dgraph for timeseries data.md

Modeling timeseries data in Dgraph with timestamp on value object

Problem is, you will likely want to query by TimeStamp and need to index the TimeStamp in your schema. However when sending tons of mutations adding new values with high frequency the index keeps rebuilding and all attempts querying the timeseries data will either timeout or fail saying something like “try again later”. Recording a few values per minute should work fine, but for anything more aggressive run more tests before building a whole product relying on it. This approach puts timestamps on the node. An alternative approach is to put the timestamps on a facet, see below.

Schema

Name: string @index(fulltext, term, trigram) .
Value: uid @reverse .
Timeseries: [uid] .
Timestamp: datetime .
Number: float .
type NumberValue {
  Timestamp
  Number
}

Mutation

upsert {
  query {
    var(func: eq(Name, "myparameter")) {
      param as uid
    }
  }
  mutation {
    set {
      uid(param) <Value> <_:Value> .\n"
      uid(param) <Timeseries> <_:Value> .\n"
      <_:Value> <Timestamp> "2020-06-01T18:00:00-06:00"^^<xs:dateTime>
      <_:Value> <Number> "22.3"^^<xs:double>
      <_:Value> <dgraph.type> "NumberValue"
    }
  }
}

Query latest value

{
  r1(func: eq(Name, "myparameter")) {
    Name
    Value {
      expand(_all_)
      dgraph.type
    }
  }
}

Query the most recent 100 timeseries values

{
  r1(func: eq(Name, "myparameter")) {
    Timeseries (orderdesc: Timestamp, first: 100) { 
      expand(_all_)
    }
  }
}

Modeling timeseries data in Dgraph with timestamp on the edge

... not tried yet.