MatrixHB opened a new issue, #4114:
URL: https://github.com/apache/rocketmq/issues/4114
We use the API `queryConsumeTimeSpan` provided by DefaultMQAdminExt when
querying message within a certain time period on a user console. We found that
the API `queryConsumeTimeSpan` takes a long time, which cause querying message
timeout all the time.
The reason is that our broker cluster is relatively large, including over
100 brokers. In the implement of `queryConsumeTimeSpan`, the client need to
request all the brokers to fetch consumeTimeSpan information, and the
request-reply are serial.
Original Code:
```
TopicRouteData topicRouteData = this.examineTopicRouteInfo(topic);
for (BrokerData bd : topicRouteData.getBrokerDatas()) {
String addr = bd.selectBrokerAddr();
if (addr != null) {
spanSet.addAll(this.mqClientInstance.getMQClientAPIImpl().queryConsumeTimeSpan(addr,
topic, group, timeoutMillis));
}
}
```
In order to improve the performce, we try to add thread pool with 10 core
threads in `DefaultMQAdminExtImpl`, and execute queryConsumeTimeSpan
concurrently using CompletableFuture. This thread pool can be shared by other
APIs in DefaultMQAdminExt.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
