End-to-end reordering of packets on Internet is investigated. Packet streams transferred over the Internet are used to analyze the long-term and shortterm trends in reordering. Reorder Density (RD) is used to capture comprehensively and concisely the nature of reordering present in a stream. Simpler metrics are derived from RD for monitoring of longterm reordering trends. Reorder Entropy characterizes the reordering in a stream using a single value, reflecting the fraction of packets displaced as well as the magnitude displacements. The mean displacement of packets, defined with respect to late packets, early packets or all the packets, together with the percentage of packets that are displaced can also be used for this purpose. The pros and cons of each of these for monitoring long-term trends in reordering are addressed. The measurements presented indicate that for some end-to-end links the packet reordering shows clear daily and weekly trends.
Bin Ye, Anura P. Jayasumana, Nischal M. Piratla