Hello, and welcome to this lecture, where I will discuss how serverless architectures in AWS can help you minimize cost. Adopting a serverless architecture can help you realize cost savings in two different ways: first, in terms of the time and effort you’ll save by not having to provision and maintain any infrastructure, and second, because you’ll never need to pay for idle resources that may sit unused for long periods of time. Now if your solution relies on any long-running processes or has constant high levels of demand, a serverless architecture may not save you any money. In fact, it could even end up costing you a lot more than a traditional instance-based deployment. But for just about any other use case, a serverless architecture is likely to save you both time and money. And you can learn more about the benefits of serverless architectures over traditional deployments, as well as see some real-life cost comparisons between the two, by checking out my course, When to Go Serverless. But I do want to quickly take a minute here to walk through a couple examples of popular serverless architectures you may want to consider leveraging within your applications, especially if you have an opportunity to develop from scratch or refactor an existing application.

So perhaps the most popular example of a modern serverless architecture is one that leverages API Gateway and Lambda to deploy web APIs that route requests to Lambda functions. And these are useful for both APIs and microservice architectures. So with API Gateway, AWS will manage your API endpoint so you don’t need to worry about provisioning any elastic load balancers or front-end application server instances. And of course, Lambda allows you to upload your code without needing to provision any back-end server instances to run it. So the savings here can be very significant as both of these services leverage a pay-per-use model. This means you’ll only ever pay for resources when they are in use, and you’ll never need to worry about paying for any unutilized or underutilized instances.

To learn more about designing and configuring HTTP endpoints using Lambda as a backend, I encourage you to check out this hands-on lab.

We’ve seen how Amazon RDS can be used to minimize the ongoing maintenance and administration costs for relational databases. But if your application needs to make use of a key-value store or what’s known as a NoSQL database instead, look no further than DynamoDB. DynamoDB is a serverless database with both provisioned and on-demand capacity modes, allowing you to choose the most cost-effective option based on your anticipated level of demand. DynamoDB also supports encryption at rest, automated backups, and high availability without any additional administrative overhead.

To learn more about DynamoDB as well as many other cost-saving serverless services that you can incorporate into your AWS solution architectures, I encourage you to check out this course.