ccraddock/beiwe-backend-serverless

Beiwe is a smartphone-based digital phenotyping research platform. This is the Beiwe backend code

beiwe-backend-serverless's Issues

Question regarding the serverless plan

Hi,

I'm coming from the beiwe-backend repo and I would like to know what are your plans for the beiwe project. I have a beiwe project deployed but I would like to know if you will be moving to serverless because I would be interested in learning about serverless and applying it to the project too, to reduce infrastructure management complexity and reduce (if possible) economic costs.

What do you plan on to make serverless?

I was thinking about:

• having the website as a static web in S3 with content pulled from DynamoDB or RDS.
• data processing made with lambda, batch jobs, or any other suitable service.
• analytics made with batch jobs too (as it is)
• data upload managed by Kinesis or IoT core.

Thank you very much.

add survey name

• create form for entering survey name
• add survey name to model
• update survey viewing code to use survey name

update all chunks to use new file name convention

add survey processing to chunker

• function to process survey_timings files to extract relevant information, should do this from RAW data rather than CHUNKS, which may have the results spread over different chunks

• function to upload results of processing survey_timings to PipelineRegistry

• script to test functionality

• modify chunker to call survey processing functionality when a survey timings file is encountered

• integrate information from app log to get presentation time

change chunker file name to first timestamp in the file, to address 'interleaved writes'