Lambda Cold Starts analysis, updated daily


This is a project that benchmarks the performance of AWS Lambda function cold starts across various runtimes, including AWS-supported ones and custom runtimes like Go, Rust, and Quarkus. A simple "hello world" function is written in each runtime and packaged into a zip file, which is uploaded to an S3 bucket. Every day, these functions are deployed and invoked 10 times as cold starts, with the results logged to a DynamoDB table. After all invocations, the data is aggregated and committed to this repository as a JSON file, which is then displayed on a static website hosted on GitHub Pages. The goal of the project is to provide up-to-date information on Lambda Cold Starts performance that can be easily accessed and compared across different runtimes. The project is not associated with any companies and has not been reviewed or certified by any company. Users are invited to hack/fork/send PR and create their own benchmarks.
Company Screenshot