14 Jan
2016
14 Jan
'16
9:36 p.m.
On 2016-01-14 21:29, Paul_Koning at Dell.com wrote:
Yes. But I think that Peter is actually advocating that cost for
ethernet should be larger than cost for Multinet link at all nodes, in
order to most of the time decide to favor the multinet links.
Ie. Not have a situation where node a uses the multinet link to talk to
node b, and b uses ethernet (bridge) to send responses back to a.
DECnet works just fine with such different paths, though. So it's just a
question of - will the multinet links give better performance than
ethernet and the bridge?
Both ends could have equal or different costs for the links. As long as
the relative order of costs between the links are the same, they should
then both favor the multinet link. However, it can become more
interesting when there are multiple paths, and multiple hops, as the
costs are the sum of the cost of all hops, meaning that potentially,
what cost other people put on their links can affect how your forwarding
matrix looks like.
Johnny
On Jan 14, 2016, at 3:02 PM, Peter Lothberg
<roll at Stupi.SE> wrote:
If costs are the same at both ends of a link, that will certainly help. Then again, it
is quite possible for two paths to have equal cost, and if so, DECnet implementations will
pick one of the two, in a way that is not specified. The values are somewhat arbitrary; it doesn't
really matter what
scheme you use but if you are inconsistent the routing may be
surprising.
The routing spec has a suggested algorithm (100,000/line speed)
which may have made sense in the old days but for modern networks
isn't terribly useful.
paul
What I wanted to get to was a scenario where traffic was symetric
between two nodes, eg, use the same links from a-b as b-a, it makes it
much easier to understand what's wrong when things behave funny...