You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by Dave Fisher <wa...@apache.org> on 2022/10/19 20:08:28 UTC
OpenMessaging Benchmark - Artemis Driver PR
There is a project at the Linux Foundation called https://openmessaging.cloud - they include Artemis drivers in their Benchmark program - for both Artemis here: https://github.com/openmessaging/benchmark/tree/master/driver-artemis and Artemis JMS here: https://github.com/openmessaging/benchmark/tree/master/driver-jms
We had someone doing benchmarks using the Artemis JMS driver and on a cross country test with about 60 ms of internet latency the results are not good.
We wanted to get feedback from the ActiveMQ project about a PR we are submitting: https://github.com/openmessaging/benchmark/pull/343
Are there additional changes that should be made to optimize Artemis?
Also if there are any JMS configurations that could improve throughput that can be made when running Artemis JMS any pointers would be helpful.
Best Regards,
Dave
Re: OpenMessaging Benchmark - Artemis Driver PR
Posted by Clebert Suconic <cl...@gmail.com>.
Franz worked with the software mills guys. There is a blog post about it I
think with some recommendations.
I will see if I can find some links in the morning to add here.
On Thu, Oct 20, 2022 at 2:04 AM Clebert Suconic <cl...@gmail.com>
wrote:
> The confirmation window size is definitely a change that needs to be done.
>
> Is your benchmark a single producer / consumer ?
>
>
> There is a lot of things we do to allow scale of multiple clients.
>
> On Wed, Oct 19, 2022 at 4:08 PM Dave Fisher <wa...@apache.org> wrote:
>
>> There is a project at the Linux Foundation called
>> https://openmessaging.cloud - they include Artemis drivers in their
>> Benchmark program - for both Artemis here:
>> https://github.com/openmessaging/benchmark/tree/master/driver-artemis
>> and Artemis JMS here:
>> https://github.com/openmessaging/benchmark/tree/master/driver-jms
>>
>> We had someone doing benchmarks using the Artemis JMS driver and on a
>> cross country test with about 60 ms of internet latency the results are not
>> good.
>>
>> We wanted to get feedback from the ActiveMQ project about a PR we are
>> submitting: https://github.com/openmessaging/benchmark/pull/343
>>
>> Are there additional changes that should be made to optimize Artemis?
>>
>> Also if there are any JMS configurations that could improve throughput
>> that can be made when running Artemis JMS any pointers would be helpful.
>>
>> Best Regards,
>> Dave
>>
>>
>> --
> Clebert Suconic
>
--
Clebert Suconic
Re: OpenMessaging Benchmark - Artemis Driver PR
Posted by Clebert Suconic <cl...@gmail.com>.
I don’t have much time myself to dedicate to the benchmark now. (For the
next month at least. )
If you have any issues after that Pr I would appreciate if you could
provide some feedback here please. I would make some time or ask someone
else here who could do jt.
Thank you.
On Thu, Oct 20, 2022 at 12:50 PM Dave Fisher <wa...@apache.org> wrote:
>
>
> > On Oct 19, 2022, at 11:04 PM, Clebert Suconic <cl...@gmail.com>
> wrote:
> >
> > The confirmation window size is definitely a change that needs to be
> done.
> >
> > Is your benchmark a single producer / consumer ?
>
> The benchmark program can be configured with many consumers and producers
> of many topics.
>
> You can see some of the extreme tests that some are doing from the
> workloads here:
> https://github.com/openmessaging/benchmark/tree/master/workloads
>
> Thanks,
> Dave
>
> >
> >
> > There is a lot of things we do to allow scale of multiple clients.
> >
> > On Wed, Oct 19, 2022 at 4:08 PM Dave Fisher <wa...@apache.org> wrote:
> >
> >> There is a project at the Linux Foundation called
> >> https://openmessaging.cloud - they include Artemis drivers in their
> >> Benchmark program - for both Artemis here:
> >> https://github.com/openmessaging/benchmark/tree/master/driver-artemis
> and
> >> Artemis JMS here:
> >> https://github.com/openmessaging/benchmark/tree/master/driver-jms
> >>
> >> We had someone doing benchmarks using the Artemis JMS driver and on a
> >> cross country test with about 60 ms of internet latency the results are
> not
> >> good.
> >>
> >> We wanted to get feedback from the ActiveMQ project about a PR we are
> >> submitting: https://github.com/openmessaging/benchmark/pull/343
> >>
> >> Are there additional changes that should be made to optimize Artemis?
> >>
> >> Also if there are any JMS configurations that could improve throughput
> >> that can be made when running Artemis JMS any pointers would be helpful.
> >>
> >> Best Regards,
> >> Dave
> >>
> >>
> >> --
> > Clebert Suconic
>
> --
Clebert Suconic
Re: OpenMessaging Benchmark - Artemis Driver PR
Posted by Dave Fisher <wa...@apache.org>.
> On Oct 19, 2022, at 11:04 PM, Clebert Suconic <cl...@gmail.com> wrote:
>
> The confirmation window size is definitely a change that needs to be done.
>
> Is your benchmark a single producer / consumer ?
The benchmark program can be configured with many consumers and producers of many topics.
You can see some of the extreme tests that some are doing from the workloads here: https://github.com/openmessaging/benchmark/tree/master/workloads
Thanks,
Dave
>
>
> There is a lot of things we do to allow scale of multiple clients.
>
> On Wed, Oct 19, 2022 at 4:08 PM Dave Fisher <wa...@apache.org> wrote:
>
>> There is a project at the Linux Foundation called
>> https://openmessaging.cloud - they include Artemis drivers in their
>> Benchmark program - for both Artemis here:
>> https://github.com/openmessaging/benchmark/tree/master/driver-artemis and
>> Artemis JMS here:
>> https://github.com/openmessaging/benchmark/tree/master/driver-jms
>>
>> We had someone doing benchmarks using the Artemis JMS driver and on a
>> cross country test with about 60 ms of internet latency the results are not
>> good.
>>
>> We wanted to get feedback from the ActiveMQ project about a PR we are
>> submitting: https://github.com/openmessaging/benchmark/pull/343
>>
>> Are there additional changes that should be made to optimize Artemis?
>>
>> Also if there are any JMS configurations that could improve throughput
>> that can be made when running Artemis JMS any pointers would be helpful.
>>
>> Best Regards,
>> Dave
>>
>>
>> --
> Clebert Suconic
Re: OpenMessaging Benchmark - Artemis Driver PR
Posted by Clebert Suconic <cl...@gmail.com>.
The confirmation window size is definitely a change that needs to be done.
Is your benchmark a single producer / consumer ?
There is a lot of things we do to allow scale of multiple clients.
On Wed, Oct 19, 2022 at 4:08 PM Dave Fisher <wa...@apache.org> wrote:
> There is a project at the Linux Foundation called
> https://openmessaging.cloud - they include Artemis drivers in their
> Benchmark program - for both Artemis here:
> https://github.com/openmessaging/benchmark/tree/master/driver-artemis and
> Artemis JMS here:
> https://github.com/openmessaging/benchmark/tree/master/driver-jms
>
> We had someone doing benchmarks using the Artemis JMS driver and on a
> cross country test with about 60 ms of internet latency the results are not
> good.
>
> We wanted to get feedback from the ActiveMQ project about a PR we are
> submitting: https://github.com/openmessaging/benchmark/pull/343
>
> Are there additional changes that should be made to optimize Artemis?
>
> Also if there are any JMS configurations that could improve throughput
> that can be made when running Artemis JMS any pointers would be helpful.
>
> Best Regards,
> Dave
>
>
> --
Clebert Suconic