PAS600 operating metrics

[PiotrJakubczyk]

Hi

Did anyone tested what are allowed operating metrics for PAS600? 

What is allowed CPU usage and memory usage? 

I plan to log data from 64 Modbus TCP/IP devices. Every device have 10 registers pooled every minute. 

Modbus devices are Ecostruxure Building Operation (EBO) AS-P devices. Model used is AS-P V1.0

Who I can talk to about performance tests of PAS devices?

I'm beginner in relation to PAS devices but want to avoid beginner errors as much as it possible 😉  

[Guillaume_Evrard]

hello @PiotrJakubczyk 

With 2.0 there should not be any issue as you can go up to 128 modbus tcp devices and 5000 datapoints if I remember well (refer to release note) 

https://www.productinfo.schneider-electric.com/ecostruxurepanelserverrn/doca0178-ecostruxure-panel-s...

What will be your usage? Connection to ETP ?

Memory usage and cpu usage are displayed on the EWP.

Performance testing has been made by validation team

[PiotrJakubczyk]

Thanks for reply @Guillaume_Evrard  

My usage will be pooling data from Modbus TCP devices to Schneider Electric Cloud witch PAS600. 

In current configuration I was able to add to PAS600L 64 Modbus TCP/IP devices. I was able to configure data sampling on 45 devices before i get bellow warning. Every Modbus device have 10 registers. Data sampling was set to 1 minute on 45 devices. 19 devices have no data sampling configured. 

"Your current configuration is close to reaching the system sampling limit.
It is recommended that you do not add any additional data samples unless existing sampling settings are reduced"

Hardware is version 001.000.000 

 And PAS600 software is 001.010.000

So am I right that sampling is limited to 45 samples per minute? Multiple register read function is used to read 10 registers). 450 registers per minute. PAS600 and Pas800 have same processor so limitations will be similar? 

[Guillaume_Evrard]

Hello @PiotrJakubczyk 

Please upgrade to 2.0 to reach 128 modbus tcp devices and 5000 data points.

You are having 2 limits:

- 5000 data points

- 30 000 measurements published / hour ( = nb measurement * 60/ poling cycle in min) or 500 points /min but I do not like this statement.
If you want to log every minute you can have max  500 data points

if you want 64 devices with 10 data each every minute it does not work, you need to log less frequently 

Kind Regards