[LARTC] QoS with Artifficial Intelligence

[LARTC] QoS with Artifficial Intelligence

[Gomi]

Hello everyone, it is not the first time i discuss this topic here, but now
it has come the time to actually do it.

My idea is to set up a daemon to run QoS on linux, with a particularity, add
some A.I. capabilities to our system and hence, be able to change QoS
"topology" every certain time to obtain the maximum performance.

I first want to teach the system which parameters should i vary, and hence i
would like all of you to tell me, which do you think i should change.

Any ideas? Anybody is welcome to join!! :)

________________________________________________
Message sent using UebiMiau 2.7.2

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

Re: [LARTC] QoS with Artifficial Intelligence

[Ed Wildgoose]

>My idea is to set up a daemon to run QoS on linux, with a particularity, add
>some A.I. capabilities to our system and hence, be able to change QoS
>"topology" every certain time to obtain the maximum performance.
>
>I first want to teach the system which parameters should i vary, and hence i
>would like all of you to tell me, which do you think i should change.
>  
>

The paramters to vary are easy enough, after all your are simply 
segmenting the network traffic by type and then throttling it to some 
lower proportion of the total network capacity.

Your problem is determining the "fitness" function that you are 
optimising?  After all if you can describe the fitness function in 
enough detail then you can simply implement an optimal traffic control 
function to inplement that desired policy... In other words I'm not sure 
where the AI bit would fit in?

Good luck

Ed W
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

Re: [LARTC] QoS with Artifficial Intelligence

[Jonathan Day]

I'm guessing the "AI" bit is a simplified way of
expressing what they're after. "AI", per se, is
meaningless, because it's undefined.

What I -think- they want to do is examine the current
behaviour of the traffic, anticipate how it is going
to behave next, set the QoS to match that expectation,
and then "learn" both from what actually happens and
from the quality of traffic flow produced.

A self-adjusting QoS is a tough problem, and I'm not
aware of anyone who has done much research into such
things. One of the problems is that traffic flow is
random, rather than periodic or constant. There is no
obvious way, at the start, to tell if a given transfer
is going to be large or small. Also, you can't just
pick a certain set of variables to change, because the
values are highly interdependent. You've got to change
them all, and that makes the problem much more
complex.

A much better approach would be to look at QoS over
the network, rather than at a single point. This is
because optimising a single point can make some
subsequent point perform worse. What you want is to
optimise the system in totality.

On a relatively small network, this is relatively
easy. Just have all the routers periodically transmit
their current settings, the statistics per interface
per traffic class (you don't care about the source for
this), the router load and the estimated latency &
packet loss. This data goes to a central server, which
determines the settings likely to work best in the
future.

Because we're dealing with relatively large pools of
aggregated random data, we can apply statistical
techniques. I'd start off with looking at queueing
theory, which deals with the forming and processing of
a linear series of random events which need
processing. This should be able to tell you how large
a bucket you want for each class. (ie: the hard
limit.)

Calculating the optimal number of classes is harder,
but you do know that the sum of the upper soft limits
for all classes must be equal to or less than the
capacity of the router. To me, that suggests you might
be able to get a good guess for the soft limits via
the SIMPLEX method (also known as Operational
Reseaarch).

Once you know the soft and hard limits, you can
determine the number of classes by queueing theory -
it is the minimum number of "queues" into which you
need to split the traffic to get maximum throughput,
avoiding "empty" queues.

This approach would not work for a single router. The
traffic is "random", but it is not random enough, and
statistics doesn't work well on single points. It will
also fail on very large networks, because the overhead
of transmitting the metadata would become too large,
and by the time the data was processed, the results
would no longer have much meaning.

For very large networks, you could "escape" the
problem by regarding it as a large collection of
overlapping medium-sized networks. You could then
process each of the medium-sized networks using the
above method. Where two (or more) manager nodes
instruct a specifc router, the router would take the
average recommended values. (If you know in advance
that one of the managers is more relevent than
another, then simply weight the average accordingly.)

***WARNING***

All of the above is speculative, in the sense that it
-should- work, but I don't have a large enough test
network to verify it. Nor do I know the optimum number
of routers/hosts where the numbers are statistically
meaningful, yet where the metadata doesn't interfere
with the traffic flow AND where the results can be
passed back and acted on within the timeframe for
which they are valid.

I say the above -should- work, because there are
methods for solving the various parts of the
problem-space. If you combine these methods correctly,
you'll end up with a solution to the whole problem
space.

Now comes crunch #1. Although traffic flow is random,
in aggregate, it is not necessarily random when split
into classes. Certain events (eg: backups over a
network, connecting to a DHCP server on power-up, etc)
are mostly going to occur at specific times. You could
always complicate the manager nodes, by adding a diary
of known large-scale events, so that it can statically
allocate the correct bandwidth for those and then
dynamically allocate whatever is left.

Crunch #2. Statistical methods, herustics, etc, are
generally slow. Changes in network behaviour can be
fast. To be meaningful, the results have to be
calculated and passed back to the routers so they can
update their QoS methods before the traffic has
changed significantly.

Crunch #3. Probably the biggest problem of all.
Transmitting the metadata and then getting the updated
QoS information is going to take up bandwidth. This is
going to alter the flow. If you're lucky, the change
will be short-lived. If you're unlucky, the knock-on
effects (eg: resent packets, changes in
load-balancing, etc) will disturb the pattern
significantly and unpredictably, making the new QoS
parameters useless.

Crunch #4. This whole system is a set of feedback
loops, which aim to produce a net negative feedback
system. Because you don't know how traffic will
actually change with time, any individual QoS loop may
become a positive feedback loop. The system, as a
whole, is therefore "meta-stable" - it's stable, but
only within certain limits. Behaviour outside of those
limits could potentially crash the whole system.

