In my discussions with our customers, partners, and while meeting industry players, the topic that surfaces the most these days seems to be the issue of “shadow IT”, or “under-the-desk IT”. I prefer “cloud sprawling” myself, as this is the term I feel captures the heart of the issue that’s plaguing enterprises of all sizes.
Adoption of new technologies is always faster in the consumer space.
Your development team has grown accustomed to the idea of self-service a long time ago
Their hobby projects are most likely kept in a public cloud of some sort.
At work however there is still a serious lack of access to resources. Let’s see how an average provisioning process would go. (The timeframes here are merely illustrations, based on feedback from my discussions.)
- Day 1: Your employees needs to provision resources for a project. They defined what’s needed, and sent a request to your operations team. Depending on the number of others requesting the same thing, and the size and agility of your operations team trying to service these requests, the waiting time can queue up to weeks.
- Day 3-14: The operations team delivers the resources to your developers.
- Day 3-28: If the resources aren’t configured right, operations will need to revise them based on the original (or updated) requests. Perhaps the requirements weren’t defined well enough, or the operations team just didn’t configure them correctly, or – while waiting for the request to be fulfilled – the configuration simply changed.
Best case scenario, your employees sit idly by for 3 days… 3 DAYS! What do you think they’re going to do if the waiting time is four weeks? Or even just half of that?
Most likely what’s going to happen is that they take matters into their own hands. They have jobs to do, after all. Spin up machines in a public cloud quickly (15 minutes at the most) to start working – chances are, by the time your operations team gets around fulfilling their requests, they’re already halfway (or completely) done.
Beyond the obvious financial issue (you’re basically paying for fiddling proverbial – and actual – thumbs for days on end), the heart of the shadow IT problem is security: your intellectual property living on the personal accounts of your employees. You’re dependent their ability to make sure that valuable data is safe — and what happens if they leave the company?
You’ve unknowingly lost control over your own property
You’d think that there’s an easy way to solve this: provisioning resources from multiple pools – that is: both from your private infrastructure and your “on-demand” resources -, selecting the most appropriate for the task, all the while maintaining a single point of control through a single-pane-of-glass interface.
Companies should be able to weigh options more flexibly
Those already partially utilizing public clouds for workloads that are not regulated by compliance mandates can optimize resource allocation with a hybrid cloud structure, while maintaining control through centralized provisioning tools linking multiple infrastructure locations and public cloud services together.
The practice doesn’t always follow the theory though. How many solutions like this exist? I can tell you right now: only a handful, if that. One of them is QStack.