The correct Cisco IPR is http://datatracker.ietf.org/ipr/2540/.
Vishal has done pioneering work in TCP modeling and in bringing in the
classical PI controller into the AQM space. I have certainly learned a lot
from him. We referenced their paper in our draft and paper, we even
mentioned his name in our PIE paper.
PIE and PI equations are similar because they are all classical
Proportional Integral controller (PI) based. We have never claimed the
ownership of classical PI controller that works like a charm. PI AQM's
design: Equation (8) in the original paper is a direct z-domain transfer
of the classical PI controller: K*(s/wg+1)/s. Our designs share a lot of
the good features inherited directly from the classical PI controller.
However, PIE is not equivalent to PI AQM:
----------------------
1) Our design is more than just a z-transform of the classical PI
controller.
Beyond math, any engineering problem has implementation and parameter
sensitivity issues associated with them. In our form of the PI controller,
we control the offset to the reference level and second moment of the
latency
independently by alpha and beta. They are often orders of magnitude apart.
This makes PIE a lot easier to tune.
2) PIE adopted the classical PI controller for latency. The latency
measurement through rate estimation turns out to be more challenging than
expected. We have ideas on how to determine when we
can have a rate sample, etc. We have to make the rate calculation work in
a robust way for real network scenarios.
3) PIE's auto-tuning:
if (PIE->drop_prob_ < 0.1%) {
alpha'= alpha/128 and beta' = beta/128;
} else if (PIE->drop_prob_ < 1%) {
alpha'= alpha/16 and beta' = beta/16;
} else if (PIE->drop_prob_ < 10%) {
alpha'= alpha/2 and beta' = beta/2;
} else {
alpha'= alpha and beta' = beta;
}
Self-tuning enables PIE to cover various network conditions.
4) We have improved the single TCP flow's performance dramatically by
reducing bursty drops and adopting bursty protection regions. We can
provide good performance for one single TCP session within seconds.
------------------------
Vishal's pioneer work certainly is an important step in advancing our
knowledge in AQM. PIE is a solid step forward given the requirements of
current challenging environments. Both PIE and PI AQM stand on the giant
shoulders of the classical PI controller.
Regards,
Rong
On 3/3/15 10:07 PM, "Dave Taht" <[email protected]> wrote:
>Two items:
>
>A) The IETF IPR filing http://datatracker.ietf.org/ipr/2187/ points
>to the wrong patent: 13/874,500. A google search for that patent
>number brings up http://www.google.com/patents/US20130239255";
>
> It is ironically relevant to the discussions at hand, as that one
>concerns:
>
>Abstract:
>
>"Provided are methods of increasing the tolerance of a plant to
>abiotic stresses and/or increasing the biomass and/or increasing the
>yield of a plant by expressing within the plant an exogenous
>polynucleotide homologous to SEQ ID NO:13."
>
>... As I consider myself a near-vegetable, and am 40 pounds heavier,
>and not responding particularly well to antibiotics, after
>participating for the past several years on all the ietf mailing lists
>I just got off of. I am sure that upon acceptance of pie in the ietf,
>that making that particular patent more generally available for all to
>use would probably have similar effects on others.
>
>The correct patent number for PIE, 13/874,600, is here:
>
>http://www.google.com/patents/US20140328175
>
>I would appreciate that the IPR filing be corrected.
>
>In the meantime, here's some more great NSFW george carlin routines!
>
>https://www.youtube.com/watch?v=tVlkxrNlp10
>
>B) Vishal Misra (author of PI) gave me pointers to his PI papers
>recently (and he had NO idea at all his work was used for pie! - he
>got his marketing department to issue a press release about it:
>http://engineering.umass.edu/news/got-bufferbloat-umass-amherst-research-b
>ehind-fix
>)
>
>I usually have a pretty strict policy about never reading patents, but
>I read all those papers [1], and both! patents above. I had not fully
>realized that the PI-AQM work went as far back as 2001. The PI update
>equation and the PIE update equation, look pretty darn similar, just
>the meanings of two variables, changed.
>
>C) I am kind of curious if any working code for the original PI
>algorithm exists for linux?
>
>D) oh, never mind, I will blog about the rest one day.
>
>[1] still prefer fq_codel.
>
>--
>Dave Täht
>Let's make wifi fast, less jittery and reliable again!
>
>https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb
>
>_______________________________________________
>aqm mailing list
>[email protected]
>https://www.ietf.org/mailman/listinfo/aqm
_______________________________________________
aqm mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/aqm
