How can I control cloud costs while still enabling high performance? originally appeared on Quora: the place to gain and share knowledge, empowering people to learn from others and better understand the world.
Controlling cloud costs can be complicated – but it usually boils down to one very important factor: data awareness. You have to be aware of your data usage, workloads and access needs in order to understand how much cloud resources to use. When it comes to data awareness, there are 3 common mistakes:
- Not thinking critically about which workloads benefit from the cloud: The cloud offers some compelling benefits, but those benefits aren’t universally worth the expense. Do the due diligence to really understand the technologies that best serve your workloads, rather than peanut buttering a cloud solution across your business.
- Not looking at data dynamically: Modern workloads like machine learning and remote collaboration demand dynamic access to data, and understanding your costs in the cloud will demand you be aware of how much compute resources you need.
- Not monitoring peaks and valleys: The cloud is a utility, and one of its primary benefits is its dynamic ability to meet different needs at different times. Your team might need more cloud compute for a particular project, or according to seasonality. In order to anticipate cost, you need to understand where these peaks and valleys happen, what drives them, and adapt your infrastructure to avoid paying for unused compute power.
By getting a better understanding of how you’re using your data and where your data needs to live, you’ll be smarter about your spending.