I was thinking along these lines as well. Possibly they have other checks and balances in place for less than death situations for servers like high CPU usage

I was thinking along these lines as well. Possibly they have other checks and balances in place for less than death situations for servers like high CPU usage and those situations result in another mechanism killing the server which is the same as what this solution provides.