Re: [RPRWG] RPR over wireless links
Geir
About the fairness part:
From simulations and some analytical work I have done. It seems like
the aggressive mode of fairness operation should allow for a maximum
total delay (sum of queueing, transmission, propagation) of < 150 ms
between head and tail in a fairness domain (with lpCoef set to 512).
The analysis covers only convergence of the rate calculation to the fair
rate and does not include analysis of the delay properties for the
different traffic classes.
I have not analyzed the conservative mode, but here, the time between
rate adjustments is calculated dynamically and a rate adjustment is
performed only when the effect of the previous can be observed by the
head node...
Topology and sub 50ms protection time:
This is a separate issue that others may probably provide more
insightful comments to...
Regards,
Fredrik
goy@xxxxxxxxxxxxxxxx writes:
> Talking to our modem people the PHY delay in a wireless hop is up to 100
> mikro seconds/hop.
> This includes SDH, modem and coding/decoding delay. Propagation delay is
> not included.
> Typically wireless ring networks have 5-10 hops and hoplengths around 5 -
> 15 km.
> This means that the physical layer delay will be the most dominant part in
> these networks.
>
> So my wonder is if the RPR functionality (fairness and protection) can
> still be conserved with
> link rates of 155 Mbits/s in the above mentioned ring structure?
> How much RTT delay is acceptable for the RPR MAC protocol?
>
> Thanks Geir Åge Øye
>
>
>
>
>
> "Castellano,
> Robert"
> <RCastellano@c-co To
> r.net> "'Mike Takefman'" <tak@xxxxxxxxx>,
> goy@xxxxxxxxxxxxxxxx
> 15.01.2004 19:40 cc
> stds-802-17@xxxxxxxx
> Subject
> RE: [RPRWG] RPR over wireless links
>
>
>
>
>
>
>
>
>
>
> Hi,
>
>
> It would be nice to know what the worst case link delays or ring RTT could
> be supported by the fairness algorithm
> at 155Mbps.
>
>
> thanks,
>
>
> robert
>
>
>> -----Original Message-----
>> From: Mike Takefman [mailto:tak@xxxxxxxxx]
>> Sent: Thursday, January 15, 2004 10:18 AM
>> To: goy@xxxxxxxxxxxxxxxx
>> Cc: stds-802-17@xxxxxxxx
>> Subject: Re: [RPRWG] RPR over wireless links
>>
>>
>>
>> Geir,
>>
>> I have not really heard of anyone planning RPR over Wireless
>> but in theory there is no reason why it could not work.
>>
>> Right now, the slowest SONET PHY that RPR supports is
>> 155 Mb/s. Which fits into the range you were looking at.
>>
>> However, the design of the MAC was assuming
>> fiber media, which had a very low bit error rate. I am
>> aware that SONET radio links do exist and often include
>> FEC coding schemes to decrease the error performance
>> of the links.
>>
>> As to the additional link delays that could occur due
>> to coding schemes, there are simulation models for
>> RPR that could be modified to include such effects.
>> If the delays were large enough, there could be
>> issues with fairness.
>>
>> I cannot comment on Silicon Providers plans / products.
>>
>> cheers,
>>
>> mike
>>
>> goy@xxxxxxxxxxxxxxxx wrote:
>> >
>> >
>> >
>> >
>> > Hello
>> > Currently I'm study the applicability of using the RPR standard and
>> > functionality over
>> > wireless ring structures, or as a combination of fiber links.
>> > This means that the typical data rates would be from STM-1
>> - 4*STM-1.
>> > Even data rates as low as 100 Mbit/s could be interesting
>> in the entreprice
>> > access market.
>> >
>> > So the I would like to ask the subscribers if anyone has planns or
>> > any other comments on using the RPR over a wireless media
>> and with such low
>> > data rates?
>> >
>> > Question of interest:
>> > Could wireless link quality be any problem?
>> > Loss rate and faiding charcteristics?
>> > Delay/hop requierement for low capacity links to maintain the RPR
>> > functionality?
>> > Silicon providers for low capacity?
>> >
>> >
>> > Geir Åge Øye
>> > Nera Research
>> > Research Scientist
>> > Phd Applied Mathemathics
>> > Phone +47 55225559
>> > Fax +47 55225299
>> > goy@xxxxxxxxxxxxxxxx
>> > (Embedded image moved to file: pic21003.gif)
>> >
>> >
>> >
>> --------------------------------------------------------------
>> ----------
>> >
>>
>>
>> --
>> Michael Takefman tak@xxxxxxxxx
>> Distinguished Engineer, Cisco Systems
>> Chair IEEE 802.17 Stds WG
>> 3000 Innovation Dr, Ottawa, Canada, K2K 3E8
>> voice: 613-254-3399 cell:613-220-6991