Scientific research communities move very large volumes of experimental data between large and small international sites, using connectivity provided by National Research and Education Networks (NRENs). NRENs are therefore particularly interested in understanding how their networks are used by which scientific activities to make informed decisions on how to scale their infrastructures.

This talk introduces Scientific Network Tags (scitags), an initiative promoting identification of the science domains and their high-level activities at the network level. Scitags provides Network Flow and Packet Marking that can be captured by NRENs to better understand existing network patterns, estimate network usage and track activities. Network packet marking uses UDP "firefly" packets, whereas flow marking uses the flow label in the IPv6 header. Scitags is an open source framework developed by the Research Networking Technical Working Group (RNTWG) of the Worldwide LHC Computing Grid (WLCG) community; it is aimed at high-energy physics experiments, but is easily applicable to other scientific communities.

The presentation also demonstrates the usage of flowd, a daemon that has different plugins to retrieve flow identifier and a set of backend to mark the traffic, to add Scitags support to an actual data transfer service (StoRM WebDAV).