Ellen Rubin is CEO and co-founder of ClearSky Data.
Don’t worry, this isn’t another one of those articles that says, “multi-cloud is coming!” We all know multi-cloud is already here, right? In fact, in its “2017 State of the Cloud” report, Rightscale found that approximately 85 percent of organizations have a multi-cloud strategy. Further, companies in the cloud are utilizing, on average, 1.8 public clouds and 2.3 private clouds.
We also know the reasons why companies have been moving to the multi-cloud model. Making use of many cloud platforms helps you avoid vendor lock-in and choose the right cloud for each app and it’s data.
One thing many companies can’t quite explain, however, is why their multi-cloud projects aren’t staying within budget. One of the major reasons to move to the cloud is the cost savings it promises, so this problem is particularly frustrating for IT departments – and budget makers.
There are many reasons why cloud projects are over budget, but we’ve found that with careful planning and management, most companies can get exactly what they need out of the cloud. Let’s look at some actions you can take to keep your multi-cloud projects on – or under – budget, both before you move to the cloud, and after you’re up and running.
Before You Move
The question is no longer about how to move to a multi-cloud model, but how to make the best use of a multi-cloud model to save your company money while optimizing efficiency and performance. To accomplish this, you can’t just “lift and shift” your apps and data. You have to truly understand each app, and what they require to run as the business demands.
Three things we suggest that our customers pay close attention to when determining which apps to move to which cloud providers are:
Some apps have a steady flow of use, others have peaks and valleys. Some require a lot of compute to crunch data, others are leaner. Develop a performance profile of each of your apps before moving, and choose the cloud provider that best fits.
Latency is one metric that many companies forget about when moving to the cloud. Remember, many cloud data centers will be far away from where your apps and data are needed most. For some apps, any latency will affect performance and increase costs, as people and customers can’t use the app efficiently. One way to solve this issue is by utilizing an edge computing model for the data that absolutely can’t afford any performance delays. By keeping this critical data and compute near users, the edge reduces frustration, performance hiccups and, likewise, costs.
After You’re There
Once you’ve determined which cloud providers you’re going to use for which apps, the temptation is to sit back and let things run. This, of course, is where companies get into trouble.
Real time usage is often different than tests indicate, and demands on apps and data shift over time. Therefore, continuous management of your multi-cloud environments over time is a crucial aspect of keeping costs where you expect. We’ve found a two-pronged strategy works best.
First, closely monitor your workloads. After your multi-cloud environment is up and running, you’ll no doubt find that some workloads would run more efficiently and less expensively on a different platform. This is not only OK, but expected.
Second, constantly monitor your budget. Are your costs matching what you thought they’d be? Are the invoices from any providers higher than expected? Are any lower? Develop a historical spending analysis to nip potential budget problems in the bud. Fees vary by provider, so think about how much you’ll need to access data from each app and see how the spend lines up.
Remember, this is why you have a multi-cloud approach in the first place. If one vendor’s bill is higher than expected, then you could be using its services inefficiently. Look to see what you can shift off that vendor to a different option, and what apps on other platforms might benefit from replacing them.
A multi-cloud model has great potential for performance and cost optimization. That doesn’t mean you can “set it and forget it,” though. The only way to guarantee your multi-cloud implementation will deliver on its great potential is through careful planning, and constant monitoring and evaluation.
Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.