Cold start times

Last time I was investigating Swift on Server for AWS Lambda, one of the areas for improvement was noted to be cold start times. Specifically, the added binary size due to the ICU dependency is listed as a performance inhibitor on the runtime GitHub repo.

Larger packages size (Zip file uploaded to AWS Lambda) negatively impact the cold start time, since AWS needs to download and unpack the package before starting the process.
Swift provides great Unicode support via ICU. Therefore, Swift-based Lambda functions include the ICU libraries which tend to be large. This impacts the download time mentioned above and an area for further optimization.

Now, I see on the Swift.org website that start times are “negligible”.

Finally, quick boot times make Swift a perfect fit for serverless applications such as Cloud Functions or Lambda with negligible cold start times.

Have the cold start time issues been resolved since the GitHub runtime repo’s README was written?

Have you measured any times? Last time I looked into our cold start times they were around 50ms.

One thing that can significantly speed up cold start times is to increase the memory allocations for the lambda