Today, many companies want to give access to back-end services to other applications through an application programming interface (API). It usually requires a tremendous amount of effort and resources to create, maintain and secure the API and make sure it’s operating correctly. Amazon hopes to simplify that by essentially creating an API as a Service.
A successful API could deal with millions or even billions of calls a day and this puts tremendous pressure on its servers and the company to make sure it’s operating smoothly. Everyone wants other services tying into theirs, but it takes work. If it’s not going smoothly, developers will simply find an alternative that is.
It also ties into other Amazon services (naturally) such as Amazon Elastic Compute Cloud (Amazon EC2) for compute power and AWS Identity and Access Management (IAM) to verify and authenticate API requests. Developers can also take advantage of Amazon Lambda, announced at Amazon’s re:invent conference last fall, which lets programmers create programmable trigger events based on certain actions.
For instance, if your API is getting extremely popular and it’s causing latency, you could have a trigger event that kicks in at a certain amount of API calls to spin up new compute resources.
This announcement puts AWS in direct competition with several companies which have been offering API management services for some time including Apigee, which went public last April, Mashery and others. Just because a big player like Amazon enters a particular market isn’t a guarantee it will succeed, but it has to making these other players a bit uncomfortable right about now.
Amazon is charging $3.50 per million API calls received, plus the cost of data transfer out, in gigabytes. It also has additional charges depending on the amount of cache memory you’re using. You can find a list of the data transfer and caching rates on the Amazon API Gateway pricing page.
There is also a free tier, which includes one million API calls per month for up to 12 months