CloudFront has the ability to cache various types of data that can be presented via HTTP or HTTPS. To configure a CloudFront distribution, we need to specify an origin and its features. An origin can be an S3 bucket or an arbitrary DNS-resolvable HTTP/HTTPS server that can be hosted within AWS or anywhere on the internet. The content delivery network (CDN) is designed to deliver the distributions with the lowest possible latency. We also have some ability to control how the cache is retained and expired within CloudFront, allowing us to get the best out of the CDN.

CloudFront is fully addressable via its API and excels at programmability and the features that are offered as a service to extend its capability. We can control the behavior of the caching environment, define how to forward requests to the origin, choose what kind of compression to use, modify responses coming directly out of CloudFront, and detect the client type within the cache. CloudFront also has the ability to integrate with Lambda@Edge so developers can introduce cone execution patterns as close to the end users as possible. This can improve the performance and user experience in applications on various types of devices and operating systems.

Any data transfer between the AWS services and CloudFront is free, meaning we can cut down costs for delivering content from S3 as we are not subject to outgoing transfer pricing when using CloudFront. This also means CloudFront can be a cost-effective way to deliver content to our users, as we have the ability to implement it on an arbitrary basis and pay only while using it. Amazon also offers the ability to set up a committed private contract for CloudFront so that clients can take advantage of the service at a reduced price when they know their traffic patterns and are looking for a long-term solution.