How to shorten the response time|OPC UA Standard|Forum|OPC Foundation

You need to determine where the bottle neck is.

For a delay that long is it likely the processing in the server that is causing the issue.

Do you have other clients which are faster?

Have you used wireshark to confirm that the delay is on the server side?

Can you instrument your server?

What SDK did you use to build it?

It is possible it is related to server design: a server that did an update and waited for the next scan to check that the write was successful would introduce delays based on the scan frequency. If this is configurable you may be able to lower it.

It could be an implementation problem with the UA server. You could confirm it is specific to the UA server if you have another way to write to the PLC, such as a engineering tool provided by the PLC vendor.

I would also test by subscribing to the data point with a SamplingInterval and PublishingInterval of 0 or the lowest possible value. See how long it takes to get an update after sending the write. If the update arrives before the write response then that is evidence that the server is likely waiting for confirmation before returning a response.