Using an HTTP(S) load balancer

Now we are getting to the final stage of our rehosting architecture changes. The HTTP(S) load balancer brings it all together. We have created our regional managed instance group from our instance template and used the existing configuration and drive from our Tomcat virtual machine to create that instance template. For the regional managed instance group, we accepted the defaults on which zones it uses (all zones in the region by default), and the default on the autoscale rules, but specified the minimum number of virtual machines as three. Remember the rule of thumb, N + 2. To ensure we have at least one server running, we need to have a target of three, so that one could go down for maintenance and one could go down due to failure, but one would still be running.

So, what is left to do? Well, we need to create an HTTP(S) load balancer, defining the backend configuration, host and path rules, and frontend configuration. The backend configuration...