Dev Ops

Transforming Traditional DevOps To A Modern Cloud-Centric Operation

Transforming Traditional DevOps

Over the last year, I’ve been hearing about more and more instances of companies asking the question that the title suggests – how do you transform a DevOps process into a more cloud-centric operation?

To start, we must all assume that there is some notion of a traditional DevOps process built into the company, and while on the face of it, many believe that it’s a technology based idea – it’s a bit more complex than that – DevOps is a business culture shift. However, for the focus of this article, please accept that the development and operations teams are on the same page, have created a sense of shared responsibility, and are working together to build revenue. It’s a grand notion, but let’s assume .

So how do you make a DevOps process more cloud-centric? By concentrating the process around infrastructure solutions and business operations. Sounds obvious, but here are the five areas enterprises need to concentrate on to accomplish the task:

Gaining Cloud Visibility and Control

Ok, on the surface, this looks like operations (and I’m lumping the security team here) returning to their old tricks of being the group that’s always saying “NO!” Let’s say in this situation, providing cloud visibility to IT Operations empowers them to enable the use of a broad selection of cloud services. IT Operations will be able to see the current service needs for both their business and development partners; while maintaining crucial governance and compliance requirements. And proactively, IT Operations can assist their business customers with understanding potential opportunities for cloud adoption and assist with a supporting business case that describes the full benefits of operating on a cloud service.

Build an IT as a Service Culture

This is the biggest challenge, just when we thought it was difficult enough to create a DevOps environment, here’s “Part Deux” – to use a bad 90’s movie title joke.

Enterprises will need to shape operations and development to act as IT brokers or internal consultants to address business challenges. They need to accomplish this utilizing multiple cloud services — actively discovering, managing, and dismissing cloud providers. The team needs to forge new relationships with their “business clients,” and learn to give up some control to provide users choice of the optimal cloud services for their needs.

So if this is the biggest challenge, why isn’t it first on the list?

Because, if you are working in a fog, you can’t build a bridge between two land masses.

Business Financials and Governance

A successful cloud-centric DevOps platform needs to provide a financial support system that can quickly and easily provide hypothetical pricing, billing, system metering, etc. And if that’s not enough, teams should be able to provide financial analytics that ensure the best budget matches for each development solution. And let’s not forget governance, a cloud-centric operation needs to be able to provide a clear, corporate approval process as well as policy management and an ongoing review for continuous improvement.

Automation is important, but be wary

Just because the shampoo bottle says, “Lather, Rinse, and Repeat,” do you do so one hundred times in a shower? Of course not. However, automation can give a sense of empowerment, thus we are push hard to give our business clients a feeling like they own the service, sometime to IT’s detriment. So it’s important to consider which cloud-centric DevOps processes are most significant to develop into an automated task. Now, I don’t have an immediate “should” answer here, rather a caution. Most enterprises focus on provisioning and orchestration as there automation process, without considering the big picture. Although “sexy,” in the long run, that approach could cause more code review issues than it’s worth. As the above subjects point out, automation needs an important focus on a means of visualizing and managing benefit, platform comparison, as well as financial administration. I suggest concentrating in these areas. Always remember, cloud providers are constantly changing and adding to their provisioning and service APIs, but when it comes to their financials, the APIs never change – just the data.

Think beyond Infrastructure as a Service

And lastly, IT Operations needs to help and train the development and business teams to “not reinvent the wheel.” As of this writing, there are approximately 11,000 cloud services available to businesses. Understanding cloud API models isn’t always easy, but there are many ways to integrate them into an enterprise platform, infrastructure, or Software as a Service. And as each cloud service has its own underlying components, enterprises could take advantage of their “modularity” to add greater functionality to their applications while at the same time decreasing the platform’s time to market.

To sum it up, the process of operating in a DevOps ecosystem is difficult and hardly “traditional.” However, although I describe 5 laborious steps to get there, once realized, IT Operations can make the realization of acting as a Cloud Services Broker; thus enabling development and business units to effectively and efficiently get the most out of cloud computing for their business and customers.

By Tony Pagano

THOUGHT LEADERS

John

China Ascendant? Maybe, Maybe Not

China seems to be beating us at leadership in tech in just about every category. In a decade they are supposed to be the largest ...
Louis Columbus

Why AI Is The Future Of Cybersecurity

Future Of Cybersecurity 61% of enterprises say they cannot detect breach attempts today without the use of AI technologies. 48% say their budgets for AI ...
Ashwin Krishnan

Get Used To It – Artificial Intelligence For Real-time Gas Pricing

Real-time Gas Pricing Get used to it – we will extract every dollar you can afford at your friendly …. ! I was reading in ...
Daren

Countdown to GDPR: Preparing for Global Data Privacy Reform

Preparing for Global Data Privacy Reform Multinational businesses who aren’t up to speed on the regulatory requirements of the European Union’s General Data Protection Regulation ...
Gilad

Accessing (HPC) High Performance Computing

HPC in the Cloud Big data and Machine Learning (ML) can provide businesses with incredible insights and an innovative edge. However, to properly analyze the ...

SPONSOR PARTNERS