[rabbitmq-discuss] Throughput observation with RabbitMQ-3.1.3 and Erlang R16B01 - Single Node and Cluster Node
mklishin at gopivotal.com
Thu Aug 1 20:19:35 BST 2013
> 1) Is this expected behaviour on throughput scaling when number of producer and consumer increases linearly ?
Given that there are hardware resources available, yes, but up to a certain point.
> 2) Can something be done to improve throughout with single node configuration without changing publisher confirm and comsumer ack configuration ( means keeping them enabled) ?
For publishers, no (besides reducing message payload). For consumers, increase prefetch count
x2-8 and add more consumer threads (probably not going to have any effect on Pika and CPython).
> 3) How to calculate server latency in approximate way ? here I thought by adding round trip time (RTT) for both publisher confirm and consumer ack, one can get latency. Is this correct understanding ? What is the effective method to calculate RTT ?
Calculating RTT for a single message correctly is quite tricky. I suggest that you calculate avg latency
for the entire run and take a mean of, say, 10 runs (or 100, if it can be automated).
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 495 bytes
Desc: Message signed with OpenPGP using GPGMail
More information about the rabbitmq-discuss