<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Hi, <br>
<br>
Just some additional informations on that topic:<br>
<br>
Regarding the performance aspect Christian didn't mention some
points that might be important.<br>
We also tested connecting the robot to one of the Laptops and just
connecting _one_ of the cameras<br>
to its gigabit ethernet port. If you then start the camera and the
robot task within syskit the delays<br>
also occur, and they not just stuck for several milliseconds but up
to 8 seconds which causes the<br>
driver delivering a time out on the connection to the robot (via
serial port) and shuts down the connection<br>
completely. I think that it is not "just" a performance issue, the
Laptop has quite high performance hardware.<br>
<br>
Also on the SeekurJr itself the problem occurs already with one
camera running. While the camera is running<br>
with 10Fps and binning, the logger has to write ~1MB per Frame which
is just 10 MB/s. I did some benchmarking<br>
on the harddrive (without system cache) which results in a write
performance between 50 - 60 MB/s constantly.<br>
<br>
Btw. the aria task and the camera task are not in the same
deployment. <br>
<br>
So it still could be some performance issue, maybe in combination
with unfortunate configured tasks, especially<br>
I am not sure about the camera task, as far as I understood the
answer from Alex Duda, I guess such a time out<br>
is already used, but maybe Christian can check that better than me.<br>
<br>
Regards<br>
Elmar<br>
<br>
<div class="moz-cite-prefix">Am 17.12.2013 08:36, schrieb Matthias
Goldhoorn:<br>
</div>
<blockquote cite="mid:52AFFF10.4070309@dfki.de" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<div class="moz-cite-prefix">On 16.12.2013 19:28, Christian Rauch
wrote:<br>
</div>
<blockquote cite="mid:52AF4634.1070205@dfki.de" type="cite">I set
the priority by: <br>
<br>
task("seekur_drv","aria::Task").priority(99) <br>
<br>
which does not affect the result (still delays). <br>
<br>
The aria::Task is periodic (100ms) and the camera_ids::Task is
triggered in the updateHook after a new image arrives. <br>
</blockquote>
I assume both in one deployment. You could (for testing reasons)
try to separate the deployments.<br>
<blockquote cite="mid:52AF4634.1070205@dfki.de" type="cite"> <br>
As the camera Logger is writing with approx. 10 MB/s (and CPU
load is below 20%) which is far away from the HDD limit
(~60MB/s) and this whole issue does not occur when running the
cameras with logger in syskit and the aria::Task outside syskit,
I assume that this is not a hardware related issue. <br>
</blockquote>
did you take a look to the load AND the system CPU Usage (top
lines from `top`)<br>
<blockquote cite="mid:52AF4634.1070205@dfki.de" type="cite"> <br>
As this problem for now seems to occur only when running both
tasks (aria, camera_ids) within the bundle; what are the options
to change the scheduling by syskit? Does syskit change the
behaviour of how processes are scheduled by the OS in any way? <br>
</blockquote>
Syskit does NOT influence the scheduling of tasks, this is defined
at deployment time implicitly. The only differences when using
syskit are:<br>
- Syskit takes also CPU Load<br>
- Syskit might use different connection types than you runscript
-> take a look to `syskit instanciate ...` the connection
policys are part of the graph<br>
<br>
I still assume some load-related problems on your machine, sinc'e
you mention that if you stopping the logger everything works fine.<br>
<blockquote cite="mid:52AF4634.1070205@dfki.de" type="cite"> <br>
Regards, <br>
Christian <br>
<br>
<br>
Am 16.12.2013 16:21, schrieb Matthias Goldhoorn: <br>
<blockquote type="cite">Several things could happen: <br>
