Kathy Gibson at Gartner Symposium, Cape Town — IT professionals are not always in charge of their own destiny — but they have to make it work.
“We aren’t always in a position to architect the solution that way we want it,” says Gartner analyst Milin Govekar.
The new trends and technologies will, however, have a huge impact on IT infrastructure and operations, so IT professionals should start thinking about that they need to do.
The new technologies cloud be strategic, tactical or democratised — the last category being those that are largely beyond IT’s control.
In terms of strategic technologies, these are driven by disappearing data centres, automation and new application development tools like containers, microservices and application streams.
On the tactical front, we are already seeing moves to DevOps, data centre as a service (DCaaS), application service monitoring (APM) and Internet of Things (IoT).
The democratised technologies start with cloud computing — which many organisations see as a way of bypassing it, micro- and edge computing environments, and the digital workplace.
Disappearing data centres is a reality. The number of computing workloads is going down quite significantly, Govekar points out.
Currently, about 70% of the enterprise workloads run in the data centre, and this will go down to about 15% by 2024. The new workloads will not only move to cloud, but to the edge as well.
Implications of this move means there will be multiple clouds and other locations where computing takes place.
“IT has manage all these environments,” Govekar says. “The type of skills we need will change, and this will have a massive impact on leadership structure.”
Gartner proposes an organisation model that looks drastically different from how it looks today. Staff will have to be more versatile rather than having specific skills.
“Ultimately, a broker role will be critical,” he adds. “You want the business to have confidence and trust in you when it comes to choosing services, rather that bypassing you.”
Automation will have the biggest impact, Govekar says. “In infrastructure and operations, the biggest form of automation is currently scripting. People are solving the automation problems tactically and opportunistically by writing scripts.
“This means there is not much visibility into how many scripts there are in an organisation.”
Automation needs to be made systematic, however, and this could be done by using strong governance that is policy-driven. It addresses the enterprise requirements, scales and offers end-to-end integration. Automation should also be metrics-driven, offering clear business benefits and a manageable total cost of ownership (TCO).
Artificial intelligence (AI) can help with automation and is becoming popular, in the service desk environment, in performance management and in the software-defined area.
Containers are the new trend in application development, to enable new forms of applications through agile development. Most containers are deployed on virtual machines, especially in cloud-based infrastructure as a services (IaaS0.
Govekar believes containers will become a primary tool for application developers, so infrastructure and operations professionals will have to find ways to support than, and provide container governance tools to support developers.
Container skills are scarce, but Govekar says its vital for organisations to get them. “You need to get fluent in all the buzzwords, and understand the impacts.”
DevOps is rapidly becoming common, with 29% of organisations already using a DevOps approach to IT. Among these, 86% have been using it for at least a year.
There is a disconnect between the speed requirement by the development people, and the metrics demanded by operations. Since the point of DevOps is agility, infrastructure and operations hass to find way to manage the disconnect.
A solution could be to align IT projects by products, Govekar says. This will result in a significant change in the organisation structure as some of the IT resources become embedded in the business.
It’s important to deal with the culture, and this should be prioritised above the tools, he adds.
DCaaS is seeing the biggest change. IT needs to make the data centre frictionless for the applications, while keeping the features of the data centre intact.
“You need to make the business aware of the fact that the technology needed to monitor the end user experience don’t yet exist,” Govekar points out. “You may have to write these yourself.”
IoT is one of the most visible and widespread changes happening in IT right now.
“Typically with IoT projects, there is a service provider implementing it on behalf of the business.” Govekar says. “But there are potentially hundreds of thousands of events being generated by IoT.”
IT simply doesn’t have the capability to perform a root cause analysis on these events, so infrastructure and operations has to find ways to integrate with the service or network supplier as well as with the enterprise ERP system.
Democratisation is seen most clearly in the implementation of cloud computing. Although the most common reason for adopting cloud is agility and cost, it is also a route to IT modernisation. “Many organisations are looking at a cloud-first infrastructure,” says Govekar. “And it is becoming more strategic in many organisations.”
It’s important to raise there is no such thing as one cloud, and any cloud strategy is going to have to include a number of solutions.
Govekar recommends that cloud be approached in a redesign rather than a redeploy operation, one workload at a time. It is also urged to focus on outcomes.
Computing is having to move closer to the edge as millions more devices are connected to the network. “You no longer have to accommodate data centre and as-a-service technologies, you have to look at combining the edge with the traditional data centre as well,” Govekar says.
The workplace is no less affected in the digital business, and workers expect a different experience now.
Govekar recommends a light-touch approach on thinner solutions. At the same time, the dependency on the operating system needs to be eliminated, letting users adopt applications as and when they are needed.
All of this means that infrastructure and operations will have to develop a range of new skills and roles. Top of the list is cloud skills, followed by business intelligence and analytics, critical thinking and problem solving, technical skills, and security skills.