DynamoDB is a NoSQL database service that stores data on solid-state disk drives and creates replicas in three different locations to improve availability. AWS improves the performance of the DynamoDB service with parallel scan, which gives users faster access to their tables.
So, an application that processes a large table of historical data can perform a parallel scan. This moves much faster than a sequential one, and thus the result sets return faster to the application. Think, a group of people leaving a theater using one door, versus twenty. Leaving through twenty doors is much faster.
The many database-as-a-service offerings out there all want to differentiate themselves on features and function. Performance is something that is typically on the critical list for those who pick databases that live in the cloud, such as DynamoDB.
While this does not seem like much in the way of new for AWS, the reality is that the cloud-based database wars have been on for some time. As improvements are made to cloud-based database offerings, the providers will be quick to point them out.