[zeromq-dev] inproc seems to be rather slow

Doherty, Kevin Kevin.Doherty at mlp.com
Fri Apr 27 18:52:22 CEST 2012


tx
Oddly though I don't get the same "low number of iterations" issue with  an ipc socket test.
Even the following in the same thread for a PAIR crawls when done one at a time a second apart.
Other transports have no problem.

struct timespec t1;
struct timespec t2;
clock_gettime(CLOCK_REALTIME, &t1);
    // Now it should be possible to send one more
    bool sent = s2.send (msg, 0);
    assert (sent);

    received = s1.recv (&msg, 0);
clock_gettime(CLOCK_REALTIME, &t2);
printf ("CLOCK_REALTIME : %ld\n", t2.tv_nsec-t1.tv_nsec );



From: zeromq-dev-bounces at lists.zeromq.org [mailto:zeromq-dev-bounces at lists.zeromq.org] On Behalf Of Brian Knox
Sent: Friday, April 27, 2012 11:33 AM
To: ZeroMQ development list
Subject: Re: [zeromq-dev] inproc seems to be rather slow

Here are my results:

[laotse at moneta perf]$ ./inproc_lat 1024 100000
message size: 1024 [B]
roundtrip count: 100000
average latency: 13.333 [us]

Try sending more than one message.

I haven't looked at the performance test code but my guess is that it is measuring some of the test setup time in it's measurements.

Brian

On Fri, Apr 27, 2012 at 7:08 AM, Doherty, Kevin <Kevin.Doherty at mlp.com<mailto:Kevin.Doherty at mlp.com>> wrote:
On a fairly decent  Linux box
These performance number don't look so good.
The test program that comes with the package below uses REQ/REP.
Also similar slowness with a PUSH/PULL socket.
$ pwd
zeromq-2.1.11/perf

././inproc_lat 1 1
message size: 1 [B]
roundtrip count: 1
average latency: 151.500 [us]
$ ././inproc_lat 1 1
message size: 1 [B]
roundtrip count: 1
average latency: 116.000 [us]
$ ././inproc_lat 1 1
message size: 1 [B]
roundtrip count: 1
average latency: 84.500 [us]
$ ././inproc_lat 1 1
message size: 1 [B]
roundtrip count: 1
average latency: 53.500 [us]
$ ././inproc_lat 1 1
message size: 1 [B]
roundtrip count: 1
average latency: 49.500 [us]


######################################################################
The information contained in this communication is confidential and
may contain information that is privileged or exempt from disclosure
under applicable law. If you are not a named addressee, please notify
the sender immediately and delete this email from your system.
If you have received this communication, and are not a named
recipient, you are hereby notified that any dissemination,
distribution or copying of this communication is strictly prohibited.
######################################################################
_______________________________________________
zeromq-dev mailing list
zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>
http://lists.zeromq.org/mailman/listinfo/zeromq-dev


######################################################################
The information contained in this communication is confidential and
may contain information that is privileged or exempt from disclosure
under applicable law. If you are not a named addressee, please notify
the sender immediately and delete this email from your system.
If you have received this communication, and are not a named
recipient, you are hereby notified that any dissemination,
distribution or copying of this communication is strictly prohibited.
######################################################################
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.zeromq.org/pipermail/zeromq-dev/attachments/20120427/4d83c2df/attachment.htm>


More information about the zeromq-dev mailing list