we've had a few issues in the past weeks with the implementation regarding performance & resource usage in umh-core and overloaded devices at customers which led to unstable connections to ignition through opcua specifically when there are tenthousands of tags on a single instance