Amazon Web Services launched a managed NoSQL database service yesterday that lets users easily launch a database and scale it up or down as needed.
The service meets the needs of web companies that are collecting, storing and processing an increasing amount of data. Without such a scalable database, AWS users would sometimes spend weeks forecasting and preparing their databases to perform during heavy usage periods, the company said. That's because traditional databases were not designed to scale quickly.
"Managing and scaling databases has always been the Achilles heal of web apps," said Werner Vogels, CTO of Amazon. To scale, companies could either buy bigger hardware or split databases across servers, he said. "Both approaches are increasingly complicated and expensive," he said. "Plus, there's a real shortage of technical people who have the specialised skills to do this."
DynamoDB, the new database service, stores data on Solid State Drives and replicates across regions in order to ensure access for customers.
The company designed the service based on its experience building large non-relational databases for Amazon.com. DynamoDB is being used internally at Amazon for the Amazon.com advertising platform, Amazon Cloud Drive, IMDb and Kindle.
In addition, companies including Elsevier, SmugMug and Formspring are already using it.
Others can get started using it now beginning with a free service that provides 100MB of storage, five writes and 10 reads per second up to 40 million requests per month.