AWS Lambda

Amazon Web Services (AWS) Lambda is a compute service that executes arbitrary Python code in response to developer-defined AWS events, such as inbound API calls or file uploads to AWS' Simple Storage Service (S3).

Why is Lambda useful?

Lambda is often used as a "serverless" compute architecture, which allows developers to upload their Python code instead of spinning and configuring servers, deploying their code and scaling based on traffic.

Python on AWS Lambda

Lambda only had support for JavaScript, specifically Node.JS, when it was first released in late 2014. Python 2 developers were welcomed to the platform less than a year after its release, in October 2015. Lambda now has support for both Python 2.7, 3.6 and 3.7.

Python-specific AWS Lambda resources

• Serverless Slash Commands with Python shows how to use the Slack API to build slash commands that run with an AWS Lambda backend.

• Zappa is a serverless framework for deploying Python web applications. It's a really slick project and used even by internal AWS developers for their own application deployments.

• How to Setup a Serverless URL Shortener With API Gateway Lambda and DynamoDB on AWS builds a non-trivial URL shortener application as an example Python application that runs on Lambda.

• How we built Hamiltix.net for less than $1 a month on AWS walks through setting up a full website that runs on AWS and scales with the Lambda free tier to minimize spend despite large traffic spikes.

• Deploying a serverless Flask app to AWS Lambda using Zappa provides a screen capture of one developer deploying their application to Lambda.

• Building Scikit-Learn For AWS Lambda follows up on the Using Scikit-Learn In AWS Lambda post which shows how to perform scientific computing with Python packages on AWS Lambda.

• Creating Serverless Functions with Python and AWS Lambda explains how to use the Serverless framework to build Python applications that can be deployed to AWS Lambda.

• Code Evaluation With AWS Lambda and API Gateway shows how to develop a code evaluation API, to execute arbitrary code, with AWS Lambda and API Gateway.

General AWS Lambda resources

• Getting started with serverless on AWS is a wonderful tutorials, example projects and additional resources guide created by a developer who used all of these bits to learn AWS services herself.

• Security Overview of AWS Lambda (PDF file) covers their "Shared Responsibility Model" for security and compliance. Although the paper bills itself as an in-depth look at AWS Lambda security it is really more of a high-level overview, but still worth the read.

• Reverse engineering AWS Lambda is an incredible, in-depth analysis of the author's work investigating the black box of how Lambda works and what he learned from it.

• The AWS Lambda tag on the official AWS blog contains all the related first-party tutorials

• Serverless Cost Calculator estimates the amount that AWS would charge based on Lambda exeuctions, average execution time and memory needed per execution.

• Serverless at Nordstrom is an awesome real-world story with the architecture behind a serverless AWS Lambda application deployment at Nordstrom.

• How was your experience with AWS Lambda in production? has a good discussion of some of the benefits and issues that developers had as of mid-2017 with using Lambda for production applications.

• Passwordless database authentication for AWS Lambda shows how to use a MySQL backend from your Lambda functions.

• How does language, memory and package size affect cold starts of AWS Lambda? investigates the performance implications of various Lambda settings.

• Best Practices for AWS Lambda Timeouts explains some of the current hard upper limits on AWS timeouts, such as 5 minutes for Lambdas, when explicitly set that high, as well as 29 seconds for API Gatway requests. There is also good advice on how the circuit breaker pattern should be applied to your Lambdas and ultimately why low time outs are likely the best way to go to prevent your application from becoming entirely unresponsive.

• X-rays for Flask and Django Serverless Applications is an instrumentation, monitoring and debugging service built into AWS Lambda specifically for Python web frameworks running on the service.

• Cutting Through the Layers: AWS Lambda Layers Explained explains how AWS Lambda now offer a "Bring Your Own Runtime" by exposing the layers that were previously controlled exclusively by Amazon. There is an overview of the layers and why they matter for customizing your functions.

What's next after learning about AWS Lambda?