Re: [PCN] "blind marking" not a big advantage

Hi Phil,

yes, this issue boils down to the tradeoff implementation complexity vs. 
overtermination in multiple bottleneck scenarios.

I can try to quantify the overtermination in future performance studies. 
Likewise, it would be good to have input from manufacturers about the 
complexity increase when going from "blind marking" to "stateful marking".

Regards,

    Michael

philip.eardley@bt.com wrote:
> In terms of 3SM edge behaviour [the 3rd of the 4 columns below), then
> termination marked packets are metered.
>
> For CL edge behaviour [the 4th of the 4 columns below), then including
> termination marked packets can lead to over-termination in multiple
> bottleneck scenarios. 
> Possibly this is a trade-off work making? Ie simpler behaviour for
> PCN-interior-node for more complexity in edges or more uncertainty in
> outcome of termination?
>
>
>
>   
>> -----Original Message-----
>> From: pcn-bounces@ietf.org [mailto:pcn-bounces@ietf.org] On Behalf Of
>> Michael Menth
>> Sent: 25 February 2008 10:32
>> To: pcn
>> Subject: [PCN] "blind marking" not a big advantage
>>
>> Hi,
>>
>> I have seen that some encoding discussions underlined the benefit of
>> "blind marking", i.e. the codepoint can be set without reading the ECN
>> bits before. I just want to point out that this advantage is not so
>>     
> big
>   
>> because the meter needs to know the marking anyway to avoid that
>> termination marked packets are metered. I just copied from the
>> comparison draft
>> http://www.ietf.org/internet-drafts/draft-charny-pcn-comparison-00.txt
>>
>>
>>     
> |-------------------------------------------------------------------|
>   
>>    |Metering and  | do not meter  |  meter all pkts  | do not meter
>>     
> |
>   
>>    |remarking of  | AM-marked     |  for admission   | TM-marked pkts
>>     
> |
>   
>>    |previously    | packets       |  and termination;| for termination
>>     
> |
>   
>>    |marked        |               |  do not re-mark  | meter all
>>     
> pkts;|
>   
>>    |packets       |               |  TM-marked pkts  | for admission,
>>     
> |
>   
>>    |              |               |                  | do not re-mark
>>     
> |
>   
>>    |              |               |                  | TM-marked pkts
>>     
> |
>   
> |-------------------------------------------------------------------|
>   
>> Regards,
>>
>>     Michael
>>
>> --
>> Dr. Michael Menth, Assistant Professor
>> University of Wuerzburg, Institute of Computer Science
>> Am Hubland, D-97074 Wuerzburg, Germany, room B206
>> phone: (+49)-931/888-6644, fax: (+49)-931/888-6632
>> mailto:menth@informatik.uni-wuerzburg.de
>> http://www3.informatik.uni-wuerzburg.de/research/ngn
>>
>> _______________________________________________
>> PCN mailing list
>> PCN@ietf.org
>> http://www.ietf.org/mailman/listinfo/pcn
>>     

-- 
Dr. Michael Menth, Assistant Professor
University of Wuerzburg, Institute of Computer Science
Am Hubland, D-97074 Wuerzburg, Germany, room B206
phone: (+49)-931/888-6644, fax: (+49)-931/888-6632
mailto:menth@informatik.uni-wuerzburg.de
http://www3.informatik.uni-wuerzburg.de/research/ngn

_______________________________________________
PCN mailing list
PCN@ietf.org
http://www.ietf.org/mailman/listinfo/pcn