[Iccrg] What's wrong with TCP Congestion control

[Radek Krzywania]

On Tue, 07 Feb 2006 16:08:38 +0100, Phelan, Tom <tphelan at sonusnet.com>  
wrote:

> Hi All,
>
> I'd like to point out that the large buffer requirement is not an
> intrinsic property of window-based congestion control.  It's the result
> of the requirement for reliability, coupled with the large transmit
> window.  If you don't need reliability (as with DCCP and its AIMD
> congestion control), the need for sender buffering is greatly reduced.

Hi,
I fully agree that it is the side effect of large window. But in sense of  
TCP, where reliability is a requirement, we need to allocate large memory  
buffers. So paraphrasing my first point:

"Maybe it is good to say explicitly, that large windows requires large
memory buffers on sender side for reliability purposes."

I am talking about memory, because if we have e.g. grid computing  
machines, which communicates each other with multiple streams over long  
fast links, the so-expensive memory may be wasted on TCP windows, rather  
than computations. Am I very wrong here?

Best regards
Radek

-- 
______________________________________________________________________
Radoslaw Krzywania       Network Research and Development
sfrog at man.poznan.pl      Poznan Supercomputing and Networking Center
+48 61 858 20 28         http://www.man.poznan.pl
______________________________________________________________________