LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Re: Release new code: Scheduler for distributed caching

Subject: Re: Release new code: Scheduler for distributed caching
Cc: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
From: Thomas Proell <Thomas.Proell@xxxxxxxxxx>
Date: Wed, 25 Oct 2000 12:17:32 +0200 (MET DST)
Hi!

> examination.  (Though a hash based netfilter module possibly could
> answer the same needs.  But someone would have to code it.)

Don't look at me that way! :-)
You know, I was not very familiar with ipchains, netfilter and all
that stuff. So, I could not think about a way to TUNNEL the packets
without LVS. How should that work?

So, say, it works. What advantages would there be to implement it in
netfilter? The advantage of "not having to track the routes", right?
If this is the only one, you may forget it, since I think it's
less effective to calculate the hash for each packet.

Moreover, if I implement the hot spot solution, I will have to 
take care of the problem (where I sent this connection before) again.

So, the impact of this (if any) is very little.

Any other advantages of netfilter?



Thomas



<Prev in Thread] Current Thread [Next in Thread>