(I cloned 520 to hopefully get the tagging right...)

This improvement describes the requirements and constraints to implement IMIX Genome support in VSPERF.

RFC 6985 describes several different methods to communicate clearly about test streams with multiple packet sizes. It is a goal for test equipment manufacturers to produce exact sequences of sizes, which are thought to be a key step toward reproduce-able results across different experiments with different traffic generators. However, some generators do not offer this capability for exact stream specification.

Perhaps the most common specification of multiple sizes across different generators is the method described in https://tools.ietf.org/html/rfc6985#section-6.2 Table of proportions. This would require a list of frame sizes, along with the fraction of total frames will have this size in the test stream.

Martin imagined something like (in TC definition, where only changes to TRAFFIC array can be defined):

TRAFFIC = {
               ‘multistream’ : 30000
               ‘l2’: {
                              ‘srcmac’ : ’#01:00:00:00:00:00:00’
                              ‘dstmac’ : ’AA:00:00:00:00:00:00’
                              'sizes'  : '64'  ...
                              'fract'  : '30'  ...
                      }
               }

where ‘#’ sign marks, what should be iterated. Of course we should figure out more generic notation, which allows to iterate also through values from given list, etc. Iterating through values in a list could serve the need for a specification of sizes, but the fraction of total packets is a different sort of specification.

As Martin has done with IXIA and Spirent comparison, we need to do the same with each generator's IMIX capabilities and see what form of general specification best aligns with their capabilities (and I didn't have enough time to do that tonight...).

This issue is related to/overlaps with VSPERF-97 and VSPERF-403.