Lesson 5 Lecture Time! - Water-Monitor/IoT-Portfolio-FJS GitHub Wiki
In the last part of the lesson, we watched videos and had to answer a few questions about IoT in general and different frameworks.
First Video: Discussing the Challenges of IoT
What are the key challenges?
- All IoT systems have to work with each other. Therefor, integration is a huge challenge since there are already so many different kind of standard and levels.
- Cost are in general a big challenge. Solving the integration means also high costs.
- Low bandwidth, connectivity, ...
What is/are generic solution(s) for it?
- Hybrid integration platform hip
Two examples for used pattern:
- Filtering
- Observer pattern
Optional: Definition/relation of cloud, edge, fog:
- The all are a device to device communication
Optional: Advantages of edge/fog computing:
- They keep data closer to the edge
Second Video: Differnet Frameworks for IoT
What are the different categories of presented frameworks?
- data pipelines, streaming, process-engine
What are the main features of each category?
- data-pipelines: transferring data
- streaming: looks at windows and sliding data
- process: integration
Example of application:
- Edge integration
Example of available tools:
- nifi
- flink
- samza
- spark
- storm
Third Video: The available tools
Common features:
- Open Source
- Connectivity
- Web UI for visual coding
- Deployable
Target audience:
- Developers, "for human beings", integration specialists
Advantages of each:
- Kura: Open Source, integration framework, best of breed ideas
- Apache camel: Open Source, integrate with different technologies, leverage gateway kura
- Node-Red: Open Source, interconnectable, great visual design, great for developing web ui, easy-to-start, use it everywhere, shareable
Disadvantages of each:
- Kura & apache camel: not easy to set up, hard to use, needs configuration before starting
- Node-Red: no binaries, you have to install runtime on the device, only good on raspberry pi, cannot do 'easily' powerful things.