Initial demo of a data streaming application using the CLNR dataset.
This demo will show a number of containers processing data streams of smart meter readings. The data was generated by the Customer Led Network Revoultion project. Initially the demo will use the TC1a dataset which has domestic smart meter readings on 30 minute intervals.
The IngestAPI module exposes a REST endpoint which will accept CSV or JSON objects representing smart meter readings. These messages are sent downstream as JSON objects on a Kafka Topic. The CSV represents one sample of smart meter containing the customerId, timestamp and kWh reading since the last sample.
The Persist to DB module will store the JSON objects it receives as Readings in a MySQL database. In a future version of the demo this will be monitored by Debezium.
In order to run this example you need to deploy a Postgres database into your project. This can be achieved with the following command
oc new-app -e POSTGRESQL_USER=luke -ePOSTGRESQL_PASSWORD=secret -ePOSTGRESQL_DATABASE=my_data openshift/postgresql-92-centos7 --name=my-database
Once the Postgres has started the IngestAPI and PersistToDB containers can be deployed using the command mvn clean package fabric8:deploy
from the module directories.
There is a script in the smart-meter-simulator project which will send sample data to the application.
Set the INGEST_URL
environment variable to point to your Ingest API application and run the start.sh
script with a parameter of one fo the files in the resources directory.
$ oc get routes
NAME HOST/PORT PATH SERVICES PORT TERMINATION WILDCARD
ingest-api ingest-api-hardcoded-test.127.0.0.1.nip.io ingest-api 8080 None
persist-to-db persist-to-db-hardcoded-test.127.0.0.1.nip.io persist-to-db 8080 None
$ export INGEST_URL=ingest-api-hardcoded-test.127.0.0.1.nip.io
$ ./start.sh small.csv
Sending data to ingest-api-hardcoded-test.127.0.0.1.nip.io
2011.05.01,00:00:00,1077,0.019
201
2011.05.01,00:00:00,1083,0.113
201
...
^C
$