Amazon has published reference architectures to assist customers of its Amazon Web Services (AWS) platform in implementing cloud-based back-end services using the Lambda feature that runs code without the need to provision server instances.
AWS Lambda enables developers to create code that runs on the cloud platform in response to some event or trigger. Lambda takes care of everything required to run and scale up the code as required, and customers pay only for the compute time consumed when the code is actually running.
With no need to provision server instances to run code, Lambda is thus a very attractive option for operating services that may need to rapidly scale up and down in response to peaks and troughs in demand, and AWS is giving developers a push in this direction with some new reference architectures.
"Building your applications with only managed components has become very popular, and AWS Lambda plays a crucial role in that," said Amazon chief technology officer Werner Vogels, writing on the All Things Distributed blog. Vogels lists a handful of reference architectures, including those for a mobile services back-end and an Internet of Things (IoT) back-end.
The reference architectures, available to download from GitHub, demonstrate how to use AWS Lambda, in conjunction with other AWS services, to build a serverless back-end for each specific application. 'Serverless' in this case means that the customer does not have to worry about provisioning servers to operate the framework.
The Mobile Backend reference architecture, for example, creates a service that enables users to upload photos and notes using Amazon Simple Storage Service (Amazon S3) and the Amazon API Gateway. The notes are stored in DynamoDB, and processed asynchronously using DynamoDB streams and a Lambda function to add them to an Amazon CloudSearch domain.
Meanwhile, the IoT Backend reference architecture uses AWS Lambda in conjunction with Amazon Kinesis, DynamoDB, Amazon S3 and Amazon CloudWatch to build a serverless system for ingesting and processing sensor data.
"By leveraging these services you can build cost-efficient applications that can meet the massive scale required for processing the data generated by huge deployments of connected devices," Vogels said.
Curiosity rover is in 'normal mode' but not transmitting scientific data back to base
NatWest outage comes a day after Barclays' IT systems shut out customers and staff
The ICO is concerned with AggregateIQ's retention and processing of data used in the Brexit referendum
Map selection, quick menus for grenades and healing items, and automatic reload all coming in PUBG update #22