Is Dev/Ops for real, or is it simply the latest marketing tool to get you to buy more stuff for your data center? Or is it a little of both, a potentially revolutionary change to enterprise infrastructure management provided you can see through all the Dev/Ops-washing that is going on?
As with most technology initiatives, the concept behind Dev/Ops is solid – it offers a more flexible approach to the data-resource allocation challenges present in hyperscale and Big Data environments. But by the same token, success or failure is usually determined by the execution, not the initial design. So the real challenge with Dev/Ops is not in selecting the right platform but in taking the designs and concepts currently in the channel and making them your own.
For those involved in developing Dev/Ops platforms, its transformative powers are substantial indeed. Chef Software CEO Barry Crist told Diginomica recently that most enterprises will see Dev/Ops go from initial deployment to widespread integration in as little as two years. And we’re not just talking about shuffling virtual servers around but a full-stack automation regime covering hardware, software runtime and even applications themselves. To achieve this, though, Dev/Ops will have to transcend the IT department to foster cross-functional capabilities, and then focus on process outcomes by incorporating development, product management, business development and other elements.
For traditional infrastructure vendors like HP, Dev/Ops is the basis for “composable infrastructure,” says The Platform’s Timothy Prickett Morgan. With hyperscale entities and the hyperscale-wannabes among the broader enterprise community anxious for an efficient means to handle vast resources, the company has launched Project Synergy which, while still largely under wraps, will likely build upon management stacks like OneView to build the kind of management regimes that Facebook, Google and other hyperscalers are developing in-house. The advantage for HP is that it will be ready-made for its installed base of server, storage and networking hardware while at the same time providing open APIs to draw in third-party developers. The downside is that this kind of management ecosystem could be years in the making, and enterprises of all sizes are looking to tap into Big Data and other heavy workloads right away.
Dev/Ops is all about software controlling hardware with little or no human oversight, so it makes sense for a software company to take the lead – which is exactly what Microsoft is trying to do, says Forbes’ Adrian Bridgwater. If the goal is to foster orchestration and automation to support rapid and agile application development, the combination of Visual Studio Online and Team Foundation Server (TNS) source code will provide the real mechanics of Dev/Ops. This combo will take care of all the project and build functions, including collaboration and reporting, that will push infrastructure management to the next level. As well, the company just released a self-assessment tool to give teams the ability to determine their own strong and weak points across a range of key Dev/Ops functions.
And like just about everything else in the converged, integrated data center these days, Dev/Ops is more than just a technology play but a cultural shift that alters the way in which all stakeholders view and leverage data infrastructure. As Contino co-founder Benjamin Wootton explained to InfoQ recently, collaboration in the Dev/Ops process is going to bring together people who have had very little contact in the past, which will in turn lead to greater empathy for the challenges that others face and ultimately a more cohesive and responsive workforce. The real challenge will be to view Dev/Ops by the way in which it can remake data infrastructure from the ground up, rather than to simply patch problems with existing architectures.
All of this may be a little hard to fathom at the moment, considering that most organizations’ experience with Dev/Ops is a few pilot projects, if that. But if early predictions are correct, the enterprise will need a way to quickly optimize emerging infrastructure for particular data needs. And since no two enterprises are alike, the heavy lifting in this optimization will have to be done in-house.
In this regard, Dev/Ops does have the capability to be truly transformative, but it will take a concerted effort to ensure that the transformation is for the better.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata, Carpathia and NetMagic.