Citadel handling 1K SV

[yakir]

Hi,

I am using NI Citadel 5 for logging about 1000 SV in one LV 8.2.1 project (Acts as a data server).

I have several clients with LV 2011.

I have some data historical viewers in  5 client PCs. Unfortunately I suffer from bad performances of the viewers.

1. Are there any performeces differences in NI Citadel between LV 8.2.1/2011/later?

2. Are there any relations to SV throughput rate upgrade between LV versions?

3. Is anyone knows any recommended archiving + viewing tool for this amount of data (or bigger)?

Thanks,

Yakir.

[Joel-I]

Hi yakir,

What is the update rate for your Shared Variables?
What viewer are you using currently?

There do not seem to be major performance changes in Shared Variables or Citadel from 8.2 to 2011.

You have several options for an archiving and viewing tool.  One option is to use NI HyperTrend, which can be found in LabVIEW by going to the Front Panel and right clicking » DSC Module » Trend Control » NI HyperTrend

Alternatively, you can look at the DSC Module pallete.

There are also examples that can be found by going to Help » Find Examples » Toolkits and Modules » Datalogging and Supervisory Control » HMI Design » NIHyperTrend Graph XML Configuration.lvproj

Regards,

[yakir]

Hi,

What is the update rate for your Shared Variables?

Update rate is varying - from 10ms to 1sec
What viewer are you using currently?

I use the historical data viewer in NI MAX

My main problem is the slow data transfer rate between my server and the NI MAX clients..

Regards,

Yakir.

[Tortu]

Greetings yakir

When you say you suffer from bad performance on the viewers, what symptoms accompany that bad performance? Is there a memory rise? Increased CPU or disk usage? Do they crash? Do you think the network connection could be saturated and that is decreasing the data transfer rate between the server and the client PCs?

What happens if you only use 1 client PC? Is it still slow? If you use LabVIEW 8.2.1 in the client PCs, does it work fine? Let us know.

Finally, because of the number and update rate of traces, you could try Option 2 of this Knowledge Base article to improve performance. (Assuming the poor performance is due to a memory problem)

http://digital.ni.com/public.nsf/allkb/F766A77CC235BD9D862576C000703A8E?OpenDocument

Warm Regards,