Repository navigation
Terminology
Jim Carroll edited this page Jul 24, 2019
·
3 revisions
#Terminology
Having gone through the example we should codify some of the terminology that was developed and will continue to be used throughout this document as well as throughout the Dempsy code.
| Term | Definition |
|---|---|
| message processor | an instance of a cloned message processor prototype responsible for processing every message of a particular type with a particular unique key. |
| message processor prototype | an instance used by Dempsy to serve as a template when it needs to create more instances of a message processor. |
| message | is an object that Dempsy routes to a message processor based on the message's key. |
| key | obtained from a message using the method on the message object that's annotated with the `@MessageKey` annotation. Each unique key addresses an individual message processor instance in a cluster |
| cluster | a cluster is the collection of all message processors or adaptors of a common type in the same stage of processing of a Dempsy application. A cluster contains a complete set of potential message processors keyed by all of the potential keys from a particular message type.That is, a cluster of message processor instances covers the entire key-space of a message. |
| node | a node is a subset of a set of clusters containing a portion of the each cluster's message processors. nodes are almost always (except in some possible test situations) the intersection of a cluster and a Java process. That is, the portion of a cluster that's running in a particular process is the cluster's node |
| container | Sometimes also referred to as a message processor container, it is the part of the Dempsy infrastructure that manages the lifecycle of message processors within an individual node. That is, there is a one-to-one between the portion of a cluster running in a particular node, and a container. |
Next section: The Message Processor Lifecycle