re:Invent 2023 (Las Vegas) – In the era of generative AI, AWS chose for the first keynote of its re:Invent 2023 event, with Peter DeSantis, Senior Vice President of AWS Utility Computing, to focus on… serverless.
AWS has been working on serverless technology for years. This term is used to describe computer applications that depend entirely or largely on applications and services that operate in cloud mode.
The goal for AWS is obvious. This involves managing the server layer in the cloud for the client, and letting them give business specifications to start an application. And to bill him for this management service of course in addition to the storage and computing services.
Automation of relational database management in the cloud
AWS Lambda is the gateway to serverless architectures at AWS. “AWS Lambda runs your business code and automatically manages the underlying compute resources for you,” Amazon says on its website. AWS Lambda then manages the other serverless bricks.
The company is therefore announcing today the marketing of serverless services AWS Aurora, Elastic Cache and a new feature on Redshift serverless.
Aurora Serverless is for cloud-based relational databases. The idea is to automate sharding, that is to say the allocation of IT resources to run databases in the cloud. And this most often involves base segmentation.
AWS therefore offers an “unlimited database” feature, which manages the sharding work in an automated manner. The goal ? When configuration needs change, Aurora serverless directly makes the necessary adjustments to automatically manage shards.
Serverless caching
The ElasticCache Serverless product is a serverless caching service that is inserted between application servers and databases. Enough to improve response times while reducing database costs.
The idea is to be able to set up highly available caches with response times in the microsecond range, ready to scale to large volumes of data “in less than a minute”.
Finally, AWS announced that Redshift Serverless, a feature that has been around since 2021, now uses AI to automatically optimize and scale Amazon Redshift data warehouses based on query patterns and data volumes.
The big idea behind these announcements is that because AWS manages all the hardware in the background and provides just the amount of resources you need, increasing capacity when needed, resource efficiency is particularly optimized.