Re: [RRG] ALT's strong aggregation often leads to *very* long paths

[jnc@mercury.lcs.mit.edu (Noel Chiappa)]

    > From: ksriram@nist.gov

    > query/response messages and first-few (or perhaps first-several) "data
    > packets will be sent (inefficiently) along the server hierarchy".
    > ...
    > Modeling needs to be done to quantify the delays due to potentially
    > very long paths along the ALT-rtr hierarchy and their performance
    > impacts. Meanwhile, I think it is worth exploring some enhancements
    > that can potentially improve performance under these circumstances.

Yes, but that behaviour (forwarding the first/first-few packets along the
resolution hierarchy) in itself is an optimization (see Brian's previous
message) affecting only a small share of user data packets.

Do you really want to try and optimize an optimization, and only for that
small share of traffic?


    > Can we not make an enhancement to the way the query/response messages
    > and first data packets are handled in ALT routers to take advantage of
    > any available shorter paths (bypassing the hierarchy as and when
    > feasible)?

I can't respond about ALT, but adding these kind of short-cuts in the
resolution hierarchy was something that was considered at some length for
CONS, in order to speed up resolution (i.e. reduce resolution delay), and
also avoid a load concentration at the root.

A number of other optimizations which had the same effect (reducing
resolution delay) were developed, and since this particular one i) added a
fair amount of complexity, and ii) didn't buy that much with the other
optimizations in place, it was dropped.

Now that it is proposed to use the resolution hierarchy to forward packets
for which the mapping is missing, I think it might be worth taking another
look at that optimization, but it would only be to avoid creating a potential
load hot-spot at the root, and not for any slight increase in perfomance for
that small share of the traffic which used it.

	Noel

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg