Yesterday, I discussed a key challenge facing today’s IT professional when it comes to their network infrastructure; a lack of agility. Today, I want to discuss another critical challenge—a lack of flexibility.
Here’s the thing, something that as an IT professional you probably already know: Workloads tend to be sticky. Essentially, you deploy a workload and it gets somewhat stuck to the server that it is deployed on, simply because there are some dependencies on the physical network that the workload may have assumed. But what if you want to move a workload running on one server to another server, one rack to another rack, or a workload from your data center into the cloud? With the workload stuck to the server, those tasks might be easier said than done.
What you, like most other IT professionals, want and need is the flexibility to control your workloads the best way you deem possible. But how do you get that flexibility? One way is by adopting Software Defined Networking (SDN) in the data center, preferably using a SDN solution that’s built on a host-based model.
With a host-based model, every entity in the network platform must be programmable. So, for example, if the entity in question is an access policy rule, it is transformed into a specific set of instructions, which are pushed out by a centralized controller down to the hosts, and the policy then gets enforced on the hosts. Such an approach is a lot more scalable than traditional approaches. The model allows network operators to compose a uniform SDN overlaid on an existing architecture. They can also utilize other SDN products or build something new without having to undertake a substantial network upgrade.
While SDN and the host-based model can most certainly deliver greater flexibility, giving IT professionals absolute control over where they instantiate their workloads requires a true hybrid platform. The platform must comprise both a public and private cloud component, each sharing the same underlying infrastructure. Such a platform gives you the flexibility to choose whether you want a workload on premises in your racks or in the cloud. Or you may opt to have pieces living in either of these places. You could even decide to put the entire workload in the cloud, but then bring it all back on premises due to compliance policies—everything just continues to work because the underlying infrastructure is the same across the cloud and on premises offering. With a hybrid platform, all of these scenarios are possible.
The key point here is that by choosing a SDN solution from a vendor that offers both a public cloud and on premises offering, you don’t have to have a 10-year strategy around what your workloads will look like. That’s because the solution will give you the flexibility and therefore, absolute control, to run those workloads any way you want. It’s the cloud on your terms and it’s something you’ll want to be sure to get from the SDN solution vendor you choose—if flexibility is one of your top priorities.
Be sure and check out my next blog for information on another networking pain point that can be overcome by using SDN in the data center. And, don’t forget to check here for additional blog posts on a range of other IT-related issues.
This blog is sponsored by Microsoft.
Cheryl J. Ajluni is a freelance writer and editor based in California. She is the former Editor-in-Chief of Wireless Systems Design and served as the EDA/Advanced Technology editor for Electronic Design for over 10 years. She is also a published book author and patented engineer. Her work regularly appears in print and online publications. Contact her at [email protected] with your comments or story ideas.