Jim Gettys' bufferbloat paper

[Sameer Verma]

On Sat, Dec 3, 2011 at 1:44 PM, Rafael Ortiz <rafael at activitycentral.com> wrote:
>
>
> On Sat, Dec 3, 2011 at 4:36 PM, Sameer Verma <sverma at sfsu.edu> wrote:
>>
>> Jim's paper is out. I'm sure some of you have already seen it, but
>> here it is: http://queue.acm.org/detail.cfm?id=2071893
>>
>> cheers,
>> Sameer
>
>
> Must read!.

I had to explain this to someone who doesn't understand bits and TCP
and such. Here's what I came up with.

I tend to think of the issue akin to pouring a liquid out of a bottle
(oil, soy sauce, coca cola). If you simply tilt the bottle quickly,
the liquid will pour out (bits go out), and then create a vacuum
inside the bottle, which stops the flow, and creates a gap in the neck
to let some air back in (like ACK). Then the liquid flows again. What
you get is a stop and go flow, very much like what bufferbloat
suggests.

Now, tilt the bottle slowly, so that we have a small opening for the
air to keep flowing back into the bottle at a steady pace, and the
liquid will flow out steadily as well. So make buffers smaller, and
the flow will be steadier.

Not quite an exact analogy, but close enough :-)

Sameer

>
>
>>
>> --
>> Sameer Verma, Ph.D.
>> Professor, Information Systems
>> San Francisco State University
>> http://verma.sfsu.edu/
>> http://olpcsf.org/
>> _______________________________________________
>> Devel mailing list
>> Devel at lists.laptop.org
>> http://lists.laptop.org/listinfo/devel
>
>