Sun Java System Message Queue 4.3 Administration Guide

Adjusting Multiple-Consumer Queue Delivery

The efficiency with which multiple queue consumers process messages in a queue destination depends on a number of factors. To achieve optimal message throughput there must be a sufficient number of consumers to keep up with the rate of message production for the queue, and the messages in the queue must be routed and then delivered to the active consumers in such a way as to maximize their rate of consumption.

The message delivery mechanism for multiple-consumer queues is that messages are delivered to consumers in batches as each consumer is ready to receive a new batch. The readiness of a consumer to receive a batch of messages depends upon configurable client runtime properties, such as imqConsumerFlowLimit and imqConsumerFlowThreshold, as described in Message Flow Limits. As new consumers are added to a queue, they are sent a batch of messages to consume, and receive subsequent batches as they become ready.


Note –

The message delivery mechanism for multiple-consumer queues described above can result in messages being consumed in an order different from the order in which they are produced.


If messages are accumulating in the queue, it is possible that there is an insufficient number of consumers to handle the message load. It is also possible that messages are being delivered to consumers in batch sizes that cause messages to be backing up on the consumers. For example, if the batch size (consumerFlowLimit) is too large, one consumer might receive all the messages in a queue while other consumers receive none. If consumers are very fast, this might not be a problem. However, if consumers are relatively slow, you want messages to be distributed to them evenly, and therefore you want the batch size to be small. Although smaller batch sizes require more overhead to deliver messages to consumers, for slow consumers there is generally a net performance gain in using small batch sizes. The value of consumerFlowLimit can be set on a destination as well as on the client runtime: the smaller value overrides the larger one.