In this special guest feature, John Gentry, CTO of Virtual Instruments, discusses how DevOps gives organizations the opportunity to increase revenue and customer acquisition, due to its ability to kick-start development and accelerate release cycles. But in order to do so, companies must adapt their approaches to reflect design, deployment and management of underlying infrastructure. As CTO of Virtual Instruments, John is responsible for being the voice of the customer and understanding the key IT infrastructure industry trends that affect product strategy and strategic alliances. He brings nearly 20 years of executive IT industry experience and has worked at industry leaders such as Qlogic, Borland, McData and CNT.
The speed of business today is either exciting, or frightening depending on how you see it. Throughout every industry, IT teams face staggering pressures to deliver-on the impossible. No matter what shifts occur, no matter how rapidly, or how frequently workloads change, they must provide a highly performant IT infrastructure that delivers flawless end-user-experiences.
At the same time, these same IT teams must also enable rapid application development, test and release cycles, without impacting the performance and availability of their other production workloads that are sharing the same infrastructure. They must deliver on this impossible set of imperatives despite the explosive data growth fueled by BYOD, IOT and 24/7 access, despite the ever-increasing complexity arising from virtualization and software defined strategies, and despite the accelerated pace of innovation.
Welcome to the new normal, where IT Development and Operations teams no longer have the luxury of time for siloed approaches to application lifecycle management. This is forcing IT teams to acknowledge an inconvenient truth—that they are trying to manage their massively complex business, cultural and technology shifts, with obsolete and ineffective tools, processes and strategies that were never intended to manage this new breed of complexity. The stakes are high, the risks higher and the uncertainty greater.
To overcome these seemingly insurmountable challenges and keep competitive pace, companies are rethinking approaches, reconfiguring teams, and requiring data-informed decisions at every turn. Though slightly different in every organization, DevOps teams now generally own this charter – to fundamentally enable IT and business agility, innovation, and competitive pace, no matter what changes occur.
To help them deliver on their charter, DevOps teams are turning to Big Data as the “killer app” that will accurately inform decision-support and enable rapid adaptation in a guided and controlled manner. But, for Big Data initiatives to work, Big Data Infrastructures need to be there – robust, reliable, and highly performant, no matter what changes occur… see the problem?
Defining DevOps these days is a bit tricky, as it has multiple meanings depending on who you talk to. Here’s one take: DevOps, which gets its name from the relationship created between development and IT operations, helps companies effectively innovate (or fail) fast in numerous ways. It provides a framework that helps organizations understand exactly what technology investments are needed, to deliver innovation.
Companies that are sticking to their old silo’d development and release models are scrambling to adapt and re-tool old technologies to keep current. In so doing, they are putting their futures at risk. Simply increasing spend and deploying more resources to ensure performance is no longer a viable option. It’s not scalable, and doesn’t help companies manage effectively or efficiently. More importantly, it doesn’t help them understand how to adapt for rapidly evolving workload requirements.
Solving-for these circumstances helps bring the DevOps charter more clearly into focus—as the core enabler for continual innovation, transformation and adaptation.
DevOps requires organizational change
Effective DevOps models require organizational change, which is never easy. It forces new collaboration, alignment and fluidity to achieve more deterministic outcomes.
On the technology side, going from test and development to production is never seamless, especially for large enterprises deploying mixed workloads across heterogeneous infrastructure and processes. Therefore, having and making the best use of both pre-production and production performance management and analytics platforms is critical. Teams must continually test, monitor and adapt their mission-critical IT infrastructures for dynamically changing workloads. Doing so as an ongoing practice provides the insights that ultimately help to cut costs, enhance performance and improve overall IT efficiency.
DevOps models evolve to incorporate more expertise and skillsets to confidently manage initiatives including cloud, Big Data, virtualization, software-defined, hyper-convergence, and innumerable permutations coming down the road. DevOps is critical for any engineering organization.
Embarking on a Big Data projects without DevOps ultimately results in less than optimal approaches and questionable outcomes. These projects are simply too complex to handle without the fast fail/succeed approach that DevOps affords. In no uncertain terms, the success of Big Data initiatives is heavily reliant on a strong, data-informed DevOps culture.
With DevOps, the days of relying solely on IT service providers are gone – a shift that spans both culture and technology. While service providers can provide value in the traditional IT tasks of support and maintenance, they cannot easily enable companies to quickly adjust to constantly evolving demands from the business and end-users. Moving away from service provider-centric models has many implications on the technology side, of course, and also necessitates cultural transformation.
Culturally, developers have traditionally seen themselves as the creators of software, products and solutions, while the operations people have managed those creations once they exist. With DevOps, the two teams work together—allowing IT and development teams to collaboratively implement solutions, and be integrally involved in the overall lifecycle.
This fundamental change requires a change in outlook. The overall strategy must involve options such as retraining existing employees, hiring new ones, and implementing new solutions. Nobody is being marginalized, however employees must be cross-trained on different skill sets for DevOps to be effective and achieve the Agility mandate.
It’s been said that change is the only constant in business. As the pace of business accelerates, the only way for the Agile methodology to continue forward is through incorporating a DevOps-focused approach. This provides the best way for businesses to evolve with the market and the quickening pace of business.
Sign up for the free insideBIGDATA newsletter.