RE: IPv6 Performance (was Re: IPv4 Pricing)

[Vasilenko Eduard via NANOG <nanog () lists nanog org>]

I completely agree that the list of reasons "why RTT of IPvX is faster/slower then IPvY" is very long. I have already 
mentioned a few, you have mentioned a few additional.
But I have never seen a good research which one has the biggest influence. It is a still a pure speculation from all 
sides.
In this situation, my preference is to think that new/recently_upgraded network has a better service and occasionally 
it has been pushed to IPv6 at the same time.
And sorry, but NO - I do not have resources to conduct such a research. I would be interested to read such research, 
not write.

> You said that so many times, it is boring now and that impact doesn't have enough weight to change the average 
> latency in a way that IPv6 is slower in general.
In the great majority of cases, it was said "IPv6 is faster" without clarification that it is for RTT that does not 
matter. The user is misled that it is for FCT that he/she needs.
People said non-important advantage about 10^6 times - you are not boring with this. I said 3 times about important 
disadvantage - you are already boring.

You did touch ASIC processing. Actually, it is not important because it would be pretty fast anyway (X us).
What is important that IPv6 architecture has 2x bigger meaningful address part, hence, the scalability of tables (for 
routing, filtering) is 2x less for all vendors and all products.
Why it is 2x not 4x? Because the second half of addresses is not the address, hence, it is typically discarded from 
such tables.
IMHO: this IPv6 deficiency is justified too, because the world needs addresses of the bigger size.

As you see, IPv6 has many deficiencies against IPv4: more overhead, less scale in memory. But most of them are 
justified.
Eduard
> -----Original Message-----
> From: Marco Moock via NANOG <nanog () lists nanog org>
> Sent: Thursday, December 4, 2025 10:37
> To: nanog () lists nanog org
> Cc: Marco Moock <mm () dorfdsl de>
> Subject: Re: IPv6 Performance (was Re: IPv4 Pricing)
>
> Am 04.12.2025 um 06:22:19 Uhr schrieb Vasilenko Eduard via NANOG:
>
> > Yes, indeed, https://stats.labs.apnic.net/v6perf shows -6.1ms
> > globally. Highly probably, Geoff did everything right, we could trust
> > the numbers. But sorry, there is no explanation why?
>
> There are various reasons:
>
> No NAT/CGNAT
> No DS-Lite tunneling (less MTU and more CPU cycles on the CPE chips) Smaller
> routing tables Different peering/routing inside AS
>
> If you really want the proof, you have to choose one ASN and examine all the
> devices and check the latencies there.
> There is no "this is the general reason for all". Some routers also handle
> SPI/NAT/routing different, especially when longer prefixes are used.
> I heard (never verified) that certain devices only handle IPv4 /24 and
> IPv6 /48 fast, the rest will be processed slower due to ASIC chip design.
>
> > At least, I have never seen anything that could be called "prove".
> > Your speculation is the same reliable as mine. Actually, mine (about
> > new networks -> high quality) looks more probable for me.
>
> Explain the term "new". Various ISPs exist for decades and I have serious
> doubt that old devices are still present there. Even entire network
> architectures were outphased, e.g. ATM, ISDN and old DSL infrastructure.
>
> > 2.
> > RTT does not matter. Not at all.
>
> That is indeed right. I saw situations where the ICMP messages of a router
> more hops away arrived faster than one from a hop closer. A reason is also
> that this stuff is most likely not processed in the ASIC based forwarding plane,
> but in a slower control plane. If you want to know more about that, you have
> to ask the router's manufacturers.
>
> > Because it is not visible for the
> > end user directly and it has negligible influence for the good
> > Congestion Control (like BBR). User cares only about the FCT (flow
> > completion time). FCT is dependent on (in priority): 1) bottleneck
> > bandwidth, 2) packet loss, 3) congestion control convergence, 4) RTT
> > (but only if congestion control is bad), and a few other things.
> > Subtracting 2.6(6)% of the goodput (because of bigger overhead) make
> > IPv6 fundamentally slower (for the all other things equal).
>
> You said that so many times, it is boring now and that impact doesn't have
> enough weight to change the average latency in a way that IPv6 is slower in
> general.
>
> --
> Gruß
> Marco
>
> Send unsolicited bulk mail to 1764825739muell () cartoonies org
_______________________________________________
NANOG mailing list 
https://lists.nanog.org/archives/list/nanog () lists nanog org/message/EQ4NGAMUG6NUZR6R4HCKMKQ5SHXF7YPD/