On Sun, 14 May 2000, Joseph Mack wrote:
> On Fri, 12 May 2000, S Ashok Kumar wrote:
>
> > How many simultaneous ipvs sessions can
> > a P-III with 512MB RAM running RedHat 6.1 handle?
>
> Are you really interested in the number of sessions, or the amount
> of throughput that can be handled by LVS?
>
> The number of sessions is handled by user configurable variables
> in the kernel.
>
> The amount of throughput in VS-DR mode is limited by the network
> layer.
>
> In neither case is LVS limiting.
>
Hmmm... Thats definetely an overstatement and misleading.
Sure if you throw infinite processing power and RAM you can achieve
anything.
Overhead of the LVS code is definetly a contributor. Has anyone done
profiling on the LVS code?
To store the tables for thousands of connections means using up relatively
more RAM.
So yes, LVS code has everything to do with it.
cheers,
jamal
|