Virtualization and Cloud executives share their predictions for 2016.  Read them in this 8th Annual VMblog.com series exclusive.

Contributed by Jason Andersen, vice president, business line management, Stratus Technologies

Predicting Availability in a Software-Defined World

As we wrap up 2015 and look ahead to 2016, I wanted to dig into one of the topics from last year's top 10 IT tech trends to watch issued by Gartner:  The rapid change taking place in Software-defined Applications and Infrastructure.

I've seen a lot of progress from companies trying to create new software-based infrastructure. However, I've also seen many of them struggle to determine how to rethink availability in our new software-based world as old hardware-based approaches for system availability and reliability don't always provide the same value they once did. Below are five predictions I see shaping application and infrastructure availability in the year to come:

Prediction #1: Although security breaches are still a big threat to IT, availability is and will continue to be the biggest concern.

The gap between security and availability is decreasing. However, as the impact of computing extends outside of the corporate environment to impact the consumer and society at-large, both are important. It's one thing when you can't withdraw $20 from your ATM; it's quite another when your driverless car does not know where to go or when to stop. People always have seen availability as a simple cost of downtime. So for example, if my system goes down, my factory is less productive and there is a cost to that. But, as more and more applications must be Always On, the impact of availability becomes proactive vs reactive. If I don't need to expend as much effort to build reliability into my applications or infrastructure, I can deliver new innovation to market more quickly. Or I can create things that I never thought possible before because reliability was preventing me from delivering that innovation. We really are beginning to see reliability as a way to help our partners and customers innovate and create new possibilities for themselves versus just being insurance. 

Prediction #2: Big changes are coming for availability at the edge of the network.

At the Edge, the change for availability is less technical but just as profound as it is in the data center. Why? The notion of low-touch at the Edge is giving way to no-touch. For example, you may have a couple people working at a hydroelectric production facility who can reboot a server or who can do some basic admin. But, at a wind farm there are no people day to day. Everything has to be no-touch. That is a big, big change. I predict that many of the automation technologies emerging in the cloud will have some role to play at the Edge because of this. 

Prediction #3: As newer ideas for innovation grow, the need for availability will evolve.

People are really starting to get the need for availability now. They are feeling the pain when they try to deploy newer technologies such as SDN and NFV, or they try and retrofit existing technologies into the Edge. It's just apples and oranges. People are demanding new ideas and the software-defined approach is an example of this.

Prediction #4: IT managers will need to think more about what to spend their budgets on next year.

I advise IT managers to take a hard look at what their applications need versus what their infrastructure provides. You may surprise yourself at what you could save.

Prediction #5: There are 3 key changes IT managers must prepare for now.

1. Massive changes in skills and employee demographics - Baby boomers are retiring and loads of institutional knowledge is leaving IT. New solutions will have to be simpler to accommodate the people who will replace the existing IT operational staff.
2. Getting over the last hurdles to become more software-defined - This has everything to do with looking at applications in a different way - it's about being consumers of services vs. providers of services. Service Oriented Architectures and middleware was a first step to this. SDN and storage are next, and new ideas such as fault and performance management are on the horizon. These are distinct ideas that require different solutions and approaches. If you aren't preparing for this, you are in trouble. 
3. Localizing the solutions - The Edge is changing a lot of industries already. Would an electric company 10 years ago think it would be BUYING electricity from its own customers? You have to start thinking about how the Edge becomes a competitive advantage vs. a cost of doing business.

##

Jason Andersen, vice president, business line management, Stratus Technologies