Thank you Tom,

that is simply great!

There I see this:

running the nvidia command alone
#nvidia-smi 
Thu Apr 17 08:30:14 2014       
+------------------------------------------------------+                       
| NVIDIA-SMI 331.38     Driver Version: 331.38         |                       
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Quadro 4000         Off  | 0000:01:00.0      On |                  N/A |
| 90%   93C    P1    N/A /  N/A |    201MiB /  2047MiB |     38%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Compute processes:                                               GPU Memory |
|  GPU       PID  Process name                                     Usage      |
|=============================================================================|
|  No running compute processes found                                         |
+-----------------------------------------------------------------------------+

and here running the line from the script

#>nvidia-smi --format=csv --query-gpu=index,gpu_name,temperature.gpu,fan.speed,power.draw,utilization.gpu,utilization.memory
index, name, temperature.gpu, fan.speed [%], power.draw [W], utilization.gpu [%], utilization.memory [%]
0, Quadro 4000, 93, 90 %, [Not Supported], 39 %, 5 %

looks to me that all the data is sent to observium,
is it something that I can do to make it show in the graphs?

Thank you!


2014-04-16 23:10 GMT+02:00 Tom Laermans <tom.laermans@powersource.cx>:
root@observium1:/opt/observium/scripts/agent-local# head nvidia-smi
#!/bin/sh
# report nvidia hardware information
# requires nvidia-smi


On 16/04/2014 22:37, Mihai - Cristian Satmarean wrote:
Thank you all,

as usual very fast replies
I would like to keep it inside observium before giving collectd a try,
but I cannot find the gpu-temp agent, I am using the latest community version.

Thank you!


2014-04-16 17:24 GMT+02:00 Tom Laermans <tom.laermans@powersource.cx>:
There's a Unix Agent script for the GPU temperature. No mempools or load (yet) :)

Tom


On 04/16/2014 03:20 PM, Peter Childs wrote:
Perhaps with collectd and this?
https://github.com/faxm0dem/cuda-collectd


________________________________________
From: observium [observium-bounces@observium.org] on behalf of Mihai - Cristian Satmarean [mihai.satmarean@gmail.com]
Sent: Wednesday, 16 April 2014 10:08 PM
To: Observium Network Observation System
Subject: [Observium] graph NVIDIA GPU parameters

Dear all,

sorry if this was already addressed,
is it somehow possible to graph the Nvidia GPU parameters in observium?
Load, memory, temp etc.

Thank you!
Mihai
_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium



_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium


_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium