ABSTRACT:

Packet forwarding prioritization (PFP) in routers is one in every of the mechanisms commonly out there to network operators. PFP can have a important impact on the accuracy of network measurements, the performance of applications and therefore the effectiveness of network troubleshooting procedures. Despite its potential impacts, no information on PFP settings is instantly obtainable to finish users. In this paper, we gift an end-to-end approach for PFP inference and its associated tool, POPI. This is the primary attempt to infer router packet forwarding priority through finish-to-finish measurement. POPI enables users to find such network policies through measurements of packet losses of various packet types. We tend to surveyed all connected network operators and received responses for about half of all confirming our inferences. Besides, we have a tendency to compared POPI with the inference mechanisms through other metrics like packet reordering (referred to as out-of order (OOO)). OOO is unable to seek out many priority methods like those implemented via traffic policing. On the opposite hand, apparently, we found it can detect existence of the mechanisms that induce delay variations among packet types such as slow processing path within the router and port-primarily based load sharing.