Abstract: Traffic sampling techniques are widely used for traffic measurements at a high link speed to prevent an exhaustion of resources and to limit the measurement costs. However, the challenge of an effective sampling method for IPv6-based networks is as yet unmet. This paper proposes a traffic sampling measurement method to take the challenge. For ensuring randomness of sample, we use entropy as an evaluation tool to analyze the bit randomness of each byte in IPv6 packet headers, and conclude that the last one byte of the Payload Length field and byte numbers 8, 12, 14, 15 and 16 of the IPv6 source and destination address fields which have both unchangeability during forwarding and high bit entropy values. We estimate whether a packet is sampled based on a hash function computed over the selected bytes. Therefore, the entire packet header content is not taken into account in our sampling method. The advantages of the method are improved randomness of the sample and the runtime efficiency of the sampling algorithm. Finally, through experiments using real IPv6 traffic traces, we prove that the sampled traffic data can correctly reflect the packet size distribution of full packet trace.

Key words: IPv6, sampling measurement, distributed measurement, information entropy