Amazon has unveiled its big data service Amazon DynamoDB, a NoSQL database that stores data on faster solid state drives.
NoSQL is a non-relational database that has become popular with e-commerce sites that deal with high volumes of data. It was also tapped by Oracle as a key component of its "big data appliance", pitched for the management of dynamic data streams, such as web log, sensor or smart meter data.
Amazon Web Services chief technology officer Werner Vogels said in a blog post the service would help cloud startups overcome constraints in rapidly scaling an application.
"With Amazon DynamoDB, developers scaling cloud-based applications can start small with just the capacity they need and then increase the request capacity of a given table as their app grows in popularity. Their tables can also grow without limits as their users store increasing amounts of data."
While the database is ideal for web applications and has been used within Amazon's shopping cart and session service, it may also prove useful for enterprise applications, Vogels told GigaOm.
Vogels explained that Dynamo needed to be simplified and offered as a service due to the high overheads involved with training staff to specialise in the technology.
"Developers strongly preferred simplicity to fine-grained control," he said. "Dynamo might have been the best technology in the world at the time but it was still software you had to run yourself. And nobody wanted to learn how to do that if they didn’t have to. Ultimately, developers wanted a service."
Like Amazon's SimpleDB and EBS, DynamoDB allows customers to set and then dial up or down required levels of read and write throughput and replicates stored data across Amazon's different availability zones.
Amazon has set a flat, hourly rate for the service based on reserved capacity, with write throughput charged at 1 cent per hour for every 10 units of write capacity and 1 cent an hour for ever 50 units of read capacity. It also includes 100 MB of free storage, as well as five writes per second and 10 reads per second.