|Network Data Structures|
|Data structures used in representing network data. |
|Data structures used for representing statistical data. |
|Data structures used to describe attributes, either of Aggregates or of Packets. |
These data structures were built to be generic, to allow the representation of various and sundry different models of network traffic. The only one that has been tested is one where the top-level aggregations are bidirectional flows (based on protocol, source and destination network addresses, and source and destination transport-layer ports), each of which has two unidirectional half-flows, one in each direction. They allow multi-layer aggregation; any Aggregate that is to contain packets should be built as a Trace, while those that are to contain other Aggregates should be a Decomposition. Mixed Aggregates (those containing packets and other Aggregates) are not properly supported yet.