[rabbitmq-discuss] another performance question

Wed Jul 17 22:14:09 BST 2013

Matthias,

OK, built erlang R16B from source, then rabbitmq 3.1.3 from source as well. Still no joy.

sh runjava.sh com.rabbitmq.examples.MulticastMain -y 0 -t topic -s 10000
starting producer #0
time: 1.000s, sent: 5092 msg/s
time: 2.000s, sent: 11501 msg/s
time: 3.000s, sent: 14754 msg/s
time: 4.001s, sent: 14339 msg/s
time: 5.001s, sent: 15070 msg/s
time: 6.001s, sent: 15238 msg/s
time: 7.001s, sent: 15329 msg/s
time: 8.001s, sent: 15335 msg/s
time: 9.001s, sent: 15329 msg/s
time: 10.001s, sent: 15237 msg/s
time: 11.001s, sent: 15168 msg/s
time: 12.001s, sent: 15160 msg/s
time: 13.001s, sent: 15220 msg/s
time: 14.001s, sent: 15311 msg/s
time: 15.001s, sent: 15180 msg/s
time: 16.001s, sent: 15241 msg/s
time: 17.001s, sent: 15194 msg/s
time: 18.001s, sent: 15064 msg/s
time: 19.001s, sent: 15193 msg/s
time: 20.001s, sent: 15163 msg/s

Unix top output:

top - 15:09:38 up 23 days,  3:13,  4 users,  load average: 1.32, 1.18, 0.86
Tasks: 151 total,   2 running, 149 sleeping,   0 stopped,   0 zombie
%Cpu(s): 23.3 us,  9.8 sy,  0.0 ni, 64.8 id,  0.0 wa,  0.0 hi,  2.1 si,  0.0 st
KiB Mem:  33019144 total,  2277464 used, 30741680 free,   136004 buffers
KiB Swap: 47313916 total,        0 used, 47313916 free,  1586932 cached

  PID USER      PR  NI  VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND
30714 root      20   0  256m  54m 2756 S 181.5  0.2   6:23.19 beam.smp
30990 geib      20   0 8476m 151m  10m S 100.7  0.5   0:46.45 java
31107 geib      20   0 23172 1540 1120 R   1.0  0.0   0:00.05 top
31016 geib      20   0 77392 1840 1000 S   0.3  0.0   0:00.02 sshd
    1 root      20   0 10648  808  672 S   0.0  0.0   0:19.60 init
    2 root      20   0     0    0    0 S   0.0  0.0   0:00.11 kthreadd
    3 root      20   0     0    0    0 S   0.0  0.0   0:04.25 ksoftirqd/0

Mark Geib
Principal Engineer
Cheyenne Software Engineering
mark.geib at echostar.com / 35-215

“We, the unwilling, led by the unknowing, are doing the impossible for the ungrateful. We have done so much, for so long, with so little, we are now qualified to do anything with nothing.” 
― Mother Teresa

On Jul 17, 2013, at 2:05 PM, Matthias Radestock wrote:

> On 17/07/13 21:00, Geib, Mark wrote:
>> No, my initial reply to you was output on a bare metal. The VM
>> results were just to confirm our performance numbers.
>> 
>> My main testing machine is bare metal.
> 
> I have no idea then what could be the cause. Try running with an R16B compiled from source, since that's what I'm running. I doubt it will make a difference, but I can't think of anything else to try.
> 
> Matthias.
> 