Hi Tony, > NPUs will likely have much larger contribution to PCPPS due to its higher power > consumption, so I see no clear reason why it would not have an effect.
I fully agree with this. Btw I said grouping will have an effect ... negative one to the end goal :) However that contribution can be carried as PCPPS of the NPU member links without introducing new grouping into link state flooding. Example: NPU consumes 100W and serves 4 active links with each laser of 10W. So you can advertise each link with the PCPPS metric of (directly mapping) 35. (25 as 100/4 + 10) How else would headend introduce Power Groups to the metric in CSPF ? Thx. R. On Thu, Sep 24, 2026 at 10:19 PM Tony Li <[email protected]> wrote: > > Hi Robert, > > > But you did not comment on the question how PCPPS would relate to RFC7810 > constraints if at all. > > > > As we have said repeatedly, PCPPS is only a modification to the metric. > CSPF still respects all constraints. > > > In respect to power groups you said: > > *Power groups allow vendors and operators to advertise the power > consumption of NPUs and other silicon components and the relationships > between those components so that external devices can make good decisions > about path computation.* > > Excellent I fully get the desire to shut NPU not just bunch of lasers. > But now you are really asking for ingress nodes to understand that per > node groupings in their path calculations. > > > > Which is exactly why this information is advertised in the IGP. > > > And here I must say that I share the concern of others that selecting the > same set of power groups (not only interfaces) by independent ingress nodes > may be a matter of luck then heuristics. And needless to say that needs to > be done bidirectionally which by itself argues against the nature of power > groups. > > Why ? > > > > It is neither a matter of luck nor heuristics. It is a matter of metric. > Larger metrics drive traffic away. PCPPS increases the metric on > interfaces where there are power-saving opportunities. If symmetrically > configured (which is NOT a given, due to asymmetric metrics because the two > ends of the link are not symmetric with respect to router architecture), > then the PCPPS addition would be symmetric and drive traffic to alternate > paths. > > > NPUs serving east ingress nodes (east-headends) may be part of completely > disjoined power groups from NPUs serving traffic coming from west ingress > nodes (west-headends). > > > > Correct. In fact, power-groups are node-local. > > > With staying on a per link granularity the likehood of success (moving > LSPs to other paths) IMO is much bigger then if adding to the mix the > concept of power groups (which again brings the question of how CSPF > actually is supposed to use them). > > > > Well, I completely disagree with you. NPUs will likely have much larger > contribution to PCPPS due to its higher power consumption, so I see no > clear reason why it would not have an effect. > > Cheers, > Tony > > >
_______________________________________________ Lsr mailing list -- [email protected] To unsubscribe send an email to [email protected]
