|Thread Links||Date Links|
|Thread Prev||Thread Next||Thread Index||Date Prev||Date Next||Date Index|
On the ring, whether a packet is store and forward or cut through does affect the delay/jitter
performance. Especially the ring is a frame based with variable frame size. The transit path is
also the shared medium so it affects every packet in flight.
I beg to differ that it is a trade-off between the ingress path or the transit path.
From: Steven Wood [mailto:swood@xxxxxxxxx]
Sent: Thursday, March 22, 2001 10:05 AM
To: Ray Zeisz
Subject: Re: [RPRWG] MAC Question
This is a design choice and is currently the subject of much
debate within RPR. My view is that the latency/jitter issues
will be there in either a store & fwd or cut-through design.
It is simply a question of where you introduce the jitter,
on the ingress path or the transit path.
In the next couple of meetings, I'm sure we will be seeing
some concrete results to compare the two approaches.
At 09:02 AM 3/22/01 , Ray Zeisz wrote:
>I am following the .17 group from afar, but I have a question:
>Is it acceptable for each node in the ring to buffer up an entire packet
>before forwarding it to its neighbor? Would the latency be to great if this
>were done? Or is the .17 direction more along the lines of 802.5 where only
>a few bits in each ring node are buffered...just enough to detect a token
>and set a bit to claim it.
Manager, Hardware Engineering
Cisco Systems Co.
365 March Rd.
CANADA K2K 2C9 | |
Phn: (613) 271-3450 :|: :|:
Fax: (613) 271-3333 :|||: :|||:
http://www.cisco.com C i s c o S y s t e m s