How To Make Complex Infrastructure Invisible

Times, they are hard. And business could stand a little help so that instead of focusing on keeping the lights on, you can focus on your business. There is no doubt that COVID-19 is changing the way we do businesses. Whether financial, manufacturing or retail. It has taken us around a sharp left. The effect of COVID-19 also means some projects that were at the forefront have been put on hold, and those related to infrastructure pushed upfront for reasons that will soon become obvious.

“Technology powers the business. It is not meant to be a stick in the mud. It empowers customers. I remember I was with this bank for an excess of 20 years. Happy. But every six months, I had to do this one little thing – and it took me around two hours. I looked around for differentiators, finding a bank that automated this and joined it. The latter also straightforwardly moved my data. I was able to move banks because the technology they were using was different. Anytime we talk about customers and competitive advantage, for me this is very key,” discerns Martin Walshaw, Strategic Account Manager, Nutanix.

“The problem we have is, infrastructure is horrendously complex. There are multiple different vendors with multiple solutions put in place. Each of these areas requires specialised skills. I worked for Cisco for over ten years, and I love the company. But at the end of the day they have a huge raft of products, and Cisco, being the behemoth that it is, does end up being quite the challenge should they be deployed,” he says.

So, what happens is that there is this complex bit of IT infrastructure, “and we spend a vast chunk of time trying to keep the lights on. Doing upgrades, making the changes, making sure there is compatibility. And not enough time on the application and services that we are providing to the business on one side, and the customers on the other.”

He adds, “What you should be doing is innovating new and unique ways of using technology with a perspective as to how this can make lives easier via infrastructure. Moving away from the 80:20 rule, while moving towards the applications themselves.”

The problem we have is, infrastructure is horrendously complex. There are multiple different vendors with multiple solutions put in place. Each of these areas requires specialised skills.

Deployment is a disadvantage too because it needs to be activated in phases. “If you want to deploy applications, it has to be checked, approved and you need to make sure it has the right patches. It is very slow and laborious. Not only that, but we are trying to make sure the data is secure and in regulation with the law. More critically, is the lack of time, resources, and budget to innovate. Changing it away from having to look at infrastructure but rather, thinking of the ways we can make it easier, embedded into use,” he observes.

Imagine for a moment, though, that you could change that. That you could scale. That you can put the solution in the tiniest business, as you would put in place for the most complex business. The only difference is the number of nerves. Imagine if it can be secured the same way as if it were on-premise.

“I did a CIO East Africa roundtable last year. I asked a question around the cloud, posing to them, ‘what are you doing around security?’ Most leaders made the presumption that going to the cloud is secure. We all know that is not the case. We should know that that is not the case. The same features and solutions put on-premise, need to be put in place in the cloud.”

What you need to do as a business to be on the same page as technology is

Make sure that you modernise your IT structure: Put a solution in place that can scale if, and when you desire, that is future proof and provide you with services that you need.
Start looking at IT-as-a-Service (ITaaS). Automate as much as possible. Why should you be involved in creating virtual machines or a web server when it can be automated? Why not have self-service portals that allow users to request and create a template of services?
The cloud is great, but it is not everyone’s cup of tea. Certain workloads are far better suited for the cloud than they are on-premise.

Most leaders made the presumption that going to the cloud is secure. We all know that is not the case. We should know that that is not the case. The same features and solutions put on-premise, need to be put in place in the cloud.

Let’s take a look at something as basic as your email address. It is something in the cloud. Everybody uses it. “If I were to ask you which servers, SSD, hard drives, virtualisation, software etc that they are running, you have no idea. I have no idea. The reason; we don’t actually care. All we care about is that when we access it, it is up and running and working. Cloud has done a great, great, great service for us by managing to make this underlying infrastructure, invisible,” Martin posits.

What you really want to do as a business is bring those cloud services inside your infrastructure. That means having the same features and functionality regardless of size. With email, for instance, you care about more space, but not about the underlying services. You just want it to be able to scale according to your requirements. You care that there is no single point of failure. And if there is a problem, hello? It heals itself! You want something like Machine Learning (ML) in place. You want it to be easy to deploy. That you pay for it as you grow. That you have one-click operations. One-click everything. Making that underlying infrastructure invisible. No downtimes, easy to use, stable upgrades, able to grow.

If you want to deploy applications, it has to be checked, approved and you need to make sure it has the right patches. It is very slow and laborious. Not only that but we are trying to make sure the data is secure and in regulation with the law.

If you look at Google, Amazon, Facebook, Microsoft, they continue to provide software upgrades regularly with no downtime. What you want is continuous innovation without downtime. Do the workloads in your infrastructure currently have multiple criteria to look at? Should they be in the cloud or on-premise?

“There are reasons why you would like to have a workload on the cloud versus on-premise. Some applications just do not allow you to move the cloud. As well there is data gravity – which means that the application needs to be close to and around the data itself. In some cases, we have a loss of control. You want to be sure that you have the same level of security in both. Lastly, in some cases, the cloud can be unpredictable when it comes to cost. We want to make sure that you can comprehensively figure out where the workload is best suited. Is it on-premise, or is it in the cloud?” he says.

He adds that “What works for a majority of our businesses is a hybrid. That is where we have some workloads on-premise, that is more predictable, and we have better performance, whereas some of them are better suited for the public cloud. The best example I can give you for an elastic workload would be, admittedly a few years ago, when Justin Bieber and a couple of other artists came to perform in one of our stadiums here. They released 65,000 tickets. The website went down within 10 seconds. They could not be able to create a server that was able to cater to that influx of people. Once that concert was sold out it went back to normal. That particular workload is very elastic rather than being a predictable performer. Organisations generally need a combination of the two. Take it one step further – we also can do multi-cloud as well with different workloads in different public clouds allowing us to simplify the process of where infrastructure is put in place.”

When Justin Bieber and a couple of other artists came to perform in one of our stadiums, they released 65,000 tickets. The website went down within 10 seconds. They did not have the ability to be able to create a server that was able to cater to that influx of people. Once that concert was sold out it went back to normal.

But, how do you actually make infrastructure invisible?

A typical infrastructure focuses on several things, the first being the actual hardware. “Whether it is Nutanix, HPE, Lenovo, IBM, etc., the big thing is hardware vendors are becoming irrelevant. Granted, an SSD 3.84TB is an SSD 3.84TB regardless of the vendor. What does make a difference is failure rates and support.”
Nutanix supports multiple vendors, so you have no tie-in. Going further up the stack is hyper-convergence, which is the pooling of your resources – storage, networking and virtualisation.
Going beyond that is virtualisation be it VMWare, MS or Nutanix. Hyperconverged infrastructure, as the Nutanix website describes it, ‘brings together compute, storage, networking, and virtualisation into a single, efficient platform’ with the bonus advantage of freeing your team from silos with radical simplicity.
“These three layers are invisible to you because we provide one-click upgrades. This would mean that the underlying infrastructure is not there because every time it is upgraded, it is with zero downtime. We have customers – financial institutions – that upgrade their infrastructure during the day. We have a page dedicated to our customers telling us unique places they upgraded their infrastructure,” breaks down Martin. A couple involved a plane.

Finally, you want to buy it, when and how you want to buy it. CIOs and heads of IT can make requests if and when they need it, not limited to a once-a-year purchase. Regardless of the vendor, cloud, hypervisor and use case, his people offer you the freedom of choice.

Here is where their sweet spot comes in. Their secret sauce is this: no matter how many tentacles you are reaching out, to or from, Nutanix provides a sole management platform. This way, you buy what is best for the business and not what is best from the vendor perspective.

To get even more in-depth conversation on this topic, watch our video.