Cookbook for Successful Cloud Adoption
There is a fundamental change in thinking that’s needed if you want to succesfully adopt the cloud. What is it and how do you make it happen? Let’s find out…
In my last blog, I talked about cloud being the forcing function to break down silos in IT. It’s an aspiration for any organization—and I speak from experience. To reiterate, cloud agility and cloud workflow will leave you in the dust if you are not ready. The business units have the ability to completely circumvent traditional IT when it comes to adopting cloud workflows. So, it’s keep up or die. The good news is that technology and ease of use in the form of SD-WAN is finally here. The days of banging away on keyboards to make the change are gone.
And while simplicity of SD-WAN is a welcome relief, there are things you need to be aware of to successfully, and more importantly, to seamlessly adopt true cloud workflows. I’m talking about a fundamental change in IT thinking process. Let’s face it, the concept of IP address is dead. In the world of cloud, the application name or function is what’s important. For example, let Office 365 traffic go direct to Internet. But let’s keep SAP traffic on MPLS. Oh, and let’s throw in ServiceNow which should go direct to Internet, but in-house call manager traffic should take Internet via VPN to the data center. It all sounds simple, but traditional tools make it almost impossible to deal with. It’s one thing to engineer it, it’s another matter to operate it on a day to day basis. SD-WAN takes care of the steering of traffic based on application. It’s laughably easy if your SD-WAN solution can route applications and not packets.
So steering has been taken care of. The next thing to address is visibility and dealing with latency. After all, the latency to your own data center is deterministic. But to the cloud, you lose that deterministic path. How many calls do you want to take as people complain about this or that application is slower. How does one fight anecdotal figures? Without proper visibility, you will be fighting fires all day, all night. With zero chance of finding root cause. So are you ready to play Whack-A-Mole every day, every hour? If you do not have end-to-end visibility into the cloud, and from the cloud, you will be answering angry phone calls and emails. It behooves you to arm yourself with the right tools to guarantee performance. But more importantly, to measure the performance—whether it’s on-prem, IaaS, or SaaS, it doesn’t matter. With the right technology in place, you can insure application delivery and fight anecdotal ‘the network is slow’ tickets. After all, how many SWAT or Tiger teams can you be on?
In my next article, we’ll dig in a litter deeper into the architecture. What do I mean when I say ‘route applications and not packets?’ What do I mean when I say, ‘measure to and from the cloud?’ What do I mean when I say ‘speed of light latency can be overcome?’ Stay tuned and we’ll dig into these in the next article.