(재전송) RE: End to End QoS across the Internet

[jinoo7 at samsung.co.kr]

Sean wrote:

>While EPD is a clever hacks to deal with cell trains,
>I am sceptical of its scalability in size-large ATM switches
>where multiple sources are feeding a single egress interface
>that cannot transmit as fast as the offered load.

EPD and PPD, and all the congestion management schemes are aiming
to resolve such problems. By the way, doesn't 'congestion' occur
when incoming traffic overloads the switch's capacity?  


>PPD is the wrong model -- discarding the middle (or front)
>part of a cell train usually adds to congestion (this is
>called partial-delivery congestion collapse).  EPD does the
>correct thing by restricting the admission of an entire
>cell train, thus assuring that there is no partial delivery.

The point of PPD is to reduce the complexity of implementing
EPD, and yes, it still works fine. If you happen to drop an ATM cell,
by congestion, there is no meaning of transmitting the remainder,
because TCP won't accept the collapsed packet.  


EPD trys to aviod such a situation by dropping 'packets' from  

the beginning when the buffer is expected to become full therefore
there will be cells dropped otherwise.

PPD doesn't drop the whole packet. Instead, it drops the 'remainder'
-- not the middle nor the front---
of packets which have already 'currupted' by losing cells.  

What do we do about the front of corrupted packet? Nothing.  

And that's it.

Thanks,
Jinoo