From a720e8e7e09c6173d962011d768357363ea3d063 Mon Sep 17 00:00:00 2001 From: Jim Bennett Date: Mon, 5 Jul 2021 16:27:04 -0700 Subject: [PATCH] Adding extras --- .../lessons/3-run-fruit-detector-edge/README.md | 8 +++++++- .../lessons/3-run-fruit-detector-edge/assignment.md | 8 ++++++-- 2 files changed, 13 insertions(+), 3 deletions(-) diff --git a/4-manufacturing/lessons/3-run-fruit-detector-edge/README.md b/4-manufacturing/lessons/3-run-fruit-detector-edge/README.md index 2d6a44be..cb163884 100644 --- a/4-manufacturing/lessons/3-run-fruit-detector-edge/README.md +++ b/4-manufacturing/lessons/3-run-fruit-detector-edge/README.md @@ -588,6 +588,10 @@ This is the expected behavior - images are not sent to the cloud for classificat ## 🚀 Challenge +Running AI models on edge devices can be faster that in the cloud - the network hop is shorter. THey can also be slower as the hardware than runs the model may not be as powerful as the cloud. + +Do some timings and compare if the call to your edge device is faster or slower than the call to the cloud? Think about reasons to explain the difference, or lack of difference. Research ways to run AI models faster on the edge using specialized hardware. + ## Post-lecture quiz [Post-lecture quiz](https://brave-island-0b7c7f50f.azurestaticapps.net/quiz/34) @@ -595,7 +599,9 @@ This is the expected behavior - images are not sent to the cloud for classificat ## Review & Self Study * Read more about containers on the [OS-level virtualization page on Wikipedia](https://wikipedia.org/wiki/OS-level_virtualization) +* Read more on edge computing, with an emphasis on how 5G can help expand edge computing in the [What is edge computing and why does it matter? article on NetworkWorld](https://www.networkworld.com/article/3224893/what-is-edge-computing-and-how-it-s-changing-the-network.html) +* Learn more about running AI services in IoT Edge by watching the [Learn How to Use Azure IoT Edge on a Pre-Built AI Service on the Edge to do Language Detection episode of Learn Live on Microsoft Channel9](https://channel9.msdn.com/Shows/Learn-Live/Sharpen-Your-AI-Edge-Skills-Episode-4-Learn-How-to-Use-Azure-IoT-Edge-on-a-Pre-Built-AI-Service-on-t?WT.mc_id=academic-17441-jabenn) ## Assignment -[](assignment.md) +[Run other services on the edge](assignment.md) diff --git a/4-manufacturing/lessons/3-run-fruit-detector-edge/assignment.md b/4-manufacturing/lessons/3-run-fruit-detector-edge/assignment.md index da157d5c..a50d264b 100644 --- a/4-manufacturing/lessons/3-run-fruit-detector-edge/assignment.md +++ b/4-manufacturing/lessons/3-run-fruit-detector-edge/assignment.md @@ -1,9 +1,13 @@ -# +# Run other services on the edge ## Instructions +It's not just image classifiers that can be run on the edge, anything that can be packaged up into a container can be deployed to an IoT Edge device. Serverless code running as Azure Functions, such as the triggers you've created in earlier lessons can be run in containers, and therefor on IoT Edge. + +Pick one of the previous lessons and try to run the Azure Functions app in an IoT Edge container. You can find a guide that shows how to do this using a different Functions app project in the [Tutorial: Deploy Azure Functions as IoT Edge modules on Microsoft docs](https://docs.microsoft.com/azure/iot-edge/tutorial-deploy-function?view=iotedge-2020-11&WT.mc_id=academic-17441-jabenn). + ## Rubric | Criteria | Exemplary | Adequate | Needs Improvement | | -------- | --------- | -------- | ----------------- | -| | | | | +| Deploy an Azure Functions app to IoT Edge | Was able to deploy an Azure Functions app to IoT Edge and use it with an IoT device to run a trigger from IoT data | Was able to deploy a Functions App to IoT Edge, but was unable to get the trigger to fire | Was unable to deploy a Functions App to IoT Edge |