SignalK plugin to publish to Amazon Timestream, a serverless time series database.
Timestream integrates with all the rest of AWS, as well as Grafana with the Grafana Timestream Plugin.
This plugin assumes you have a Timestream database and table pre-created. If
you're running in AWS you can use a role. Otherwise you'll need to create a
user with an access key and secret key. The role/user needs permission to
timestream:DescribeEndpoints and timestream:WriteRecords, but that's it.
Note that if you're using an AWS config file, since this runs in node.js,
you'll need to set the environment variable AWS_SDK_CONFIG_FILE.
At the moment the plugin is hardcoded to only write self. The configuration
consists of the following parameters
-
Database: this is the name of your database, not the ARN
-
Table: this is the name of the table, not the ARN
-
Write Interval: a write to timestream will happen on this frequency, in seconds. Note that there is a cost to the size of the writes, as well as the volume of data stores and this is the biggest lever to control costs.
-
Filter List: this controls what signalk paths are published, the list either contains glob patterns describing the paths that should be included or excluded from publishing, for example, you might publish
"environment.*".
The plugin can fetch and stream historical data. At the moment, only one AWS profile is allowed, so if you're both reading and writing, ensure that user has both write and query permissions.