DevOps and operations teams are always focused on problem solving. They are not only under increasing pressure from tech-savvy, app-centric business users to collaboratively solve challenging business problems with IT, but also trying to keep up with a rapidly changing landscape of tools, technologies, processes and adoption. So, as we head into 2015, it is important to take stock of all the potential new challenges ahead … as well as some ways to get ahead of them.
IT infrastructure used to be a service published to the rest of the organization – all new requests were made to and executed by the operations team. However, the process is becoming more democratized and collaborative. And while that is certainly a good thing for business, it shifts the dynamic and responsibilities considerably.
Luckily, knowing what is coming will help DevOps and operations teams better prepare to overcome it. In this slideshow, operations expert Dr. Trevor Parsons of Logentries explores the five challenges teams need to address – from cultural changes and shifting ownership to hybrid environment complications. He also shares steps to take now to overcome future challenges and save some headaches along the way.
Top Challenges Facing DevOps/Ops in 2015
Click through for the five primary challenges DevOps and ops teams will face in 2015, as well as steps teams can take now to help overcome future challenges, as identified by operations expert Dr. Trevor Parsons, Logentries.
Challenge #1: Ownership Wars Might Be Stifling Culture
While culture is a fundamental part of DevOps, it goes well beyond Ninjas, Rockstars, and hoodies. At its core, DevOps culture is about fostering collaboration and results-driven effort. However, until operations can figure out how to better foster culture within its daily activities, this aspect of DevOps will not flourish. Infrastructure is still in operations’ full control, but the culture needs to span across development silos and embrace shared data, shared tools and shared approaches. Without a sound DevOps culture, getting the proper processes in place and tools implemented will be much more challenging.
There’s an App/Tool for That
Challenge #2: The “Tool/App for That” Generation
The pressure from business users who believe everything can be solved with “a simple app” has become downright destructive. Business users are adamant in their knowledge about great tools, and thanks to the emergence of open source or SaaS-based offerings, users can often start using a free trial of the tool instantly – cutting out operational oversight. Even better, most users assume every tool should/does have an app. But even if there is an app, it must be properly tied to everything else to be useful.
Operations must support the tool/app evolution to maintain awareness, governance and control. Managing the adoption, integration and quantification of an endless supply of tools and apps is no small task so leveraging the shared services concept to balance tool delegation and management must be done early on.
Legacy Meet Cloud
Challenge #3: Making Legacy Meet the Cloud Era
Development teams and business users may want to jump at all the latest tools, but operations simply can’t. It isn’t because they don’t want to, though. Tools need to fit into an existing application ecosystem, and it’s not easy to make legacy apps work well with modern tools. An added plus? There has been a major move to mobile devices, which are self-contained and voracious for enterprise-wide cloud services.
Meanwhile, the new hybrid environment includes on-premise tools, IaaS tools like Amazon, Google and Azure, PaaS compute services, and SaaS business applications. Get ready to become a matchmaker – 2015 will be all about making seemingly incompatible tools jive together to meet business needs.
Challenge #4: The Networking Roadblock
Networking, not servers like many assume, is often the unalterable object that resists the agile evolution at every turn. Need a quick example? Even vLANs have a spider web of interdependent IPs, ports, routers, routing tables, perimeter networks, etc. Changing, breaking, moving or replicating any part of that web often seems impossible.
Challenge #5 – Marginalized and Changing Budgets
Shrinking budgets paired with increasing demands on DevOps and operations is nothing new. The big change is that the budget structure is also morphing, and most organizations are struggling to align that with the massive technology shifts we explored earlier. While capital expenses may be separate from operational ones, CapEx spends are now often mirrored by OpEx ones. Trying to quantify spend on an on-premise project management tool with a SaaS application is a major headache that can no longer be left solely in the hands of the CFO.
This shift impacts the structure of IT budgets – it could even take away IT’s control if OpEx spend becomes more distributed.
How DevOps and Operations Can Get Relief
So while overcoming challenges is nothing new, DevOps and operations teams need to address the mounting pressure from the rapid rate of change before it turns into an uncontrollable pressure headache.
Relief Step #1: Collaborate
Discuss these issues in an open forum. It might seem to get a little tense at times, but it is well worth it. Collaboration opens the doors for all teams to learn from each other and suggest solutions – and yes, also to raise important red flags. Make it your mission to break the decision-making hierarchy and foster a flow of shared thoughts and ideas. The trick? Keep the ideas to specific discussions so they don’t detract from normal work activities.
Relief Step #2: Empower
Here is a novel idea: Let teams pick their own tools. However, ask them to report them all to IT, of course. If that seems too risky, another approach that empowers business users and lets IT keep control is to introduce a library of vetted tools that they can select from at their leisure. Encourage business users to suggest new tools so they have a say. The trick? Keep the tools current and make sure they actually address user needs.
Learn a New Language
Relief Step #3: Speak Accounting
Address the chargeback and OpEx issues head on so they don’t become obstacles to tool adoption. If IT can help finance understand what is going on in the world of technology licensing vs. cloud service, it can fast track the inevitable shift. The trick? Learning a new language.
Break Network Shackles
Relief Step #4: Break the Network
Remember that immutable networking infrastructure we discussed earlier? Well, now is the time to break the shackles that old infrastructure created. Why wouldn’t you want to allow development work for large business systems to be tested on exact duplicates? You might even push forward so fast that you will inevitably break something, but it likely won’t impact anyone else. The trick? Leverage advanced virtualization tools (i.e. converting legacy systems to virtual machines) and the power of modern software-defined networking (SDN) technology (rather than staying locked into vLAN and physical devices).
Adopt Robust Analytics
Relief Step #5: Prioritize Analytics
Build analytics and monitoring into everything – and Parsons means everything. Adopt a robust analytics platform, where all application and tools data can be stored, so that operations can have the systems visibility it needs. Do this even for the systems not under operations’ control – give them awareness without hindering adoption. Leverage analytics to guide and manage the effort. The trick? Become a facilitator while maintaining all needed operations practices.