Re: [RRG] thoughts on the design space 3: caching

[Jari Arkko <jari.arkko@piuha.net>]

Olivier,

> The only case where you would detect problems with mapping are when a 
> single host is opening a single TCP connection. In this case, the SYN 
> packet will be delayed by the mapping. Note that using the DNS also 
> causes such a delay...
>
> If EXPLISP needs to perform an evaluation, I would suggest to perform 
> the evaluation based on packet traces or netflow traces as we did and 
> not only on lab experiments with limited traffic.
Indeed, that would be the reasonable way to do it. I'm glad to learn 
that you've done some evaluations already! (Is there a chance that you 
could talk about some of your experiments in EXPLISP?)

I am having network trouble accessing the papers for some reason, but I 
wanted to point out two specific things that I'm worried about.

First, we often see a heavily tail in traffic distributions. While a 
large network concentrates much of its traffic to a small set of 
destinations (say, google, microsoft, and youtube), there is a very 
large set of other destinations as well that some small number of users 
are accessing. I'm pretty sure I'm the only one accessing certain 
domains in the IETF network, for instance.

Second, I'm not necessarily worried about the DNS lookup + SYN case, and 
I don't believe the impact on that is very significant. As was stated 
earlier, DNS already introduces some delay. But I'm more worried about 
situations such as applications that do round-robin or failover over a 
set of servers. The servers might come from DNS at that point, be 
resolved much earlier, or be communicated from someone else as IP 
addresses (such as would often be the case in P2P systems). What are the 
effects on those applications?

Jari


--
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