It’s the stuff of nightmares for every CFO of every company that uses cloud computing: finding out at the end of the month that you’ve blown way past your budget and have overspent significantly. It happened in a big way at Adobe just a couple of years ago: “an Adobe development team last fall unintentionally racked up $80,000 a day in charges for a computing job that ran on Azure and wasn’t discovered for more than a week”. If you’ve ever used a cloud computing platform, you know that this is a very real possibility at any given time. So what’s the best way to avoid it?

Identifying Anomalies in Cloud Spend

A good practice for anyone who wants to control Kubernetes and cloud costs is to set a budget. Just like with home budgets, cloud budgets typically define monthly spending limits. The problem is that you can use up a whole month's budget in a few days with little to no warning. Harness has figured out how to avoid this situation, with cost anomaly detection.

How does Cost Anomaly Detection work?

Here’s how it works: Harness Cloud Cost Management has access to the cost and usage reporting features of the major cloud providers. It pulls this data (hourly for Kubernetes) and uses ML algorithms to look for anomalies and trends which indicate that you will incur significant increases to your cloud costs. When an anomaly is identified, you get alerted. In the shortest amount of time possible, you know when there is a major cloud spend problem brewing that needs to be addressed.

Kubernetes Benefits and Cloud Costs

The team at Harness also understands kubernetes basics realizes that not every significant increase in cloud spend is a mistake. Sometimes, you intentionally deploy new systems and services that will cause a spending increase. We have this scenario covered with a feedback mechanism. When an anomaly is detected, simply push the “thumbs down” icon to let the system know that this is not really an anomaly. You can also add a comment that describes the scenario so that everyone on the team is up to speed. Oh, and out ML algorithms learn weekly, monthly, and yearly seasonality patterns too!

Start Your Cloud Cost Savings Journey Today

Every company is overspending on Kubernetes and cloud costs. It’s time to take control and equip ourselves with the right data and the right tools to combat this common and expensive issue. Get a free trial of Harness Cloud Cost Management today and get started down the path of reducing your cloud spend.