructure, for example. The world is yours!
If you’re like every other business on the planet, you want to grow over time. That means that your IT systems have to be able to scale in lockstep with your business. However, that doesn’t mean that today’s scaling options are always particularly good. At times, it can be downright difficult to grow resources the way you need and linear scaling available in hyperconverged infrastructure isn’t always the most desirable choice for businesses.
With composable infrastructure, scaling is both simple and granular. You can scale any resource you need individually by simply adding more of that resource and the management system seamlessly assimilates those resources into the collective whole and makes them available for use.
You may have heard and, if you haven’t, will certainly start hearing the term “infrastructure as code” bandied about. What this really means is that there is a need to more tightly couple applications and the infrastructure on which those applications reside. There are a number of outcomes that can be realized from treating infrastructure as code, such as:
- Enables developers to programmatically deploy new virtual machines and other structures so that they can more quickly test their code
- Allows an application to automatically instantiate new infrastructure based on performance conditions at the present time
- Enables automation-based user self-service, which is a prerequisite to be considered a private cloud
You should expect to see deeper and deeper integration between applications and infrastructure over time. There is no reason that applications can’t be far more self-managing than they are today. With the right tools and the right infrastructure, admins can be brought to bear on an exception-only basis.
It’s true that “the cloud” offers a compelling model for IT to use as we move into the future. It’s not true, however, the “the cloud” has to happen outside of your data center. Some of the characteristics that major cloud providers have been engineering into their data center operations for years are just what was described above. There’s no reason this same model can’t be replicated in private data centers all over the world to begin to realize cloud-like benefits with on-premises economics and control.