diff options
Diffstat (limited to 'doc/book/src/How-to-Tune-M3-Java-Broker-Performance.xml')
| -rw-r--r-- | doc/book/src/How-to-Tune-M3-Java-Broker-Performance.xml | 151 |
1 files changed, 151 insertions, 0 deletions
diff --git a/doc/book/src/How-to-Tune-M3-Java-Broker-Performance.xml b/doc/book/src/How-to-Tune-M3-Java-Broker-Performance.xml new file mode 100644 index 0000000000..109e283a22 --- /dev/null +++ b/doc/book/src/How-to-Tune-M3-Java-Broker-Performance.xml @@ -0,0 +1,151 @@ +<?xml version="1.0" encoding="utf-8"?> +<chapter xmlns:html="http://www.w3.org/1999/xhtml"> + <title> + Apache Qpid : How to Tune M3 Java Broker Performance + </title> + <section role="h3" id="HowtoTuneM3JavaBrokerPerformance-ProblemStatement"> + <title> + Problem + Statement + </title> + <para> + During destructive testing of the Qpid M3 Java Broker, we tested + some tuning techniques and deployment changes to improve the Qpid + M3 Java Broker's capacity to maintain high levels of throughput, + particularly in the case of a slower consumer than produceer + (i.e. a growing backlog). + </para> + <para> + The focus of this page is to detail the results of tuning & + deployment changes trialled. + </para> + <para> + The successful tuning changes are applicable for any deployment + expecting to see bursts of high volume throughput (1000s of + persistent messages in large batches). Any user wishing to use + these options <emphasis>must test them thoroughly in their own + environment with representative volumes</emphasis>. + </para> + <!--h3--> + </section> + + <section role="h3" id="HowtoTuneM3JavaBrokerPerformance-SuccessfulTuningOptions"> + <title> + Successful + Tuning Options + </title> + <para> + The key scenario being taregetted by these changes is a broker + under heavy load (processing a large batch of persistent + messages)can be seen to perform slowly when filling up with an + influx of high volume transient messages which are queued behind + the persistent backlog. However, the changes suggested will be + equally applicable to general heavy load scenarios. + </para> + <para> + The easiest way to address this is to separate streams of + messages. Thus allowing the separate streams of messages to be + processed, and preventing a backlog behind a particular slow + consumer. + </para> + <para> + These strategies have been successfully tested to mitigate this + problem: + </para> + <table> + <title/> + <tgroup cols="2"> + <tbody> + <row> + <entry> + Strategy + </entry> + <entry> + Result + </entry> + </row> + <row> + <entry> + Seperate connections to one broker for separate streams of + messages. + </entry> + <entry> + Messages processed successfully, no problems experienced + </entry> + </row> + <row> + <entry> + Seperate brokers for transient and persistent messages. + </entry> + <entry> + Messages processed successfully, no problems experienced + </entry> + </row> + </tbody> + </tgroup> + </table> + <para> + <emphasis>Separate Connections</emphasis> + Using separate connections effectively means that the two streams + of data are not being processed via the same buffer, and thus the + broker gets & processes the transient messages while + processing the persistent messages. Thus any build up of + unprocessed data is minimal and transitory. + </para> + <para> + <emphasis>Separate Brokers</emphasis> + Using separate brokers may mean more work in terms of client + connection details being changed, and from an operational + perspective. However, it is certainly the most clear cut way of + isolating the two streams of messages and the heaps impacted. + </para> + <section role="h4" id="HowtoTuneM3JavaBrokerPerformance-Additionaltuning"> + <title> + Additional + tuning + </title> + <para> + It is worth testing if changing the size of the Qpid read/write + thread pool improves performance (eg. by setting + JAVA_OPTS="-Damqj.read_write_pool_size=32" before running + qpid-server). By default this is equal to the number of CPU + cores, but a higher number may show better performance with some + work loads. + </para> + <para> + It is also important to note that you should give the Qpid broker + plenty of memory - for any serious application at least a -Xmx of + 3Gb. If you are deploying on a 64 bit platform, a larger heap is + definitely worth testing with. We will be testing tuning options + around a larger heap shortly. + </para> + <!--h4--> + </section> + <!--h3--> + </section> + + <section role="h3" id="HowtoTuneM3JavaBrokerPerformance-NextSteps"> + <title> + Next + Steps + </title> + <para> + These two options have been testing using a Qpid test case, and + demonstrated that for a test case with a profile of persistent + heavy load following by constant transient high load traffic they + provide significant improvment. + </para> + <para> + However, the deploying project <emphasis>must</emphasis> complete their own + testing, using the same destructive test cases, representative + message paradigms & volumes, in order to verify the proposed + mitigation options. + </para> + <para> + The using programme should then choose the option most applicable + for their deployment and perform BAU testing before any + implementation into a production or pilot environment. + </para> + <!--h3--> + </section> +</chapter> |