Franz Seiser

So you need to do overcapacity.

If you go to 10 and one or two fail, again, you do the same, but you need to build less.

So the more sites you have, the less money you need to invest in additional hardware to keep pace if you lose a certain amount of sites that you still can deal with the traffic.

So it's even making it cheaper.

to run your network in that setting.

And that's one of the big benefits of this whole architecture.

If a single instance failed, you will not recognize because the whole thing is done in a way that it automatically is taken over by others.

And here you go.

That's very easy.

The new one is 100%.

So in our new architecture, automation is part of the design.

We call it the zero, one and two.

So preparation, first rollout and regular operations.

The testing is automated and even the rollout, it's all completely automated.

So no manual intervention at any point of the system.

In the past, you do the critical maintenance you do at nighttime.

And if this poor guy starts to work at midnight and is doing configurations all night long, the probability that you do something wrong is increasing because you somehow get tired, right?

If this all is done fully automated and everything is prepared and checked, this is not going to happen.

Have you seen the roles that the people are doing have changed since automation has come in?

Yes, of course, of course.