The mighty, magical interwebs are awash with metaphorical majesty about the benefits of a strong DevOps culture and healthy collaboration. Oftentimes, much of that information stops just short of saying this: These benefits derive from empowerment — specifically, empowering the people in operations, business, and creative teams to make their own decisions.
Real empowerment delivers real benefits. When teams are trusted to make their own decisions, they’ll usually do what’s best for themselves, without negatively impacting the business over the long term. Since business value today derives from speed of delivery, the choice of tools with which teams work are becoming more local, more exclusive, less enterprise-driven. Ease of use, avoiding vendor lock-in, expediting project readiness, and promoting openness are key factors in each of these decisions. This way, new mandates for features or new requirements can be quickly addressed with new solutions, or by swapping out.
Perhaps the best thing about these more local development environments is the “winners are grinners” factor. When teams continuously adopt innovative technologies to solve real and immediate problems, other teams will gravitate towards the solutions they find. Call it DevOps-style collaboration, DevOps culture, whatever — people will back the winners rather than continue to uphold centrally maintained decisions, rigid policies, and calcified practices.
Today, IT operations are faced with a barrage of questions about their relevance and viability in the enterprise, especially since centralized IT management services are now easily obtainable from cloud service providers. These new services are perceived — true or not — as easier to access, simpler to use, and more effective.
How should IT operations teams respond? Perhaps the only viable response is not with argument but with action: Build and deliver integrated services that ease the burden on those tasked with delivering quality software at speed and scale. Give developers easy access to observability and monitoring services that can profile the performance impacts of their coding decisions before deployment.
IT operations teams should not be faced every day with hurling code over the castle walls into the production environment, and ferociously defending their entrenched structures with rigid change management policies and standardization dictates — the boiling oil of enterprise IT. By giving these teams the option of competing, they take on the mantle of service provider and adopt some of their standards and practices as well: going above and beyond, treating colleagues as customers, and crafting valuable services.
In traditional production environments, application monitoring has been used in a break/fix context. In this new environment, monitoring becomes essential for increasing competitiveness, but it requires new approaches in terms of tools and practices:
1. Monitor to Communicate
Highly competitive internal service providers know that other teams — who are, after all, their customers — have unlimited choice. So it’s important that they vigorously demonstrate the value of their service offerings through constant communication.
Here, modern monitoring can be highly effective. For example, when ops teams provide the entire organization with business service status screens, everyone has shared visibility into the health of their running applications. Ideally, teams can configure these services according to their own requirements, although the greatest value comes from providing a single source of truth.
Naturally, the truth isn’t always palatable, especially when one is on the receiving end of it. No matter how good alerting systems may be, it’s crazy for response teams to keep putting out the same fires. Therefore, modern monitoring services should always mix the good news with the bad news, without using the same color of brush to paint both. For instance, make use of analytical hotness to do a little “dashboard shaming” on all the repeat systemic offenders. This might sound painful, but fragile systems can only get fixed when the chaos they cause is made visible to management.
2. Targeted Marketing
Prospective customers for monitoring services won’t necessarily be other operations teams — they could be developers, product managers, data scientists, even business analysts. These folks need simple explanations of what these services provide and how they personally benefit.
Trying to dump existing monitoring services, no matter how complex and unwieldy, will be fruitless. So ensure that ops teams provide these services with clear instructions and that they deliver seamless integration. For example, with an application monitor rooted in Jenkins, developers can see the performance impacts of their code. An agentless Docker monitoring service may be marketed and packaged as providing immediate microservices performance visibility, without painful configuration.
3. Extended Service and Support
With so many choices and alternatives open to their customers, internal teams must differentiate the breadth and depth of their services. Support is one potential differentiator.
Leveraging their built-in advantage of proximity to the customer’s businesses, the best teams work upstream with their customers. Perhaps you’ve heard of Gemba walking, where one interacts with users, listens to their ideas and their complaints, and offers empathy for their business pains. Less personal, more technical examples include a pre- and post-deployment monitoring snapshot service, or a fully integrated APM including a load testing function.
In any case, every offering should be backed with superior customer service and a great experience. That way, your service isn’t just addressing problems, but also helping forge strong relationships — which are the key ingredient of DevOps.
With more and more cost-effective alternatives available to the people they serve, an organization’s internal teams must become competitively savvy about the services they provide. Building off of traditional production functions such as monitoring, to produce tailored, seamless offerings that are differentiated by ease-of-use, customer service, and support, everyone prospers.
CA Technologies is a sponsor of The New Stack.
Title image of a painting of track and field athletes by Raffaello Fabio Ducceschi, released under Creative Commons 3.0.
The New Stack is a wholly owned subsidiary of Insight Partners. TNS owner Insight Partners is an investor in the following companies: MADE, Docker, Real.