Amazon launches DynamoDB cloud database service

Amazon Web Services has unveiled an online database service designed to let customers store and analyse data without any of the normal administrative overheads, allowing them to scale up as their data requirements dictate and apply analytics tools to their datasets.

Amazon DynamoDB is a fully managed database service based on NoSQL non-relational technology that allows customers to quickly provision a database and scale it up as needed, on a pay-as-you-go basis.

DynamoDB also integrates with Amazon's Elastic MapReduce (EMR), a pay-as-you-go Hadoop framework hosted on Amazon's cloud infrastructure, which allows customers to perform complex analysis of their data.

Amazon said traditional databases are not designed to scale up to meet the performance needs of the latest applications, which can generate huge volumes of data garnered from social networks and other sources.

This is one area where Amazon has gained expertise from operating the AWS infrastructure itself, according to chief technology officer Werner Vogels.

"Amazon DynamoDB is the result of everything we've learned from building large-scale, non-relational databases for Amazon.com and building highly scalable and reliable cloud computing services at AWS," he said.

As a managed service, DynamoDB removes the administrative process from the customer and addresses scalability issues by automatically partitioning and re-partitioning data as needed to meet latency and throughput requirements.

DynamoDB utilises SSD storage for high performance, with data replicated synchronously across multiple AWS Availability Zones to provide high availability and data durability, Amazon said.

Thanks to this, the DynamoDB service is claimed to offer low, predictable latency, with database read and write operations taking just a few milliseconds.

As with AWS's EC2 cloud service, DynamoDB is available on a free tier for customers to evaluate. This offers 100MB of storage and up to 40 million requests per month free of charge.