IoT Edge device high availability RRS feed

  • Question

  • Hello,

    We are planning a new project and would like to use azure IoT Edge. However we have a requirement for low downtime. We're trying to determine if azure IoT edge devices can be clustered for high availability? We need all the devices in our stores to be grouped by store and be able to run azure functions on the edge device(s) while they are offline. However we are concerned that it *seems* like we can only have one edge device. We'd like to have an automatic failover option to another local edge device. Is that possible? How are people handling high availability? I can't imagine everyone has a single point of failure in their edge device. Some guidance would be much appreciated.

    Our use case is: we have tanning beds with odroids connected to them that "set" the beds using arduino microcontrollers. However since these beds use immense amounts of power, it isn't uncommon for power issues to take out a part of the salon while still leaving part of it available for use. Because of this, we're worried about implementing IoT edge across our store chain; if the edge device goes down, it would prevent all of our tanning beds from working. How would we prevent this single point of failure? 

    Tuesday, March 12, 2019 1:00 PM


All replies