The two internet search giants, Google and Yahoo!, have partially based their success on the way they manage their database over clusters of servers. But this is not something every company can accomplish, mainly because of the very high cost of managing the network of servers. This is where Aster Data System comes in.

The company is offering a clustered database for analytics for companies that are rich enough to afford it, but poor enough to do it themselves. This cluster-based approach means that the data is spread over a network of processing units, thus speeding up the computing process. The unique idea behind Aster’s system is that the nodes are divided into three layers: loader nodes, worker nodes and queen nodes. The loader nodes must take the data and load it or export it from or to external locations. The worker nodes store the data on disks, while the queen nodes control the whole process and say which data where should be stored. This way, the system makes sure that data will be located where it is needed, or at least somewhere in the vicinity.

The company already has a big client: MySpace. The social network site has hired Aster to manage its terabytes large database. Aster Data System was founded in 2005 by a group of PhD students at Stanford, and it took three years to get to the market. During this time, the company received founding from former Google early investors Sequoia Capital and computer science professor David Cheriton, among others.

The price a company will pay for the software will depend on the size of the database it will be used for, and will start at around $100,000.