Bus members
The members of a service integration bus can be application servers, server clusters, or IBM MQ servers. Bus members that are application servers or server clusters contain messaging engines, which are the application server components that provide asynchronous messaging services. Bus members that are IBM MQ servers provide a direct client connection between a service integration bus and queues on a IBM MQ queue manager.
To use a service integration bus, you must add at least one application server or server cluster as a bus member. 您也可以新增本身是 IBM MQ 伺服器的匯流排成員;服務整合利用這些匯流排成員,在 IBM MQ 佇列中寫入及讀取訊息。
When you add an application server or a server cluster as a bus member, a messaging engine for that bus member is created automatically. If the bus member is an application server, it can have only one messaging engine. 為了管理佇列類型目的地,傳訊引擎包括一個訊息儲存庫,必要之時,它可以將訊息存放在這裡,直到消費端應用程式準備好接收它們,或保留訊息來預防傳訊引擎失敗。 如果匯流排成員是伺服器叢集,它可以有更多傳訊引擎來提供高可用性或工作量分散性質。 如果匯流排成員是 IBM MQ 伺服器,它沒有傳訊引擎,但它可讓您直接從 IBM MQ 佇列管理程式和(針對 WebSphere® MQ for z/OS®)佇列共用群組存取 IBM MQ 佇列。
A WebSphere Application Server application does not have to be running on a service integration bus member to use its messaging services. If necessary, WebSphere Application Server automatically provides a connection to a suitable bus member.

If required, you can remove members from a bus. However, this action deletes any messaging engines that are associated with a bus member, including knowledge of any messages held by the message store for those messaging engines. Therefore, you must plan this action carefully.
When a bus member is deleted, the data source associated with this bus member is not automatically deleted, in case you are also using the data source for some other purpose. This also applies to bus members created using the default data source: the data source is not automatically deleted and you must remove it manually. You need not remove the default data sources because they use a universal unique identifier (UUID) in the name of the Apache Derby database. However, you might choose to delete the data source to avoid wasting disk space.