The conditional trace service is hosted in an Akka cluster which runs on one or several SC processes. For more information, see 1.7 Akka Cluster.
When Conditional Trace is running on multiple SCs, the service is distributed across all of the nodes. This allows you to scale up and down, and optimize resource utilization and footprint. In addition, by running on multiple processes, if a node experiences downtime, the service can continue to operate as long as some part of the cluster is up and running.
Any ECs running a real-time workflow starts up an Akka Cluster client which connects to the Akka Cluster via remoting. For further information, see https://doc.akka.io/docs/akka/2.4.12/scala/remoting.html.
When trace information is sent from a running workflow instance to the conditional trace service, it is stored on one of the nodes in the cluster. This information is stored in-memory so that if a node goes down, that information might be lost. The filter status information is replicated in-memory across all of the SCs, and is only lost if the whole cluster is restarted. For information on how to manage a cluster, see 1.7 Akka Cluster. The Conditional Trace templates that you create and save are persisted on the configuration server.
Conditional Trace architecture