The CollectNetworkMessageData() function currently processes data of which the result is not needed and can be skipped:

• When decoding a message with the purpose of only getting the PublisherId, the decoding of the rest of the message can be skipped after the PublisherId
• When the number of new messages in the message cache exceeds the 'MaxNumberOfMessages', processing the exceeding messages should be skipped, as they get sliced from the result

Furthermore the decoding of UADP PubSub messages involves inverting serveral ByteStrings which could be done faster when using a binary-side invert() method (see Mantis 0010171).