Many companies have the development of digital initiatives high on their agenda. Together with the growing adoption of cloud, this means a bigger part of IT budgets is moving to the cloud. Gartner predicts public cloud spending will keep growing rapidly, from $258B in 2020 to $364B in 2022. Managing cloud costs can be ‘cloudy’. Can you use your data to optimize your cloud spending?
Digital projects are typically developed on a foundation of cloud technology. In addition, we increasingly see IT enterprises moving their workloads to the cloud. And, while there are many benefits to this, managing costs is often a hassle. Whether you are able to manage all cloud subscriptions centrally, or each department is using their own credit card to generate cloud instances – the cloud invoices you receive rarely provide usable information in terms of application, department or cost center. This makes it difficult to identify on which part of the business your money is spent. Spreading workloads over multiple cloud subscriptions (Azure, AWS, Google) means an extra layer of complication since you need knowledge of each of these providers. Another challenge lies in the elastic nature of cloud, making it complex to forecast your spending.
All of this means you are probably losing money by spending more than you should. That is why we developed a cost control methodology to optimize cloud costs. It is based on three steps: subscription optimization, cost insights, and rightsizing.
The first step in cloud cost optimization is to analyze existing subscriptions. A multi-cloud strategy will lead to multiple subscriptions with cloud providers like Azure, AWS, and Google. Through our knowledge of the subscription models of these providers, and our analysis methodology, we start with identifying low-hanging fruit. Examples are consolidating subscriptions to generate discounted volume pricing or switching to a different payment plan. Workloads that predictably run 365 days per year may be committed to a long-term subscription, giving you better pricing than when they are subject to a flexible, hence more expensive, subscription. Optimizing subscriptions often yields savings ranging from 20% to 40%.
The second step is to improve cost insights. We start by tagging individual resources to providers, applications, cost centers, etc. Categorizing each asset enables us to provide valuable information on where your money is being spent. This information is accessible in a handy online tool and can be integrated with your existing reporting, like Power BI, via an API. Advanced analytics enables you to create usage forecasts, detect anomalies, and generate thresholds and alerts. These insights give you increased control over your cloud costs.
The third step is rightsizing your assets. Because it’s easy to create cloud assets, it’s also easy to make them ‘just a little bit bigger, just to be on the safe side’. However, by measuring the actual consumption of cloud resources, we create a baseline of usage patterns. This allows us to identify areas where resources might be allocated more efficiently, which we translate into usage recommendations that we discuss with the product owner. For example, if an application has a peak usage one day per month and is dormant the rest of the time, it typically makes sense to adjust the allocated resources accordingly to create significant cost savings. Our automation tools then automatically schedule to scale up and down in a completely hands-off manner, making these cost savings an ongoing activity.
Consequently, when more of your company begins to adopt cloud, it becomes necessary to pay attention to cloud cost optimization. Cost control should not be a one-time exercise, but part of your ongoing IT operations. Using data to optimize your subscriptions, generate cost insights, and rightsize your assets will enable you to directly improve the ROI of your digital projects.