Advanced scale-out

Congress needs to support scale-out to support high throughput for simulate and table query API calls. This blueprint describes a design that adds arbitrary scale-out to Congress by replicating the Congress policy engine to multiple machines. Each policy engine machine will hold a complete copy of the data source data, and will be able to satisfy simulate and table query API calls independently. The datasource driver functionality will be replicated using a master-standby server pair, and only the master will actively receive data from the datasources.