So far, rich Internet applications (RIA) don’t have to be concerned with data management on the client side, as most connected implementations are designed to be simply a visualization and interaction layer to data and services on the server side. In these cases, data is retrieved from the server as JSON or XML data types (or serialized binary objects for some platforms), then cached in-memory as objects and collections, part of the application, and eventually discarded as part of the application lifecycle.

But for relatively more complex applications that require larger and more distributed data sets, lower application latency, reduced chattiness, data relationships (even basic ones), the implementation starts to get a little complicated, and becomes more difficult with procedural languages like JavaScript. In that case, a simple and light-weight in-memory embedded database may become quite useful.

Read the full post on David Chou’s blog at