RE: [bmwg] Comments on draft-ietf-bmwg-mcastm-12.txt

["Hickman, Brooks" <[e-mail]>]

Jerry,

Correction(Pushed send button to soon). The two sections I sent specified
6.1 and 8.1 in the example. Should
have been 6.1 and 8.2.

For example;
 
- Group join measurement in section 6.1 is obtain with no 
unicast entries in the FDB.
- Baseline measurement obtained in section 8.2 is obtained with unicast
  entries in the FDB, but no actual forwarding of unicast frames.
 
 The test is repeated with the DUT/SUT forwarding the unicast 
 addresses.

 
 You will note the "difference" between using the results of 
 the join group delay using the procedure in section 6.1 and the
 "baseline" as described in section 8.2 is the presents of unicast
 addresses in the FDB of the DUT/SUT.


> -----Original Message-----
> From: Hickman, Brooks [mailto:[e-mail]]
> Sent: Wednesday, May 21, 2003 4:22 PM
> To: Perser, Jerry; '[e-mail]'
> Subject: RE: [bmwg] Comments on draft-ietf-bmwg-mcastm-12.txt
> 
> 
> Jerry,
> 
> That was the point I was trying to convey. You are correct 
> that the tests
> described in sections 5.1(Multicast latency) and section 
> 6.1(Group Join
> Delay) describe only multicast traffic being offered. There 
> is no discussion
> of unicast in those sections. The intent of the burdened test 
> is to first
> populate the FDB of the DUT/SUT with unicast addresses and 
> obtain a latency
> or group join delay measurement as described in the 
> applicable sections 5.1
> or 6.1, respectfully. In the baseline(Section 8.1/8.2), no 
> unicast frames
> are being offered to the DUT/SUT, only the FDB be populated 
> with unicast
> addresses.
> 
> For example;
> 
> - Group join measurement in section 6.1 is obtain with no 
> unicast entries in
> the FDB.
> - Baseline measurement obtained in section 8.1 is obtained 
> with unicast
> entries in the FDB, but no
> actual forwarding of unicast frames.
> 
> The test is repeated with the DUT/SUT forwarding the unicast 
> addresses.
> 
> 
> You will note the "difference" between using the results of 
> the join group
> delay using the procedure
> in section 6.1 and the "baseline" as described in section 8.1 is the
> presents of unicast addresses in the FDB of the DUT/SUT.
> 
> 
> > -----Original Message-----
> > From: Perser, Jerry [mailto:[e-mail]]
> > Sent: Wednesday, May 21, 2003 3:02 PM
> > To: '[e-mail]'
> > Subject: RE: [bmwg] Comments on draft-ietf-bmwg-mcastm-12.txt
> > 
> > 
> > Brooks,
> > 
> > I do not understand.  Where in the 'baseline measurement' 
> > (section 5.1) do
> > you 'populating the FDB with unicast addresses'?  I do not 
> > see any unicast
> > traffic in the Multicast Latency Test.  From my reading of 
> > section 5.1, it
> > was a pure multicast measurement.
> > 
> > I also do not understand how the measurement done in section 
> > 5.1 is used to
> > calculate the measurement in section 8.1.  From my 
> > understanding, these are
> > two independent tests with independent metrics.
> > 
> > Jerry.
> > 
> > > -----Original Message-----
> > > From: Hickman, Brooks 
> > > Sent: Wednesday, May 21, 2003 10:57 AM
> > > To: Perser, Jerry; '[e-mail]'
> > > Subject: RE: [bmwg] Comments on draft-ietf-bmwg-mcastm-12.txt
> > > 
> > > 
> > > Jerry,
> > > 
> > > The baseline measurement in the case of the burdened section 
> > > includes populating the FDB with unicast addresses. The test 
> > > is then repeated with the DUT/SUT actually forwarding the 
> > > multicast traffic. Since the
> > > "Join Delay' test in section 6.1 does not include this 
> > > populating of the FDB, the "baseline" would not be the same. 
> > > This was why the text specifies a baseline test first.  
> > > 
> > > > -----Original Message-----
> > > > From: Perser, Jerry [mailto:[e-mail]]
> > > > Sent: Wednesday, May 21, 2003 9:00 AM
> > > > To: '[e-mail]'
> > > > Subject: RE: [bmwg] Comments on draft-ietf-bmwg-mcastm-12.txt
> > > > 
> > > > 
> > > > Brooks,
> > > > 
> > > > Why perform the baseline at all?  How is the baseline 
> > metric used in
> > > > determining the burdened metric?
> > > > 
> > > > The point is that these are two separate tests.  They even 
> > > > have their own
> > > > sections.  The way it is written now, you can run test A, but 
> > > > you can not
> > > > run test B without first running test A.  It's the dependency 
> > > > that is the
> > > > issue.
> > > > 
> > > > If you remove the baseline measurement dependency from 
> > > > Forwarding Burdened
> > > > Multicast Latency, what breaks?  Are the burdened results 
> > > > INVALID if you
> > > > don't already have baseline results?
> > > > 
> > > > Jerry.
> > > > 
> > > > > -----Original Message-----
> > > > > From: Hickman, Brooks [mailto:[e-mail]]
> > > > > Sent: Tuesday, May 20, 2003 1:41 PM
> > > > > To: '[e-mail]'
> > > > > Subject: [bmwg] Comments on draft-ietf-bmwg-mcastm-12.txt
> > > > > 
> > > > > 
> > > > > >8.1. Forwarding Burdened Multicast Latency
> > > > > >
> > > > > >- Why 'Perform a baseline measurement of Multicast Latency'? 
> > > > >  It is not
> > > > > used
> > > > > >in determining the Forwarding Burdened Multicast Latency.  
> > > > > You can run both
> > > > > >tests for comparison.  The burdened test is a separate test. 
> > > > >  Suggest to
> > > > > >remove this.
> > > > > 
> > > > > 
> > > > > Jerry, the FDB of the DUT/SUT is populated with the unicast 
> > > > > addresses prior
> > > > > to performing a baseline
> > > > > test. The test is then performed with the DUT/SUT actually 
> > > > > forwarding the
> > > > > unicast traffic. This
> > > > > also applies to section 8.2. 
> > > > > _______________________________________________
> > > > > bmwg mailing list
> > > > > [e-mail]
> > > > > https://www1.ietf.org/mailman/listinfo/bmwg
> > > > > 
> > > > _______________________________________________
> > > > bmwg mailing list
> > > > [e-mail]
> > > > https://www1.ietf.org/mailman/listinfo/bmwg
> > > > 
> > > 
> > _______________________________________________
> > bmwg mailing list
> > [e-mail]
> > https://www1.ietf.org/mailman/listinfo/bmwg
> > 
> _______________________________________________
> bmwg mailing list
> [e-mail]
> https://www1.ietf.org/mailman/listinfo/bmwg
> 
_______________________________________________
bmwg mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/bmwg