January 24, 2023 · 2 min read
AWS Credits are promotional coupons offered by AWS that encourage companies to build their IT infrastructure on AWS.
AWS credits are promotional coupons offered by Amazon Web Services (AWS) in order to allow new customers to use their AWS services for a certain amount of time. Credits can be redeemed straight from AWS’s billing page and are applied until they are exhausted or expired.
If you’re reading this, chances are your AWS credits are about to run out. Firstly, check whether or not you are eligible to receive more credits. Startups can receive up to $100,000 in lifetime AWS credits ($250,000 for qualifying companies with ML workloads), so confirming your eligibility to receive more credits is always the best place to begin.
If you are ineligible to receive more credits, reducing your consumption of AWS resources is the only way to extend the ones you currently have. Bearing in mind that building a good, cost effective architecture is the best way to keep costs low, there are exercises that you can do today to reduce your consumption of AWS resources short of re-architecting your entire organization.
A full list of AWS-prescribed cost optimization exercises is available here to view, but included below are the low-touch exercises that can quickly and significantly lower costs:
Using AWS Cost Explorer Resource Optimization can generate a report of EC2 instances that are either idle or have low utilization. Stopping these instances or downsizing them into smaller workloads will reduce the associated costs. The AWS Instance Scheduler is a great tool to automatically stop idle instances; The AWS Operations Conductor is a helpful tool to automatically resize the EC2 instances from the Cost Explorer report.
Spot instances are spare, unused EC2 compute capacity that is available to customers at a lower price than On-Demand instances. It is important that applications using EC2 Spot Instances be stateless, fault-tolerant, and loosely coupled. You can follow this tutorial to learn how to create a stateless, fault-tolerant workload using AWS EC2 Auto Scaling to deploy Spot Instances.
In exchange for committing to either a 1 or 3 year term, AWS will provide significantly discounted hourly rates for your RDS, Redshift, ElastiCache, and Elasticsearch usage. To get a view of what you have available to save, use the recommendations provided in the AWS Cost Explorer under “Reservations - Recommendations” to guide your purchasing decisions.
Similar to RDS, Redshift, ElastiCache, and Elasticsearch, AWS will provide a significantly discounted hourly rate (up to 72%) for your EC2 compute resources in exchange for committing to a 1 or 3 year term. Again, referencing the purchase recommendations provided in the AWS Cost Explorer under “Reservations - Recommendations” will demonstrate where you can immediately lower the run rate of your compute resources.
Leveraging the above exercises is a great way to begin reducing your consumption of AWS resources so you can get the absolute most out of your credits. By the time they run out/expire, you can start building on top of a very cost-conscious cloud environment!
* The contract terms for Reservation purchases are often non-starters. At Usage AI, we underwrite Reservation purchases - in the event you’ve overcommitted to capacity, our platform automatically exchanges the remainder of that commitment with another AWS customer!
Want to learn how Usage can help extend your credits? Book a time with a member of our team at our calendar link here!
Kane Daniel is the Head of Savings at Usage AI.