Wednesday, February 17, 2021
With serverless computing, you pay for the resources that your application consumes. The longer your serverless code runs, the more you pay. The longer your serverless code remains idle, waiting for a backend response, the more you pay. The longer your database handles a write-request sent by serverless code, the more you pay. At the extreme, the idle time of serverless functions can represent two-thirds of your monthly bill, and you question why somebody selected the serverless architecture in the first place.
In this session, you learn how to use in-memory computing to reduce idle time and accelerate the execution of serverless code. The provisioning of an in-memory cluster is a prerequisite, not a game-changer. Dramatic differences become apparent when you consider several tactics while fastening serverless code to your just-provisioned in-memory cluster.
Download these images to your phone and post using the Instagram app.