Re: Rogers outage - House of Commons INDU hearing

Let's see how this goes. A political hearing will (IMO) likely not resolve in the identification of core technical issues nor meaningful changes to avoid what was experienced on July 8th. A third party conducting a real technical investigation is likely the way to find a RCA and set of

Re: HE.net and BGP Communities

I think the key difference here is that I really just wanted HE to treat my routes no differently no matter how they learned them. I wanted them to apply normal BGP routing rules to them.. that is, pick the path with the shortest AS path. >From a strictly technical basis, its silly to prefer a

Rogers outage - House of Commons INDU hearing

Canada is part of North America, and may be of interest to NANOG. Rogers had a nationwide outage on July 8, 2022. The House of Commons held a hearing today (July 25, 2022) on the outage and invited the Rogers CEO, other company CEOs, regulators and academics to testify. Recording of the

Re: 400G forwarding - how does it work?

> > It wasn't a CPU analysis because switching ASICs != CPUs. > > I am aware of the x86 architecture, but know little of network ASICs, > so I was deliberately trying to not apply my x86 knowledge here, in > case it sent me down the wrong path. You made references towards > typical CPU features; >

Re: 400G forwarding - how does it work?

On Mon, Jul 25, 2022 at 11:58 AM James Bensley wrote: > On Mon, 25 Jul 2022 at 15:34, Lawrence Wobker wrote: > > This is the parallelism part. I can take multiple instances of these > memory/logic pipelines, and run them in parallel to increase the throughput. > ... > > I work on/with a chip

RE: HE.net and BGP Communities

> > I do understand the reasoning behind preferring customer routes. > However in the case where a customer of a customer also connects to > you directly via peering doesn't it make sense to prefer the direct > connection? or at least not prefer the customer learned routes. So from my

Re: HE.net and BGP Communities

most networks prefer customers over peers. a few networks don't, some large. if they changed, customers would scream at them about the newly overloaded customer circuit. we are not required to like this :) randy

Re: 400G forwarding - how does it work?

On Mon, 25 Jul 2022 at 21:51, James Bensley wrote: > I have no frame of reference here, but in comparison to Gen 6 Trio of > NP5, that seems very high to me (to the point where I assume I am > wrong). No you are right, FP has much much more PPEs than Trio. For fair calculation, you compare how

Re: 400G forwarding - how does it work?

Hi Lawrence, thanks for your response. On Mon, 25 Jul 2022 at 15:34, Lawrence Wobker wrote: > This is the parallelism part. I can take multiple instances of these > memory/logic pipelines, and run them in parallel to increase the throughput. ... > I work on/with a chip that can forwarding

Re: 400G forwarding - how does it work?

Thanks for the responses Chris, Saku… On Mon, 25 Jul 2022 at 15:17, Chris Adams wrote: > > Once upon a time, James Bensley said: > > The obvious answer is that it's not magic and my understanding is > > fundamentally flawed, so please enlighten me. > > So I can't answer to your specific

Re: HE.net and BGP Communities

On Mon, 25 Jul 2022 at 16:23, Forrest Christian (List Account) wrote: > > I do understand the reasoning behind preferring customer routes. > However in the case where a customer of a customer also > connects to you directly via peering doesn't it make sense to prefer > the direct connection? or

RE: HE.net and BGP Communities

> Google has let you down, 6730 (sunrise) is not 6939 (HE) ;) > Oops I should check thing better at the end of the day .. My bad

Re: HE.net and BGP Communities

On Mon, 25 Jul 2022 at 17:19, Brian Turnbow via NANOG wrote: > > Hi, > > >I do understand the reasoning behind preferring customer routes. However > >in the case where a customer of a customer also connects to you directly via > >peering doesn't it make sense to prefer the direct connection?

Re: 400G forwarding - how does it work?

So the most important bits are pipelining and parallelism. And this is substantially simplified, but hopefully it helps. Pipelining basically means that you have a whole bunch of different operations that you need to perform to forward a packet. Lots of these are lookups into things like the

RE: HE.net and BGP Communities

Hi, >I do understand the reasoning behind preferring customer routes.   However in >the case where a customer of a customer also connects to you directly via >peering doesn't it make sense to prefer the direct connection?  or at >least >not prefer the customer learned routes. Business not

Re: HE.net and BGP Communities

> Am 7/25/22 um 12:45 schrieb Forrest Christian (List Account) > : > >  > I wish they'd add one more that turns off their "prefer routes learned from a > customer" rule. I'm having to split my blocks in half and announce them > that way to get them to send my traffic directly to me

Re: HE.net and BGP Communities

I do understand the reasoning behind preferring customer routes. However in the case where a customer of a customer also connects to you directly via peering doesn't it make sense to prefer the direct connection? or at least not prefer the customer learned routes. In our situation, we were

Re: HE.net and BGP Communities

>I wish they'd add one more that turns off their "prefer routes learned from a >customer" rule. I'm having to split my blocks in >half and announce them >that way to get them to send my traffic directly to me through our IX peering >session as opposed to >one of my transit providers. >I'd

Re: 400G forwarding - how does it work?

I'm not sure what your specific question is. So I answer my question instead. Q: how can we do lookup fast enough to do 'big number' per second, while underlying hardware inherently takes longer A: we throw memory at the problem I.e. say JNPR Trio PPE has many threads, and only one thread is

Re: 400G forwarding - how does it work?

Once upon a time, James Bensley said: > The obvious answer is that it's not magic and my understanding is > fundamentally flawed, so please enlighten me. So I can't answer to your specific question, but I just wanted to say that your CPU analysis is simplistic and doesn't really match how CPUs

400G forwarding - how does it work?

Hi All, I've been trying to understand how forwarding at 400G is possible, specifically in this example, in relation to the Broadcom J2 chips, but I don't the mystery is anything specific to them... According to the Broadcom Jericho2 BCM88690 data sheet it provides 4.8Tbps of traffic processing

Re: HE.net and BGP Communities

You always want to prefer customer routes over non customer routes as a service provider. Of course having a robust set of communities to let adjustments happen helps. Without proper tiering of routes you may see unstable routing. Having a standard set of customer, peer, transit set of local

Re: HE.net and BGP Communities

I wish they'd add one more that turns off their "prefer routes learned from a customer" rule. I'm having to split my blocks in half and announce them that way to get them to send my traffic directly to me through our IX peering session as opposed to one of my transit providers. I'd rather they

Re: Request for help: academic study (questionnaire)

***Apologies for cross-posting*** Interim results (available until 26th July, 10 am CET) Link to fill in survey (for those who have been unable to do so and might be