[Rpm] Test procedures for TCP stacks ?

[Rpm] Test procedures for TCP stacks ?

[Toerless Eckert]

Are thre any well defined (RFC or the like) test procedures for the "quality"
of TCP stacks ? I am especially thinking about measuring how good the
TCP stack behaves in the face of various path problems. Latency, loss,
reorder,.

Related:

How big a difference between goodput and throughput would one be able to
see anyhow between worst... best TCP stacks ?

Thanks!
    Toerless

Re: [Rpm] Test procedures for TCP stacks ?

[Christoph Paasch]

Hello,

> On Jan 31, 2022, at 7:27 AM, Toerless Eckert via Rpm <rpm@lists.bufferbloat.net> wrote:
> 
> Are thre any well defined (RFC or the like) test procedures for the "quality"
> of TCP stacks ? I am especially thinking about measuring how good the
> TCP stack behaves in the face of various path problems. Latency, loss,
> reorder,.

I don't think that there is anything like that. All depends on what would be defined as "quality".

> 
> Related:
> 
> How big a difference between goodput and throughput would one be able to
> see anyhow between worst... best TCP stacks ?

That is interesting. It would be a measure of the efficiency of the TCP stack.


Christoph

> 
> Thanks!
>    Toerless
> _______________________________________________
> Rpm mailing list
> Rpm@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/rpm

Re: [Rpm] Test procedures for TCP stacks ?

[Dave Taht]

"Quality" is an attribute I think about based on Deming's work on TQM,
and Zen and the Art of Motorcycle Maintenance.

You know it when you see it, feel it, taste it, hold its heft in your
hand, and wheel your bike out of the garage after a long winter and it
starts right up.



On Mon, Jan 31, 2022 at 7:27 AM Toerless Eckert via Rpm
<rpm@lists.bufferbloat.net> wrote:
>
> Are thre any well defined (RFC or the like) test procedures for the "quality"
> of TCP stacks ? I am especially thinking about measuring how good the
> TCP stack behaves in the face of various path problems. Latency, loss,
> reorder,.
>
> Related:
>
> How big a difference between goodput and throughput would one be able to
> see anyhow between worst... best TCP stacks ?
>
> Thanks!
>     Toerless
> _______________________________________________
> Rpm mailing list
> Rpm@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/rpm



--
I tried to build a better future, a few times:
https://wayforward.archive.org/?site=https%3A%2F%2Fwww.icei.org

Dave Täht CEO, TekLibre, LLC