Re: Naive question on multiple TCP/IP channels and please dont start a uS NN debate here unless you really want to.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 6 Feb 2015, at 16:57, Michael Richardson wrote:

> 
> Jim Gettys <jg@xxxxxxxxxxxxxxx> wrote:
>> ​What effect does this algorithm have in practice? Here are some examples:
>> o real time isochronous traffic​ (such as VOIP, skype, etc) won't build
>> a queue, so will be scheduled in preference to your bulk data.
>> o your DNS traffic will be prioritized.
>> o your TCP open handshakes will be prioritized
>> o your DHCP & RA handshakes will be prioritized
>> o your handshakes for TLS will be prioritized
>> o any simple request/response protocol with small messages.
>> o the first packet or so of a TCP transfer will be prioritized: remember,
>> that packet may have the size information needed for web page layout in it.
>> o There is a *positive* incentive for flows to pace their traffic (i.e.
>> to be a good network citizen, rather than always transmitting at line rate).
> 
>> *All without needing any explicit classification.  No identification of
>> what application is running is being performed at all in this algorithm.*
> 
> This last part is I think the part that needs to be shouted at residential
> ISPs on a regular basis.  I wish that the IETF and ISOC was better able to
> do this... in particular to ISPs which do not tend to send the right people
> to NANOG/RIPE/etc.
> 
Explicit class-based queueing is seeing fairly substantial deployment in some places - such as the UK - where for a few years now the default home routers (Thomson/Technicolor TG587/582 etc) for a number of the big ISPs (Plusnet, O2/Sky, Talk-talk and others) have shipped preconfigured with 5 queuing classes that classify traffic and provide for differing treatment. They also have some ALGs that work with SIP/H.323. I'm not aware of AQM enabled on the individual queues but at least they separate the traffic into different queues - albeit based on port number or ALG classifiers. Better than nothing anyway.

Also the DOCSIS3.1 standard now mandates the use an AQM - namely PIE, though others can be implemented. I'm not sure where that is in terms of deployment though. There's a good report on it here:
http://www.cablelabs.com/wp-content/uploads/2013/11/Active_Queue_Management_Algorithms_DOCSIS_3_0.pdf

Piers


> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     mcr@xxxxxxxxxxxx  http://www.sandelman.ca/        |   ruby on rails    [
> 






[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]