2008-04-21

Self-organizing cluster

Imagine a sparkling white box which serve your data with dazzling speed and scales out automatically if you just add another unit by plugging in the cord. No fuzz with configurations, IP-addresses or tuning. Imagine.

You would need to supply some simple rules about your data model for the system to be able to know what you need: A few data sources define the entry point. They get associated with a network interface (not tied to a specific appliance mind!) and then you click on the scale-out pattern which you want to use each time a new unit is added to the hive. Simple. Can it work? Absolutely! :)

0 kommentarer:

Arkivet