This project is designed to simplify deployment and scaling of Connected Datacloud Devices.
Following GCP services are used:
- Pub/Sub
- Cloud Storage
- Compute Engine
- App Engine
- Cloud Functions
- DialogFlow
- IAM
The demo will deploy dexcom cloud function to connect with Dexcom API and stream data into the BigQuery using DataFlow.
- This demo uses Dexcom API to pull data from the sandbox every 15 minutes using Cloud Scheduler and CLoud Function.
- Cloud Function uses egvs endpoint and ingests data via Pub/Sub into the BigQuery
dataclouddataset,dexcomtable. - Retrieved data contains user's estimated glucose value (EGV), including trend and status information.
Created GCP project and assign Billing Account. Note down the Project_ID.
Authorize gcloud to access the Cloud Platform with Google user credentials.
gcloud auth loginSet Project ID
export PROJECT_ID=<your_project_id>git clone https://github.com/evekhm/gcp_cloud_function.git demo
demo/bin/doitCheck its working. If all steps completed sucessfully, data will begin appearing in BigQuery and could be queried:
bin/querySample Output:
Alternatively, following Query could be run from the Cloud Console:
- Navigate to the Cloud Console and search for Big Query.
- Find dataset called
datacloudand table calleddexcom. This is data retrieved from the dexcom device using API with glucose monitoring activiting. - Run following Query
- Replace <PROJECT_ID> with your Project ID
SELECT SELECT userId, JSON_EXTRACT(message, '$.systemTime' ) AS date,
JSON_EXTRACT(message, '$.value' ) AS value,
JSON_EXTRACT(message, '$.trend' ) AS trend,
JSON_EXTRACT(message, '$.trendRate' ) AS trend_rate
FROM `<PROJECT_ID>.datacloud.dexcom`
ORDER BY dateFollowing command will delete previously created resources. Project itself will not be deleted.
./bin/clean