|
On Mon, 10 Dec 2001, Julian Anastasov wrote:
> > Radu-Adrian's mail (about Tiscali's LVS) was saying that they mesured the
> > load and came
> > to a "formula" that looked like this (For PIII @ 866MHz):
> >
> > CPU_Load[%] = 0.92 * Traffic[Mbps]
>
> I'm not sure CPU_Load is valid parameter. May be only if you
> have one process eating your CPU while the packets interrupt it.
That should read CPU_time, and is not in pure load-balancing conditions. Using
pure LVS/NAT (no firewall, no traffic control, 1500 bytes packets, 3Com card)
I reached 100 Mbps on a PIII/550 with 35-40 % CPU usage in system.
Adding masquerade, packet filtering, traffic control, smaller packet size
and a "not so happy NIC-switch combination" leads to a performance disaster
Radu-Adrian Feurdean
mailto: raf (a) chez.com
----------------------------------------------------------------------------------
"Russian components, american components..all are made in Taiwan" Armaghedon,
1999
|
