Configuring Insight Server
You must create and configure an Insight Server or servers so that the runtime environment receives and processes events correctly. If you plan to create a multiple server topology that is continuously available, determine how many computers and servers you require for your topology, then follow the instructions on configuring for production.
Insight Server receives events from
multiple sources such as business systems, social networks, mobile devices, and sensors. These
events are processed as they happen and business rules act on the events and data that is stored in
memory to derive actionable insights. To accomplish this type of processing in an efficient way,
Insight Server requires a run time that can
handle large numbers of events, in the order of tens of thousands per second, and that can scale in
a linear way. You must configure the runtime environment to meet your non-functional requirements,
including high availability, security, and recoverability. There is no one-size-fits-all
topology. Different solutions have different levels of non-functional requirements and risk
tolerance.
When a solution includes connectivity, one or more servers are configured as inbound connectivity servers to act as a bridge between external messaging endpoints and the solution gateway. Outbound events that target external systems are routed to one or more outbound connectivity servers. A multi-server topology is shaped by the inbound and outbound connectivity servers, catalog servers, runtime servers, and a backing database that is used for persistence of grid data to disk storage.