1.7. Notes on IOT#

More related papers:

  • Google Scholar:

This is an active area of research. There seem to be several proposals, how effective are they, and are they used anywhere?

1.8. Terms#

SCADA

supervisory control and data acquisition. A category of software applications for controlling industrial processes, which is the gathering of data in real time from remote locations in order to control equipment and conditions.

1.9. Considerations and Observations#

  1. CAP theorem (https://en.wikipedia.org/wiki/CAP_theorem) In a distributed system one has to choose 2 out of the 3 of consistency, availability, and partition tolerance. In the presence of network partitions (compute nodes being in a different subnet, like edge compute and cloud compute) one has to choose between consistency, returning an error or time-out in case of failure, and availability, returning the latest value in case of failure.

  2. Azure IOT edge (https://azure.microsoft.com/en-us/products/iot-edge/#iotedge-overview) offers compute for a on-premise server with container support. Not the actual edge devices (not this service at least)

  3. There are a number of choices for IoT connectivity:

    1. OpenConnectivity.org OPEN CONNECTIVITY FOUNDATION (OCF) With reference implementation: Architecture Another standard with implementation they’ve defined Home

    2. OPC Foundation for Industry 4.0 Home Page - OPC Foundation With specifications: Documents - OPC Foundation Introduction: What is OPC UA? A practical introduction