<br>
The logger consumes too much time to get out the images,
caused by a <br>
slow hdd. <br>
Try to increase the priority level of the pioneer task within
your <br>
deployment to make this more urgent from the point of
scheduling. <br>
<br>
Matthias <br>
<br>
P.S. i assume you task is periodic <br>
<br>
On 16.12.2013 16:17, Christian Rauch wrote: <br>
<blockquote type="cite">Hi, <br>
<br>
I am using the aria task within syskit as a device which is
supposed to <br>
be called every 100ms. The aria task uses an additional
thread <br>
provided by the aria driver to communicate with the robot. <br>
<br>
If I run that task from the syskit shell without additional
sensors the <br>
updateHook gets called as expected. If I add one or more
cameras <br>
(camera_ids) with loggers it happens that the updateHook
gets stuck <br>
for some time: <br>
<br>
{{{ <br>
[DEBUG] - aria::updateHook at: 1386784751.580.629 <br>
[DEBUG] - aria::updateHook at: 1386784751.680.668 <br>
[DEBUG] - aria::updateHook at: 1386784751.780.542 <br>
ArRobotPacketReceiver::receivePacket: bad packet, bad
checksum <br>
Warning: Task 'Packet Handler' took 1213 ms to run (longer
than the 250 <br>
warning time) <br>
Warning: ArRobot sync tasks too long at 1213 ms, (100 ms
normal 250 ms <br>
warning) <br>
[DEBUG] - aria::updateHook at: 1386784753.673.374 <br>
[DEBUG] - aria::updateHook at: 1386784753.677.372 <br>
[DEBUG] - aria::updateHook at: 1386784753.681.379 <br>
[DEBUG] - aria::updateHook at: 1386784753.685.211 <br>
[DEBUG] - aria::updateHook at: 1386784753.689.226 <br>
[DEBUG] - aria::updateHook at: 1386784753.697.023 <br>
[DEBUG] - aria::updateHook at: 1386784753.701.405 <br>
[DEBUG] - aria::updateHook at: 1386784753.705.386 <br>
[DEBUG] - aria::updateHook at: 1386784753.709.383 <br>
[DEBUG] - aria::updateHook at: 1386784753.713.216 <br>
[DEBUG] - aria::updateHook at: 1386784753.717.292 <br>
[DEBUG] - aria::updateHook at: 1386784753.721.253 <br>
[DEBUG] - aria::updateHook at: 1386784753.725.392 <br>
[DEBUG] - aria::updateHook at: 1386784753.729.400 <br>
[DEBUG] - aria::updateHook at: 1386784753.733.392 <br>
ArRobotPacketReceiver::receivePacket: bad packet, bad
checksum <br>
[DEBUG] - aria::updateHook at: 1386784753.737.227 <br>
[DEBUG] - aria::updateHook at: 1386784753.741.226 <br>
[DEBUG] - aria::updateHook at: 1386784753.745.274 <br>
[DEBUG] - aria::updateHook at: 1386784753.749.305 <br>
[DEBUG] - aria::updateHook at: 1386784753.781.249 <br>
[DEBUG] - aria::updateHook at: 1386784753.881.453 <br>
[DEBUG] - aria::updateHook at: 1386784753.981.492 <br>
[DEBUG] - aria::updateHook at: 1386784754.081.523 <br>
[DEBUG] - aria::updateHook at: 1386784754.181.392 <br>
[DEBUG] - aria::updateHook at: 1386784754.281.428 <br>
[DEBUG] - aria::updateHook at: 1386784754.381.462 <br>
}}} <br>
<br>
<br>
For some reason this problem does no longer occur as soon as
the camera <br>
loggers are stopped. It also does not occur when the aria
Task is <br>
started outside of syskit by using the default deployment
and the <br>
TaskInspector or the orocos.rb API. <br>
<br>
It looks like the whole task is blocked somehow (the
updateHook is not <br>
really taking that much processing time). <br>
Any ideas where this might come from and why this only
occurs when <br>
running both tasks with loggers in syskit? <br>
<br>
Regards, <br>
Christian <br>
<br>
<br>
<br>
<br>
_______________________________________________ <br>
Rock-dev mailing list <br>
<a moz-do-not-send="true" class="moz-txt-link-abbreviated"
href="mailto:Rock-dev@dfki.de">Rock-dev@dfki.de</a> <br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev">http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev</a>
<br>
</blockquote>
<br>
<br>
-- <br>
Dipl.-Inf. Matthias Goldhoorn <br>