Pro-active QoS is certainly possible. However, I have
absolutely no idea as to how you could build a
pro-active QoS that was both stable AND responsive in
the sorts of times a network would need to respond in.

--- Ed Wildgoose <lists@wildgooses.com> wrote:

> 
> >My idea is to set up a daemon to run QoS on linux,
> with a particularity, add
> >some A.I. capabilities to our system and hence, be
> able to change QoS
> >"topology" every certain time to obtain the maximum
> performance.
> >
> >I first want to teach the system which parameters
> should i vary, and hence i
> >would like all of you to tell me, which do you
> think i should change.
> >  
> >
> 
> The paramters to vary are easy enough, after all
> your are simply 
> segmenting the network traffic by type and then
> throttling it to some 
> lower proportion of the total network capacity.
> 
> Your problem is determining the "fitness" function
> that you are 
> optimising?  After all if you can describe the
> fitness function in 
> enough detail then you can simply implement an
> optimal traffic control 
> function to inplement that desired policy... In
> other words I'm not sure 
> where the AI bit would fit in?
> 
> Good luck
> 
> Ed W
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
> http://lartc.org/
> 



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Find what you need with new enhanced search.
http://info.mail.yahoo.com/mail_250
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

Re: [LARTC] QoS with Artifficial Intelligence

[Chris Bennett]

I'm not sure what you mean by AI.  I suppose you could mean that you're 
going to feed various QoS parameters into a neural net and "teach" the 
neural net to vary the parameters according to conditions... but somehow I 
think it unlikely that this is what you mean.

What is the specific situation you're trying to deal with, and what exactly 
are you referring to when you say maximum performance?

Another person recently asked about how to implement QoS in a heavily 
oversubscribed environment, for example.  If, even during the times when the 
network is overburdened, there is always at least enough bandwidth to handle 
the high priority latency sensitive data (possibly a big assumption), I 
suggested the possibility of monitoring a steady ping to see when the 
buffers fill and the ping value skyrockets.  At that time, a QoS script 
could be run that would assume a lesser bandwidth rate, and hopefully slow 
up lower priority traffic and cause the buffer to empty in short order.  In 
this way one might be able to maintain at least somewhat reliable level of 
low latency, while still trying to maximize the use of the dynamic 
bandwidth....  a sort of "artificial intelligence".

You could be talking about something completely different, so perhaps you 
could provide more information about what specific situation you are dealing 
with?

----- Original Message ----- 
From: "Gomi" <gomi@perezoso.net>
To: <lartc@mailman.ds9a.nl>
Sent: Wednesday, December 01, 2004 8:50 AM
Subject: [LARTC] QoS with Artifficial Intelligence


> Hello everyone, it is not the first time i discuss this topic here, but 
> now
> it has come the time to actually do it.
>
> My idea is to set up a daemon to run QoS on linux, with a particularity, 
> add
> some A.I. capabilities to our system and hence, be able to change QoS
> "topology" every certain time to obtain the maximum performance.
>
> I first want to teach the system which parameters should i vary, and hence 
> i
> would like all of you to tell me, which do you think i should change.
>
> Any ideas? Anybody is welcome to join!! :)
>
> ________________________________________________
> Message sent using UebiMiau 2.7.2
>
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
> 

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

Re: [LARTC] QoS with Artifficial Intelligence

[Chris Bennett]

You're really serious?  Hmm... okay.

As far as giving any specific help, the last time I worked with neural net 
software was in college which was... um.. over 12 years ago.  So I really 
have no idea what the latest available software for doing something like 
this is.  Sorry.

From a theoretical perspective I'll say this:  as a research project I think 
you could get interesting results by trying out something along the lines of 
having the neural net vary parameters and receive positive or negative 
feedback based on the latency, throughput or something like that.  By 
"interesting" I mean exactly that... interesting but not necessarly useful. 
I'd hate to be the one who has to suffer through using a network that is in 
training mode.  And while there may be patterns that a neural net could 
learn, I'm not sure how well the neural net would move from reacting to 
conditions to actually anticipating conditions.  And to the extent that it 
ever did correctly anticipate certain conditions, what would be the cost of 
incorrect guesses..?  I suppose it all depends on how predictable the 
activity on a particular network is.

My one practical thought is this: I'd try to shorten the feedback loop 
considerably.  Every minute or so is, I'd guess, way too long.  I'd go for 
every 10 seconds or so if possible.

Anyway, guess I'm saying its a "neat idea" but I'm not personally interested 
in pursuing it.  Good luck.  Try it out.. write a paper, become famous.

----- Original Message ----- 
From: "Gomi" <gomi@perezoso.net>
To: "Chris Bennett" <chris@symbio.com>; 
<"lartc@mailman.ds9a.nl"@alpha.symbio.com>
Sent: Tuesday, December 21, 2004 8:04 AM
Subject: Re: [LARTC] QoS with Artifficial Intelligence


>I was actually thinking in every minute or so, read statistics from queues,
> and SNMP from dsl routers for example, and vary the queues bandwith, their
> limit, their queuelenght or even the burst and cburst.
>
> I was actually thinking in implementing a neuronal network to do so, what 
> do
> you think?
>
> --------- Mensagem Original --------
> De: Chris Bennett <chris@symbio.com>
> Para: lartc@mailman.ds9a.nl <lartc@mailman.ds9a.nl>
> Asunto: Re: [LARTC] QoS with Artifficial Intelligence
> Fecha: 20/12/04 23:44
>
>>
>> I'm not sure what you mean by AI.  I suppose you could mean that you're
>> going to feed various QoS parameters into a neural net and
> &quot;teach&quot; the
>> neural net to vary the parameters according to conditions... but somehow 
>> I
>> think it unlikely that this is what you mean.

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/