As per the data flow, the Interface node performs the Exception Reporting for the values received from the data source and sends these values to the PI Data Archive node and here the Compression will be performed and this compressed data is sent to Event Queues and then to Archive files. This the general data flow if there is no buffering configured in the Interface node.
But, if we enable buffering in the Interface node, the Buffer Subsystem will perform the Compression Deviation too. And my question here is, if we enable buffering in the Interface node, does this Buffer Subsystem handle the compression all by itself by skipping the compression taking place in PI Data Archive node? Or will it perform the compression only during buffering (when PI Data Archive is down) ?