Space and Underwater Robotic <br>
<br>
Universität Bremen <br>
FB 3 - Mathematik und Informatik <br>
AG Robotik <br>
Robert-Hooke-Straße 1 <br>
28359 Bremen, Germany <br>
<br>
Zentrale: +49 421 178 45-6611 <br>
<br>
Besuchsadresse der Nebengeschäftstelle: <br>
Robert-Hooke-Straße 5 <br>
28359 Bremen, Germany <br>
<br>
Tel.: +49 421 178 45-4193 <br>
Empfang: +49 421 178 45-6600 <br>
Fax: +49 421 178 45-4150 <br>
<a moz-do-not-send="true" class="moz-txt-link-abbreviated"
href="mailto:E-Mail:matthias.goldhoorn@informatik.uni-bremen.de">E-Mail:matthias.goldhoorn@informatik.uni-bremen.de</a>
<br>
<br>
Weitere Informationen:<a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="http://www.informatik.uni-bremen.de/robotik">http://www.informatik.uni-bremen.de/robotik</a>
<br>
<br>
<br>
<br>
_______________________________________________ <br>
Rock-dev mailing list <br>
<a moz-do-not-send="true" class="moz-txt-link-abbreviated"
href="mailto:Rock-dev@dfki.de">Rock-dev@dfki.de</a> <br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev">http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev</a>
<br>
<br>
</blockquote>
<br>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Rock-dev mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:Rock-dev@dfki.de">Rock-dev@dfki.de</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev">http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev</a>
</pre>
</blockquote>
<br>
<br>
<pre class="moz-signature" cols="72">--
Dipl.-Inf. Matthias Goldhoorn
Space and Underwater Robotic
Universität Bremen
FB 3 - Mathematik und Informatik
AG Robotik
Robert-Hooke-Straße 1
28359 Bremen, Germany
Zentrale: +49 421 178 45-6611
Besuchsadresse der Nebengeschäftstelle:
Robert-Hooke-Straße 5
28359 Bremen, Germany
Tel.: +49 421 178 45-4193
Empfang: +49 421 178 45-6600
Fax: +49 421 178 45-4150
E-Mail: <a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:matthias.goldhoorn@informatik.uni-bremen.de">matthias.goldhoorn@informatik.uni-bremen.de</a>
Weitere Informationen: <a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://www.informatik.uni-bremen.de/robotik">http://www.informatik.uni-bremen.de/robotik</a></pre>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Rock-dev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Rock-dev@dfki.de">Rock-dev@dfki.de</a>
<a class="moz-txt-link-freetext" href="http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev">http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev</a>
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
Dipl.-Inform. Elmar Berghöfer
Junior Researcher
Cognitive Robotics
Besuchsadresse der Nebengeschäftstelle:
DFKI GmbH
Robotics Innovation Center
Robert-Hooke-Straße 5
28359 Bremen, Germany
Postadresse der Hauptgeschäftsstelle Standort Bremen:
DFKI GmbH
Robotics Innovation Center
Robert-Hooke-Straße 1
28359 Bremen, Germany
Tel.: +49 421 178 45-4104
Zentrale: +49 421 178 45-0
Fax: +49 421 178 45-4150 (Faxe bitte namentlich kennzeichnen)
E-Mail: <a class="moz-txt-link-abbreviated" href="mailto:elmar.berghoefer@dfki.de">elmar.berghoefer@dfki.de</a>
Weitere Informationen: <a class="moz-txt-link-freetext" href="http://www.dfki.de/robotik">http://www.dfki.de/robotik</a>
-----------------------------------------------------------------------
Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
Firmensitz: Trippstadter Straße 122, D-67663 Kaiserslautern
Geschaeftsfuehrung: Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster
(Vorsitzender) Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
Sitz der Gesellschaft: Kaiserslautern (HRB 2313)
USt-Id.Nr.: DE 148646973
Steuernummer: 19/673/0060/3
-----------------------------------------------------------------------
</pre>
</body>
</html>