Many ISPs need the kinds of quality shaping cake can do
 help / color / mirror / Atom feed
* [LibreQoS] the k8 hairball
@ 2022-11-21  2:27 Dave Taht
  2022-11-21 13:43 ` MORTON JR., AL
  0 siblings, 1 reply; 3+ messages in thread
From: Dave Taht @ 2022-11-21  2:27 UTC (permalink / raw)
  To: libreqos

without reading this, I am willing to offer 3x1 odds this report does
not mention one specific tcp related metric.

https://humanitec.com/whitepapers/kubernetes-benchmarking-study-2022

Any takers?

-- 
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
Dave Täht CEO, TekLibre, LLC

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [LibreQoS] the k8 hairball
  2022-11-21  2:27 [LibreQoS] the k8 hairball Dave Taht
@ 2022-11-21 13:43 ` MORTON JR., AL
  2022-11-21 16:42   ` Dave Taht
  0 siblings, 1 reply; 3+ messages in thread
From: MORTON JR., AL @ 2022-11-21 13:43 UTC (permalink / raw)
  To: Dave Taht, libreqos

I'll take that bet, Dave, because I'm sure it mentions Throughput (and I only looked at part of the web page). It wouldn't be benchmarking without some definition of Throughput...

Full Disclosure: I have seen an earlier K8s benchmarking study, it could be related work. Among those of us who discussed the findings, one "knock" was that they only tested with TCP streams. Benchmarking for RFC 2544 Throughput requires UDP streams, no TCP flow-control complexities...

Al

> -----Original Message-----
> From: LibreQoS <libreqos-bounces@lists.bufferbloat.net> On Behalf Of Dave Taht
> via LibreQoS
> Sent: Sunday, November 20, 2022 9:28 PM
> To: libreqos <libreqos@lists.bufferbloat.net>
> Subject: [LibreQoS] the k8 hairball
> 
> without reading this, I am willing to offer 3x1 odds this report does
> not mention one specific tcp related metric.
> 
> https://urldefense.com/v3/__https://humanitec.com/whitepapers/kubernetes-
> benchmarking-study-2022__;!!BhdT!gy_08-Lw3wEiAgdspSMpcAGcLbfMq-my_-
> YYi47om7KKd1gPKSdrT6L3o95U3sYPPpciDoMZMouCUh54K8qkmxM$
> 
> Any takers?
> 
> --
> This song goes out to all the folk that thought Stadia would work:
> https://urldefense.com/v3/__https://www.linkedin.com/posts/dtaht_the-mushroom-
> song-activity-6981366665607352320-FXtz__;!!BhdT!gy_08-Lw3wEiAgdspSMpcAGcLbfMq-
> my_-YYi47om7KKd1gPKSdrT6L3o95U3sYPPpciDoMZMouCUh54RS9Rq-c$
> Dave Täht CEO, TekLibre, LLC
> _______________________________________________
> LibreQoS mailing list
> LibreQoS@lists.bufferbloat.net
> https://urldefense.com/v3/__https://lists.bufferbloat.net/listinfo/libreqos__;
> !!BhdT!gy_08-Lw3wEiAgdspSMpcAGcLbfMq-my_-
> YYi47om7KKd1gPKSdrT6L3o95U3sYPPpciDoMZMouCUh54XdGMIEA$

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [LibreQoS] the k8 hairball
  2022-11-21 13:43 ` MORTON JR., AL
@ 2022-11-21 16:42   ` Dave Taht
  0 siblings, 0 replies; 3+ messages in thread
From: Dave Taht @ 2022-11-21 16:42 UTC (permalink / raw)
  To: MORTON JR., AL; +Cc: libreqos

On Mon, Nov 21, 2022 at 5:44 AM MORTON JR., AL <acmorton@att.com> wrote:
>
> I'll take that bet, Dave, because I'm sure it mentions Throughput (and I only looked at part of the web page). It wouldn't be benchmarking without some definition of Throughput...

Nope. Not mentioned once. I shoulda hung an amount on that bet. Also,
I was overly vague, I'd meant latency or latency under load...
I have no idea why the k8 crowd thinks the network stack isn't
something to worry about...

>
> Full Disclosure: I have seen an earlier K8s benchmarking study, it could be related work. Among those of us who discussed the findings, one "knock" was that they only tested with TCP streams. Benchmarking for RFC 2544 Throughput requires UDP streams, no TCP flow-control complexities...
>
> Al
>
> > -----Original Message-----
> > From: LibreQoS <libreqos-bounces@lists.bufferbloat.net> On Behalf Of Dave Taht
> > via LibreQoS
> > Sent: Sunday, November 20, 2022 9:28 PM
> > To: libreqos <libreqos@lists.bufferbloat.net>
> > Subject: [LibreQoS] the k8 hairball
> >
> > without reading this, I am willing to offer 3x1 odds this report does
> > not mention one specific tcp related metric.
> >
> > https://urldefense.com/v3/__https://humanitec.com/whitepapers/kubernetes-
> > benchmarking-study-2022__;!!BhdT!gy_08-Lw3wEiAgdspSMpcAGcLbfMq-my_-
> > YYi47om7KKd1gPKSdrT6L3o95U3sYPPpciDoMZMouCUh54K8qkmxM$
> >
> > Any takers?
> >
> > --
> > This song goes out to all the folk that thought Stadia would work:
> > https://urldefense.com/v3/__https://www.linkedin.com/posts/dtaht_the-mushroom-
> > song-activity-6981366665607352320-FXtz__;!!BhdT!gy_08-Lw3wEiAgdspSMpcAGcLbfMq-
> > my_-YYi47om7KKd1gPKSdrT6L3o95U3sYPPpciDoMZMouCUh54RS9Rq-c$
> > Dave Täht CEO, TekLibre, LLC
> > _______________________________________________
> > LibreQoS mailing list
> > LibreQoS@lists.bufferbloat.net
> > https://urldefense.com/v3/__https://lists.bufferbloat.net/listinfo/libreqos__;
> > !!BhdT!gy_08-Lw3wEiAgdspSMpcAGcLbfMq-my_-
> > YYi47om7KKd1gPKSdrT6L3o95U3sYPPpciDoMZMouCUh54XdGMIEA$



-- 
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
Dave Täht CEO, TekLibre, LLC

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2022-11-21 16:42 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-11-21  2:27 [LibreQoS] the k8 hairball Dave Taht
2022-11-21 13:43 ` MORTON JR., AL
2022-11-21 16:42   ` Dave Taht

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox