# Increase CPU load after add and use tag from IEC 61850 driver **URL:** \ **Category:** Ignition\ **Tags:** ignition81\ **Created:** [June 20, 2023, 1:16am UTC](https://forum.inductiveautomation.com/t/increase-cpu-load-after-add-and-use-tag-from-iec-61850-driver/75249 "2023-06-20T01:16:18Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![indi](https://avatars.discourse-cdn.com/v4/letter/i/f475e1/32.png) [@indi](https://forum.inductiveautomation.com/u/indi)\ **Post date:** [June 20, 2023, 1:16am UTC](https://forum.inductiveautomation.com/t/increase-cpu-load-after-add-and-use-tag-from-iec-61850-driver/75249/1 "2023-06-20T01:16:18Z")
I take datas from iec 61850 OPC server from different machine (remote server) to my gateway through OPC UA connection, the CPU load only 1%. After i try to connect directly to devices with iec 61850 driver modul in ignition, 39 devices connected, cpu load increase to 30-50℅. I use poll mechanism, access datas directly in model folder from device in tag browser, i havent use RCB mechanism yet. Is it normal in increasing cpu load? If normal, what can i do for decreasing cpu load? Use RCB? ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [June 20, 2023, 1:30am UTC](https://forum.inductiveautomation.com/t/increase-cpu-load-after-add-and-use-tag-from-iec-61850-driver/75249/2 "2023-06-20T01:30:42Z")
I'm not up to speed on this particular driver, but polling is always going to be more work for a driver than a simple OPC UA subscription (which makes the remote server do the bulk of the work). What kind of polling paces are you using, and for how many data points? What are the hardware specs of your Ignition gateway (CPU cores/clock & RAM)? ---
**Author:** ![Kevin.Herron](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/kevin.herron/32/7259_2.png) [@Kevin.Herron](https://forum.inductiveautomation.com/u/Kevin.Herron)\ **Post date:** [June 20, 2023, 1:31am UTC](https://forum.inductiveautomation.com/t/increase-cpu-load-after-add-and-use-tag-from-iec-61850-driver/75249/3 "2023-06-20T01:31:54Z")
Using RCB is standard practice with the 61850 protocol. You are not meant to poll data points as the primary means of data acquisition. I’m not certain it’s the cause of your CPU usage, so maybe you can work with support on that, but you should be switching to reporting regardless. ---
**Author:** ![indi](https://avatars.discourse-cdn.com/v4/letter/i/f475e1/32.png) [@indi](https://forum.inductiveautomation.com/u/indi)\ **Post date:** [June 20, 2023, 1:33pm UTC](https://forum.inductiveautomation.com/t/increase-cpu-load-after-add-and-use-tag-from-iec-61850-driver/75249/4 "2023-06-20T13:33:37Z")
I take 25000 tags every second. Here are the hardware specs of the gateway Windows Server 2019 64bit Intel(R) Xeon(R) E-2224G CPU @3.50GHz (4CPUs), -3.5GHz 32 GB RAM. ---
**Author:** ![indi](https://avatars.discourse-cdn.com/v4/letter/i/f475e1/32.png) [@indi](https://forum.inductiveautomation.com/u/indi)\ **Post date:** [June 20, 2023, 1:38pm UTC](https://forum.inductiveautomation.com/t/increase-cpu-load-after-add-and-use-tag-from-iec-61850-driver/75249/5 "2023-06-20T13:38:17Z")
![aaad](https://us1.discourse-cdn.com/inductiveautomation/original/3X/2/1/21318d862a10cd6b7222888b4e9fc8b413aa863f.png) I take data directly form model folder. Next, I will switch to reporting, hopely can lower CPU usage