Disconnected operations
In this section, we will cover unique options and opportunities for not only planning for interrupted network connections but actively planning for those scenarios as an operating requirement. You will learn how some organizations anticipate and respond to those situations in three separate approaches.
In the Delving into edge-in architectures section of Chapter 1, the first practice listed was Tolerate interruptions and unavailability of service dependencies and connectivity. In the edge device hub pattern, you should expect that each individual connection might be unavailable and plan for that eventuality. That may mean planning for devices to lose connectivity to their hub and for performance fluctuations. If that happens, do you plan for the devices to buffer enough untransmitted data, accept the potential data loss, or take some other approach? That may mean that the hub cannot provide the data in response to remote queries or cannot re-transmit the...
