Comment 15 for bug 680763

On 24 March 2011 02:00, John A Meinel <email address hidden> wrote:
> @martin "initialize_on_transport_ex" times for me are:
> 0.0958 chinstrap
> 2.4302 lp
> 3.4193 lp://qastaging
>
> so the speed saved by the forking server is lost in the
> initialize_on_transport_ex

I think (unless my arithmetic was wrong) I saw qastaging being faster.
 But it's a smaller database server, though less loaded, so perhaps it
is reasonable to expect it be more variable. Anyhow, the next step is
in the other bug for someone to profile or look at the Launchpad code.