Per Flexera’s 2021 State of the Cloud Report, around 30% to 35% of cloud computing resources purchased from public cloud providers such as AWS go to waste. Moreover, on average organizations go 24% over the planned annual cloud budget.
The better news, however, is that AWS provides a host of native cloud cost optimization tools you can use to cap your spending. Here are the four services we recommend.
AWS Cost Explorer
As the name implies, AWS Cost Explorer is an add-on cost monitoring interface that lets you slice-and-dice AWS usage patterns and zoom in on costs for all connected instances, both reserved and on-demand.
Cost Explorer visualizes historical and current usage in neat charts. Plus, it automatically creates 12-month forecasts, so that you can plan your cloud computing budgets ahead and prevent cost overruns.
Here’s how to get the most out of AWS Cost Explorer:
- Use resource level granularity control to zoom in on individual EC2 instances and identify the best candidates for switching to reserved instances (which can be up to 75% cheaper).
- Analyze hourly level granularity to determine if you can switch off certain instances outside of working hours. Instances, reserved for development or test environments are strong contenders.
You can use the web app free of charge or integrate AWS Cost Explorer API into another analytics dashboard. In the case of the latter, you’ll pay $0.01 per 1,000 UsageRecords per month.
AWS Budgets helps you stay within your allocated cloud spending range by setting spending or usage caps for your projects.
First, assign maximum budgets to tagged projects. Then receive alerts when the actual or forecasted costs are nearing the set threshold. Also, you can configure AWS Budgets to notify you if you are not profiting enough from your reserved instances (RI) e.g. if the usage drops below 80% on any given day.
Arguably, the biggest boon of AWS Budgets is simple process automation. Instead of jumping manually on the alert, you can program respective actions whenever a certain event occurs. For example, when a certain project is about to spill beyond the set spending cap, you can code to automatically apply an IAM policy that will deny the ability to provision extra resources.
AWS Cost and Usage Report
If you prefer to analyze your AWS cost using a custom data science model, download your copy of an AWS Cost and Usage Report. This is a jumbo-sized record of all your AWS billing and usage data available across all accounts. Plus extra metadata on AWS services, pricing tiers, RIs, and Savings Plans.
The dataset is available in CSV format and is auto-added to your S3 bucket. You can also easily ingest it into a data warehousing service such as Amazon Redshift for hot-key analytics access.
AWS Cost and Usage Report provides a high level of granularity (hourly, daily, monthly) and lets you filter out records by product codes, usage, or operations type. Also, you can shape up the data using Cost Allocation tags and Cost Categories.
Pro tip: Don’t forget to consolidate your AWS accounts before requesting the report copy. Without doing so, you’ll only get a partial view of your usage patterns.
Amazon S3 Analytics (S3 Storage Lens)
Cloud data storage can easily rack up your cloud bill. Thus, it’s important to regularly move assets between hot and cold storage types. Amazon S3 Analytics scans through your stored data and helps match different assets to the right storage class.
Using this tool you can:
- Track object storage usage and activity across all accounts.
- View insights on a bucket and prefix level
- Analyze storage patterns using 30 metrics
- Receive contextual recommendations on class selection
- Auto-apply data protection best practices
- Compare storage costs across regions
- Visualize activity trends, forecasts, and historical patterns
Additionally, Amazon S3 analytics can be queried with data with Amazon Athena if you need an even deeper analysis.
Runner Up: AWS CloudWatch
The above four AWS cloud costs tools can help managers discover all the data they need to make informed decisions on current cloud spending and future budgets. But to truly contain costs, both the management and the end-users (developers) should be well-aligned on the allocated spending, resource tagging decisions, and overall cloud usage policies.
To make that happen, both teams need to understand what goes into the total cloud spending bill. That’s where AWS CloudWatch comes in handy. AWS CloudWatch is an application and infrastructure monitoring solution for DevOps teams, offering analytics about resource utilization, infrastructure health, and overall performance. Such “grassroots” data is essential for assigning the optimal budgets for different projects and implementing policies that allow scalability.
Managing AWS Cloud Costs Requires Precision
AWS is a complex ecosystem with many moving parts. While you have plenty of native analytics tools for cost forecasting, budgeting, and analysis, the cost-containment decisions are still yours to make. If you need extra guidance for cloud cost optimization, Indellient has launched a new Cloud Cost Management service — a managed offering, where we help our clients optimize their infrastructure and curtail their cloud bill without undermining the application performance.