OpenX has support for the Distributed Statistics. Distributed Statistics Architecture can be used for the really large setups where single core database is not enough. In Distributed Statistics architecture there will be a number of web servers. Each web servers runs a slave database (partial replication of a core database) and logs its own statistics to the local database, and perform a partial statistics summary locally. These all data from the all slave databases are pushed to the core database, where the remaining statistics generation that requires a complete view of the raw data can be performed. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

The basic architecture is shown above. As you can see the architecture is based around a series of delivery servers reporting back to a central database.

Performance
The performance of each of the delivery servers is almost exactly the same as a stand alone ad server.

Scalability
The server farm can be scaled to greater ad serving volumes by adding more delivery servers. This assumes that the central database can handle the extra reporting load which should be taken into account when choosing the central database hardware.

Redundancy
The architecture provides good redundancy as a failure of any single server will not disable the system. The load balancer will handle a delivery server outage by distributing the extra load across the other delivery servers, and the delivery servers will continue to serve ads in the event that the central database cannot be contacted.

Conclusion
This new functionality takes the headache out of coordinating the delivery of ads across several servers. It’s a proven system used by OpenX publishers serving billions of ads each month.

With this basic architecture you will be able to scale to serve billions of impressions each month but you can scale even further. Distributed Statistics architecture will both eliminate most single points of failure, and allow almost unlimited scalability.