AWS goes all in on serverless, on-demand solutions with quartet of launches

Amazon's new slate of cloud-based offerings use serverless and on-demand versions of existing solutions to entice cost-hesitant users and increase agility for businesses requiring rapid scalability.
Written by Michael Gariffo, Staff Writer

AWS CEO Adam Selipsky debuted a quartet of new serverless and on-demand solution for its Redshift, EMR, MSK and Kinesis solutions.

The overall goal of the update is to create a more agile channel through which customers can utilize cloud-based services to cost-effectively fulfill service obligations at scale, Selipsky noted.

For Amazon Redshift this means a serverless version that will run without any required data warehousing infrastructure, while still providing the same access to structured and semi-structured SQL data across warehouses, operational databases, and data lakes. The service is compatible with Amazon S3, Aurora, and Relational Database Service (RDS). It also supports automatic provisioning and scaling to adapt to changing demand.

Similar automatic scalability is also now supported by Amazon Kinesis Data Streams, a flexible, on-demand version of the streaming solution that provides real-time processing and provisioning that adapts to usage fluctuations. The service supports dynamic shard scaling, with each shard able to ingest 1MB/s and 1,000 records/second, with 2MB/s of supported egress. Stream capacity can also be adjusted via the Kinesis Data Streams API to better balance even massive workloads.

Handling large workloads is where Amazon also expects its new EMR Serverless option to shine. The solution was presented as "easy and cost-effective for data engineers and analysts to run petabyte-scale data analytics in the cloud." Configuration, management, optimization, and security are all handled automatically, with support for running applications built on open-source frameworks such as Apache Spark, Hive, and Presto.

Lastly, MSK Serverless is designed to be a new type of cluster that makes it easier for developers to run Apache Kafka without manually handling capacity management, AWS said. Instead, these needs are handled via automation of provisioning and scaling compute and storage resources to more accurately control the data that is streamed and retained.

The other main selling point of all four AWS solutions, aside from their increased agility, is the on-demand and as-needed pricing schemes they bring with them. This potentially offers significant savings to users that were underutilizing existing versions of each service, or where hesitant to buy into any of the four offerings over fears of exceeding their cloud budgets by over-purchasing for their specific needs.

Selipsky noted that the company expects all four offerings to be particularly attractive to businesses that need to rapidly scale their online presences, such as ticket sellers bracing for a major event or online retailers responding to unexpected spikes in sales. 

Editorial standards