How companies can tell good cloud sprawl from bad: A guide
Now that operating in the cloud is officially mainstream, it’s gotten a reputation for costing more than expected. Cloud sprawl, however, is rarely the problem – in fact, we like to think of sprawl as a symptom. On the bright side, it’s often a symptom of creativity and innovation by your IT team. On the dark side, it’s also often a symptom of poor planning and a lack of governance.
Here’s a guide to structuring (or restructuring) your cloud adoption so that any growth in expenses is tied to commensurate improvements in outcomes.
Unplanned sprawl is always bad
The on-demand nature of cloud infrastructure and services make growth frictionless. It’s as easy to spin up a new server as it is to download a document. While that’s fantastic for a team’s ability to innovate and grow, it’s also a little frightening.
When there’s a complete elimination of hardship in commissioning new cloud services, it’s entirely too easy to exceed your budget.
What we often see happen is that a company decides to dip a toe in cloud transformation. They don’t worry about creating detailed governance guidelines because they’re treating their cloud use as a trial – they want to see if it’s right for them.
But as soon as engineers see what’s possible in the cloud, they get ideas. They start to figure out creative solutions for the problems they’ve been dealing with for years. And because the cloud is frictionless, they can spin up new resources without batting an eye. Before you know it, the cloud bill is double what you budgeted and you have no plan in place for ensuring that any of your experiments become ROI positive.
...but sprawl itself can be transformative
Of course, the lack of friction is also what makes the cloud such an amazing tool.
For example, in the pre-cloud era, I was working at a hedge fund. A colleague and I wanted to take massive amounts of data and run it through a bunch of scenarios to start doing predictive data arbitrage.
We knew we’d be able to glean valuable insights if we ran our data through enough scenarios and models. But in order to process all the data we’d bought in time for our analysts and quants to evaluate and make decisions about it, we’d have needed about 200 servers, which would have meant millions of dollars in capex spend.
And keep in mind: this was just for research. We had no way to guarantee that our analysis would have a positive ROI. Of course we didn’t get it approved.
With the cloud, though, we could spin up enough servers in a few days, run the scenarios, and spin down the servers. The cost would be much lower, which means the additional revenue needed to justify the project would be lower.
And that’s true in nearly every situation: because you can provision resources on an as-needed basis, you can achieve much smaller returns to justify your investment. And as you discover ways to increase profit incrementally, you can, thanks to the way the cloud works, rinse and repeat.
To ensure positive sprawl, invest in planning and governance
So how can you enjoy the many benefits of cloud infrastructure without suffering from its pitfalls? Start with a plan.
Even if you only have plans to run a cloud trial, start with a plan.
The cloud offers near-infinite ways for your engineers to solve problems, and as soon as they get a glimpse, I guarantee they’ll want to try things out. Without a plan in place, they’ll do exactly that, and the next thing you know, you’ll have dozens of active cloud accounts paid for with dozens of different credit cards. (This is a real thing we see.)
So, again: make a plan. Make a plan for how new cloud spend is approved and paid for. Make a plan for reviewing cloud usage and spinning down resources that are no longer useful or active. Make governance guidelines.
If you’re not sure what should go into your cloud plan, consider working with consultants who have experience creating these plans. With a plan in place, you’ll have a clear idea of how you’ll be using cloud resources to improve your business’s operations. Without a plan, you will see sprawl. Most importantly, without a plan, you may end up spending much more than you expected – without achieving the outcomes you were after in the first place.
Creative solutions start with practical guidelines
Sprawl is almost inevitable when a company launches a cloud transformation: because the cloud is capable of more than in-house servers, moving to the cloud inevitably means that companies do more than they used to.
To ensure that you do more in a budget-conscious way, invest time and energy upfront in establishing a plan and governance rules for your cloud usage. Once you’ve laid out rules of the road, you can let your developers explore safely, without racking up unexpected bills.
Interested in hearing industry leaders discuss subjects like this and sharing their experiences and use-cases? Attend the Cyber Security & Cloud Expo World Series with upcoming events in Silicon Valley, London and Amsterdam to learn more.
- » Kubernetes and multi-cloud: How to monitor your modern applications effectively
- » Cloud services and infrastructure spending breaks $150bn in six months, says Synergy
- » AWS, Azure or Google: Do the differences between cloud providers really matter?
- » SQL Server high availability and disaster recovery for AWS, Azure and GCP: A guide
- » OVH rebrands as OVHcloud, claims more than 70% of revenues are cloud-based