Hi, Steven, > I don't think your conclusion follows. One thing you want for LAG/ECMP is > for each flow from a given <src_addr, dst_addr> to have a unique FL value. > Fernando's algorithm achieves this by incrementing counter for each new > flow from that address pair. > > With that said, I don't think this algorithm is necessarily ideal. The FL > value for any two flows from a <src_addr, dst_addr> pair may only vary by a > few bits, so if a switch/router uses a poor hash algorithm for LAG/ECMP, > you may not get a good load spread.
I think that in this case both the hash algorithm used for generating the Flow Label and the hash algorithm used at the router would need to be be "poor". Nevertheless, if the hash algorithm at the router is poor, you're dead anyway. Because different hosts could use different algorithms for generating the Flow Label which on the aggregate could result in Flow Label values that differ in only a few bits. Unless I'm missing something.... Thanks! Kind regards, -- Fernando Gont e-mail: ferna...@gont.com.ar || fg...@acm.org PGP Fingerprint: 7809 84F5 322E 45C7 F1C9 3945 96EE A9EF D076 FFF1 -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
