Provider + AWS Time stream

This provider extends the DBProvider to manage time-series with AWS Time stream. Use HAYSTACK_PROVIDER=shaytack.providers.timestream to use this provider. Add the variable HAYSTACK_DB to describe the link to the backend to read the ontology and HAYSTACK_TS to describe the link to AWS Time stream. The format of HAYSTACK_TS is :

timestream://\<database\>[?mem_ttl=<memory retention in hour>&mag_ttl=<magnetic retention in day>][#<tablename>]

The parameters mem_ttl and mag_ttl are optionals and be used only to create the table. Read this for the maximum value. The default value for mem_ttl is 8766 (1y+6h) and 400d for mag_ttl.

The table schema is

id (varchar)            -- The haystack id
customer_id (varchar)   -- The associated customer_id
unit (varchar)          -- Unit, use only with time series of quantity
hs_type (varchar)       -- python type of the time serie
measure_name (varchar)  -- 'val'
time (timestamp)        -- The timestamp of the value is microseconds
measure_value::<double> -- The value (adapt the name with the type of value)

You can publish data in this table, via AWS IoT Core for example.

• Use the same id as for Haystack.
• Add eventually a value for customer_id

$ HAYSTACK_PROVIDER=shaystack.providers.timestream \
  HAYSTACK_DB=sample/carytown.zinc \
  HAYSTACK_TS=timestream://SHaystackDemo/?mem_ttl=1&mag_ttl=100#haystack \
  shaystack

With this provider, all the time-series are inserted in AWS Time Stream. You can use shaystack_import_db with a third parameter to describe the link to the time-series database:

$ shaystack_import_db sample/carytown.zinc \
    sample/carytown.zinc \
    timestream://SHaystackDemo

Limitation

• The entities with history must have a tag kind to describe the type of values and a tag id
• AWS Time stream refuse to import a data outside the memory windows delays. See here