When to use API Gateway vs. Lambda Function URLs Read on my blog Read time: 4 minutes “Lambdalith” is a monolithic approach to building serverless applications where a single Lambda function serves an entire API, instead of one function per endpoint. It’s an increasingly popular approach. It provides portability between Lambda functions and container applications. You can lift and shift an existing application into Lambda without rewriting it. You can use web frameworks you are already familiar with, and lean on the existing ecosystems of tools, ORMs and middleware. It also makes testing easier, because you can apply familiar testing methodologies. Tools like the AWS Lambda Web Adapter [1] have made this approach more accessible. In addition, Lambda Function URLs [2] also work well with this pattern. Which brings up a good question. “In 2024, if you want to build a REST API using serverless technologies. Should you use Lambda Function URLs or API Gateway?” Here are some trade-offs you should consider when making this decision. Function URL pros
Function URL cons
Where Function URL makes senseIf you want to (i.e. not forced into it by the choice to use Function URL) build a Lambdalith and you don’t need any of the additional features that API Gateway offers. Then Function URLs make sense — it’s cheaper, faster and has fewer moving parts. Similarly, if you need to return a large payload (> 10MB) or to run for more than 29s, then Function URL can also make sense. That is if you can’t refactor the client-server interaction. Given the limited support for authentication & authorization, it’s not suitable for user-facing APIs. These APIs often require a Cognito authorizer or a custom Lambda authorizer. This leaves function URLs as best suited for public APIs or internal APIs inside a microservices architecture. By “internal API”, I refer to APIs that are used by other services but not by the frontend application directly. These APIs usually require AWS_IAM auth because the caller is another AWS resource — a Lambda function, an EC2 instance, an ECS task, etc. API Gateway pros
API Gateway cons
When API Gateway makes senseGiven the vast array of features that API Gateway offers, it makes sense in most cases if you’re OK with the additional cost that comes with the convenience. The 29s and 10MB response limits can be problematic in some cases. But they can be mitigated with patterns such as Decoupled Invocations [6] and S3 presigned URLs [7]. However, these workarounds require you to refactor the client-server interaction, so they are not always viable. SummaryBecause of its flexibility, I prefer API Gateway over Function URLs or ALBs. But Function URL is a useful tool, especially when cost and performance are your primary concerns. It’s also an important lift-and-shift option for people migrating from an existing EC2 or container-based application. It lets them enjoy the benefits of Lambda without rewriting their application. Finally, as Ben eloquently puts it, this tradeoff represents a deeper choice we often face. What’s easier at author time might mean more ownership at runtime. For example, we don’t get per-endpoint metrics from function URLs so we have to bring that capability to the table ourselves and be responsible for them. Something for you to think about ;-) Links[2] AWS Lambda: function URL is live! [3] Introducing AWS Lambda response streaming [4] Embedded metric format specification [5] API Gateway: Why you should use Service Proxies [6] How to use the Decoupled Invocation pattern with AWS Lambda and serverless [7] Hit the 6MB Lambda payload limit? Here’s what you can do Whenever you're ready, here are 3 ways I can help you:
|
Join 17K readers and level up you AWS game with just 5 mins a week.
Step Functions is often used to poll long-running processes, e.g. when starting a new data migration task with Amazon Database Migration. There's usually a Wait -> Poll -> Choice loop that runs until the task is complete (or failed), like the one below. Polling is inefficient and can add unnecessary cost as standard workflows are charged based on the number of state transitions. There is an event-driven alternative to this approach. Here's the high level approach: To start the data migration,...
Lambda Durable Functions comes with a handy testing SDK. It makes it easy to test durable executions both locally as well as remotely in the cloud. I find the local test runner particular useful for dealing with wait states because I can simply configure the runner to skip time! However, this does not work for callback operations such as waitForCallback. Unfortunately, the official docs didn't include any examples on how to handle this. So here's my workaround. The handler code Imagine you're...
Lambda Durable Functions is a powerful new feature, but its checkpoint + replay model has a few gotchas. Here are five to watch out for. Non-deterministic code The biggest gotcha is when the code is not deterministic. That is, it might do something different during replay. Remember, when a durable execution is replayed, the handler code is executed from the start. So the code must behave exactly the same given the same input. If you use random numbers, or timestamps to make branching...