Last in a four-part series on taming cloud complexity, with Forrester
At a basic level, CloudOps requires breaking down walls between DevSecOps and site reliability engineering to create a single, unified plan, build, run, manage process to optimize outcomes. Especially in distributed and hybrid cloud environments. Yet, in most cases, it’s a work in process. According to Brent Ellis, Sr. Analyst at Forrester, most organizations are “still in the adoption phase, with most operating in a divide and conquer model.” “The big question for many is how and where do we start?” adds Prem Balasubramanian, CTO – Digital Solutions, Hitachi Vantara. “We've got a large development team and a lot going on, workloads in production, so how do we do this in a way that brings people together?” This was the focus of the final episode of our four-part conversation with Forrester, Taming the Complexity of Cloud Operations, exploring the broader topic of optimizing CloudOps, costs, and workload design to achieve success in the always-on era.
Begin with an empathetic mindset
While getting started requires a combination of people, processes, tools, and governance, it’s more about a new operational philosophy and mindset. “You really have to encourage people to work collaboratively and think of the business’ goals,” Ellis says. “Getting people to holistically understand what the business is trying to do and how their particular role fits in.” To avoid territorial tendencies, you need to create shared empathy, so dev people understand how choices they make impact operations and security, and so on. It’s also important everyone agrees on tools and nomenclature to build a framework that self-enforces improvement and collaboration.
Implementing SRE isn’t about hiring
Building an SRE mindset enables buy-in and iteration towards SRE practices. But hiring SRE experts alone won’t create change. “You can accelerate the process by hiring seasoned talent, but the goal isn't to just put them in a role and say, now we have it,” Balasubramanian says. “It’s more a way to help evangelize and teach others across your organization to use SRE principles.”Ellis concurred. “When I speak about this, I’m talking about the process of SRE or resilience engineering, not the role,” he says. “It requires changing the way you approach the work you do as an organization.”
Start small and build on your success
Complexity remains a challenge to melding DevSecOps and SRE together. “It sounds like we should be there but we’re not,” says Hitachi Vantara’s Samta Bansal, who led the discussion, Redefining CloudOps – Breaking Down the Walls between DevSecOps and SRE. The production environment can be especially complicated and there isn’t any sort of map showing where things might break. So, you need to form and test a hypothesis, then measure and improve from there.
Balasubramanian says the best approach may be to start small and use those successes to specific, important challenges to spread the culture across the organization. “When you think about reliability and resilience coming together for always-on business, you need to be able to embrace failure, and use that to build a system that can tolerate it.”
Taming the Complexity of Cloud Operations
A four-part webinar discussion series with Forrester
Episode 4: Redefining CloudOps – Breaking Down the Walls Between DevSecOps and SRE, featuring:
Be sure to catch the first three episodes in the series:
Episode 1: What’s In and What’s Out? Optimize Your Hybrid and Distributed Cloud Environments.
Episode 2: Optimizing Cloud Economics and Scale
Episode 3: Design Your Workloads for Reliability and Cost
A proud part of Hitachi Vantara
© Hitachi Vantara Corporation. All Rights Reserved.