The paper we published at HPSR 2015 
(http://ieeexplore.ieee.org/document/7483103/ ) indeed emphasizes the 
application of GSP to high-speed links, where the simplicity of the scheme is 
most beneficial.

If “conjunction with flow isolation” means combination of the algorithm with a 
flow queueing arrangement, there is logically no restriction in realizing it. 
We tested FQ-GSP on ns2, getting similar results as with other FQ-AQM schemes 
(never worse, never overwhelmingly better in the scenarios we looked at). Since 
the algorithmic simplicity is not as critical in lower-speed links, we thought 
there was little value in trying to add one more scheme to an already crowded 
space.

Also (and this is just my opinion), I don’t think that combining FQ and AQM is 
a good idea, because it imposes a single policy on all flows despite the 
variety of their needs. I like a plain FQ with large buffer much better, 
because it guarantees bandwidth fairness and makes every application solely 
responsible for the queuing delay it gets.
Andrea


From: Jonathan Morton [mailto:[email protected]]
Sent: Friday, December 15, 2017 11:16 AM
To: Francini, Andrea (Nokia - US/Murray Hill) 
<[email protected]>
Cc: Roland Bless <[email protected]>; Wesley Eddy <[email protected]>; 
[email protected]; Lautenschlaeger, Wolfram (Nokia - DE/Stuttgart) 
<[email protected]>
Subject: Re: [aqm] Status of the GSP AQM?


Reading the spec, it looks very much as though it's tuned for implementation in 
relatively simple, high-speed nodes.  It doesn't look at all like it would work 
in conjunction with flow isolation, which is inherently a much more effective 
idea when feasible to deploy - which it should be at speeds up to at least 
1Gbps.

However, I could see some use for GSP when combined with host isolation, at 
nodes aggregating a large number of subcriber hosts' traffic and thus requiring 
very high aggregate throughput.  Host isolation doesn't require as many 
resources as full flow isolation, and is typically implemented anyway as part 
of per-subscriber provisioning.

If tests are carried out, that might be the best scenario to start with.

- Jonathan Morton
_______________________________________________
aqm mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/aqm

Reply via email to