Document:  draft-ietf-ipfix-biflow-04.txt
Reviewer:  Christian Vogt
Review Date:  07 June 2007
IETF LC date:  03 May 2007

Summary:

Ready with comments.

Comments:

The document is in a good shape overall.  One thing I liked in
particular is the "Rationale and History" section, which describes
possible design alternatives and the reasons for not choosing those.

For a reader not so familiar with traffic measuring, it would be good if
the key technical terms were defined in the introduction, where they are
used the first time.  E.g., one can quickly assume what the "metering
and collection processes" do, and why these ought to be separate
processes.  But it would be great if the document itself would clarify
this, possibly including citations to existing documents.

Another thing that I think could be improved is that the figures on
pages 9 and on are neither referenced nor described in the text.  The
text would be nicer to read if it would directly refer to the figures
and explain matters based on the figures.

Also, the abbreviations used in the figures should be explained.  It is
currently left to the reader to guess that "MP" is a metering process,
and that an "Nx" is a node whoose traffic is being measured.

Otherwise, go ahead with the publication.