Saturday, Dec 16, 2017
HomeTopicsApplicationInterview with John Gentry, CTO of Virtual Instruments

Interview with John Gentry, CTO of Virtual Instruments

Ask people over the age of 25 if they remember the sound of dial-up internet and you will likely get a yes. Gone are the days where humans wait for technology to catch up to them, unless you are still waiting for your personal jet pack.

Paired tightly with the rapid pace of innovation is extreme pressure for technology teams to meet the resultant demands. End users, whether they are customers or employees, continuously demand new products, features and functionalities in their application-driven universes. This creates a never-ending cycle of challenges to meet ever-changing needs.

Companies facing these challenges now and forever forward must rethink their organizational structures, specifically their development and operations teams. By enabling their teams to work more closely together, businesses are accelerating their innovation cycles while maintaining standards for consistency and efficiency.

This is the idea and promise of DevOps. In this Q&A with John Gentry, CTO of Virtual Instruments, we’ll learn more about how DevOps is now a critical business enabler.

How do you realize the benefits of DevOps in terms of faster time to market while mitigating the risks associated with agile development and release cycles in production environments?

JG: This is one of the most important questions that DevOps raises. It’s patently clear that there is no “end state.” With development, test and release cycles happening so quickly between development and operations, collaboration must be founded on accurate and trustworthy data and insights, from which quick and confident decisions can be made. If teams don’t have this, they run the risk in every cycle of releasing code that negatively impacts end-user experiences.

The key to minimizing and eliminating these risks is having highly acute real-time situational awareness of how changes in application workloads could impact IT infrastructure performance—and proactively tuning the infrastructure to support these ever-changing shifts without end users being affected.

New features and functions create new demands on IT infrastructure, period. As the market broadens and adoption increases, performance requirements increase both organically, and exponentially. Being able to anticipate how these workload demands will increase over time and using that information to deploy your infrastructure accordingly helps you better meet performance and availability requirements.

How do you create a continuous feedback loop for DevOps to ensure that what is being built is actually meeting customer requirements?

JG: DevOps requires a very close, highly responsive relationship with end users. In order to deliver ROI, you need to quickly convert continual feedback into features and functionality that meets end-user requirements. This is actually a positive.

However, to keep pace with end-user demand and meet expectations across the board, your teams must fully understand the impacts of feature/functionality releases (and what they will be) on the underlying infrastructure supporting and delivering the related workloads.

An example of how to solve to this via a data-driven approach is to leverage workload performance simulation, testing and validation analytics in test/lab environments to ensure that network and storage infrastructure is performing for the workload requirements before deploying into production. Then, once in production, monitoring infrastructure in real time to ensure that the performance delivered is the performance expected. Integrated solutions that address the full application workload development-to-infrastructure lifecycle from test/lab to production (and back to test/lab) are ideally suited for this complex challenge.

How can companies determine which projects are best suited for DevOps?

JG: To truly benefit from DevOps, companies must develop, test and deploy applications and workloads at a rate much faster than they’re used to. Fail fast. Succeed fast. As companies try to shape and refine their DevOps models, they must continually adapt their infrastructure configurations and application functionality requirements to meet end-user demands.

To determine which projects are best suited for DevOps, you need to know how much they will cost, what benefits they’ll provide and how long they’ll take. A major step toward understanding this is implementing infrastructure performance monitoring and analytics. The insights provided with these technologies, in addition to real-world testing, gives companies an accurate picture of how projects will perform and the benefits they’ll deliver. With this knowledge, companies then can make confident, informed decisions about where to deploy resources and how to methodically evolve.

Why does DevOps fail?

JG: DevOps often fails not because of technology, but because of people. This is because DevOps isn’t just a shift in process, it’s a shift in the whole company culture – a shift that can really disrupt traditional roles and alignments.

Development and operations teams have traditionally had two different tasks: developers make the new functions and features, operations manages these capabilities in production. This relationship requires different, less intense, interactions between the teams.

With DevOps, the two teams closely collaborate. While this collaboration is part of what allows the quick development and deployment of solutions, it also creates friction. The key to enabling smooth collaboration is commonly accepted and universally trusted data and analytics. With a common analytical platform approach to DevOps enablement, teams are educated, informed and aligned for whatever shifts occur. If this cultural shift doesn’t happen in a company, DevOps is in real danger of not delivering on the promise.

What steps can companies take to ensure they can leverage the right technologies for DevOps?

JG: As mentioned above, using a combination of workload performance simulation, testing and validation analytics in pre-production lab/test environments, in combination with real-time performance monitoring and analysis in production environments are the two essential processes. A comprehensive solution that spans the full development-to-deployment cycle provides the data and insights needed for authoritative decision support and optimal control.

When done correctly, the information gleaned will ensure you and your teams have a complete understanding of how any new technologies will perform no matter what changes occur. Through leveraging integrated pre-production and production infrastructure performance analytics for every release cycle, your company builds a knowledge base that leads to shorter/accelerated project timelines, lower costs and assured performance.

Virtual Instruments