Sent from my Samsung Galaxy smartphone.
___
AusNOG mailing list
AusNOG@lists.ausnog.net
http://lists.ausnog.net/mailman/listinfo/ausnog
Overall, the article being discussed only does very precursory testing.
In my recent adventures of deploying OSS routing for mission critical services,
there has to be a mix of raw performance, stability and performance with varied
feature-sets (NAT, QoS, Load-Balancing etc). The article above l
>> "if performance matters (and it does very much so), why would you be
using *_anything_* virtualised at all..."
Because it's not actually possible to write meaningful SLAs for time
multiplexed services.
At the end of the day I agree with Brad, if you need a performant system
you want resident h
Cisco TRex does a pretty decent job of generating ridiculous amounts of traffic
if you have a supported NIC, and it’s open source.
Sent from my iPhone
> On Oct 4, 2019, at 8:57 PM, Rob Thomas wrote:
>
> Well, that's kinda the point of actually doing performance tests! 8)
>
> If you didn't re
Well, that's kinda the point of actually doing performance tests! 8)
If you didn't read the article, using 4 cores of a 7 year old CPU can
route 12gbit of traffic at 13% cpu (or 17gbit on a 4 year old CPU), I
think the actual data speaks for itself.
Unfortunately, after 20gbit, it gets hard to GE
Because the amount of CPU available on virtualised platforms is more than
sufficient for most people's needs? Because the numerous advantages of
virtualisation far outweigh the (perceived) performance penalties? Because
sometimes you don't actually have a choice?
Like any good tool, virtualisation
if performance matters (and it does very much so), why would you be
using _ANYTHING_ virtualised at all...
On 03/10/2019 23:19, Guy Ellis wrote:
> Has anyone bothered to evaluate TNSR which I will think replace pfsense where
> performance really matters?
--
Kind Regards,
Noel Butler