Talks, Presentations & Events
The demo validates improvements of the application "AWS Feature Roll" on lambda coming from findings out of a load test against the application deployed on Beanstalk.
The AWS Feature Roll app is basically a JSON representation of the "What's new?" RSS feed from https://aws.amazon.com/de/new/
/feature– get a JSON Document with features
/slides– get a reveal.js Slidedeck, ZIPed
You can filter both endpoints with
The first version is deployed on Beanstalk using a classic Beanstalk stack:
Watch the video of the Load Test against Beanstalk here:
/slidesnow checks if the requested slide deck is already available on S3, if not a third function
createSlideDeckwill create it
/slideswill know the future url of the slide deck asap and can respond to the user.
/featuresis a lot faster
/slidesis a lot faster, since the slidedeck creation is async
It depends. :)
We can't say that Beanstalk is slower. An async solution for
/slides can be implemented on Beanstalk as well.
However, a simple app like this is maybe the right use case for Lambda.
Find the slides of #AWSSummit Berlin at the the presentation page.
If you have any questions get in touch, please: firstname.lastname@example.org