The number of services generating billing events being added to system has implications for the amount of usage event records being generated for usage rating and mediation.
The system needs to employ a combination of scaling mechanisms to handle throughput. For deployments generating smaller sized Usage Data Records (UDRs) with similar characteristics, it is possible to aggregate the records into larger size event records prior to mediation and then pricing.
Care must be taken during the aggregation process to ensure granular attributes specific to each individual event record are not lost during the aggregation process.
Usage Data Record aggregation is not always optimal in all circumstances but can be of value in scenarios with large subscription and data volumes but small event record sizes. Once the UDR arrives at the input of the rating and mediation logic, horizontal scaling techniques are employed to further handle on-demand processing.