Switching to MongoDB-based driver for companion notebook for reading data

[sgbaird]

I.e., https://ac-microcourses.readthedocs.io/en/latest/courses/hello-world/1.5-data-logging.html linking to

https://ac-microcourses.readthedocs.io/en/latest/courses/hello-world/1.5.1-pymongo.html

instead of

https://ac-microcourses.readthedocs.io/en/latest/courses/hello-world/1.5.1-aws-lambda-read.html

and then we can make a note in 1.5.1-pymongo.html pointing to 1.5.1-aws-lambda-read.html as the mirrored method.

Essentially, I think it would make more sense for people to use PyMongo directly within orchestrators, which are running full Python environments. Though, this would need to be updated in 1.5 and 1.6 assignments again.

This came to mind again when Yanghuang mentioned he was going to use Lambda AWS for the orchestrator database integration.

@SissiFeng thoughts?

[SissiFeng]

Going straight to pymongo will be more straightforward and ease the burden on everyone for the first microcourse. I will tweak the 1.5 and 1.6 assignments again.

The use of aws as a mirroring method will echo the content of courser4, and is considered to be a real-world application.

[SissiFeng]

I redesigned the original folder for the 1.5/1.6 assignments (removing the aws api for the new version) https://github.com/ACC-HelloWorld

The repo used for creating the assignment needs to be hard-forced to sync to this version in order to update the assignment template for students.

[sgbaird]

Thanks! Can you also make a PR with updates to the tutorial files?

[sgbaird]

It looks like the AWS piece got taken out completely, not just on the "read from database" section on the orchestrator side. I realized my wording above was unclear. Is the idea now that insert and find both happen at the orchestrator level, without any direct interactions with the microcontroller? Or is there still the assumption that data is being uploaded from the microcontroller and being read by the orchestrator?

ACC-HelloWorld/5-data-logging@92a9894

[SissiFeng]

The current version is still maintaining the interaction between the microcontroller and the database.
The microcontroller executes the experiment and generates the data, which is logged immediately after the experiment is completed and which the orchestrator needs to read and analyze later.

[sgbaird]

🙏

[SissiFeng]

Thanks! Can you also make a PR with updates to the tutorial files?

https://github.com/AccelerationConsortium/ac-microcourses/pull/141/files

The new PR removed the aws api part, and also included some tutorial words added in course2 considering the assignment changes.