You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by mickhayes <mi...@gmail.com> on 2012/04/23 11:42:58 UTC
Re: Message rate comparison with RabbitMQ
You can look at the specJMS2007 published results here:
http://www.spec.org/jms2007/results/jms2007.html
but the results aren't normalized so they cannot be directly compared.
Note for example the HW used by RHEL, and compare with that used by the
other test runs.
At the risk of stating the obvious:
It always costs more to handle more messages.
It always costs more to handle larger messages.
It always costs more to persist messages to disk.
Faster HW/SW does the same things faster, or more things in the same time.
-----
Michael Hayes B.Sc. (NUI), M.Sc. (DCU), SCSA SCNA
--
View this message in context: http://activemq.2283324.n4.nabble.com/Message-rate-comparison-with-RabbitMQ-tp4575958p4579935.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: Message rate comparison with RabbitMQ
Posted by mickhayes <mi...@gmail.com>.
That's SPECjms2007 of course, not specJMS2007...
-----
Michael Hayes B.Sc. (NUI), M.Sc. (DCU), SCSA SCNA
--
View this message in context: http://activemq.2283324.n4.nabble.com/Message-rate-comparison-with-RabbitMQ-tp4575958p4579938.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: Message rate comparison with RabbitMQ
Posted by mickhayes <mi...@gmail.com>.
It's a throughput performance value for either the vertical or the horizontal
topology.
http://www.spec.org/jms2007/docs/DesignDocument.html gives more details.
http://www.spec.org/jms2007/docs/DesignDocument.html#S22 gives some pictures
of the JMS traffic.
BASE is the variable used along with some constants for each benchmark
interaction.
SUT is the System Under Test.
I /think/ that a value of X for e.g. SPECjms2007@Horizontal means that "a
BASE of X was the highest value we could run the benchmark *successfully
*for the SUT."
I.e. we tried with X+1 but it failed.
-----
Michael Hayes B.Sc. (NUI), M.Sc. (DCU), SCSA SCNA
--
View this message in context: http://activemq.2283324.n4.nabble.com/Message-rate-comparison-with-RabbitMQ-tp4575958p4580500.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.