Streams

Sequential Data Store (SDS) stream data are values or events of the same SDS type. SDS stores stream data and indexes it by one or more properties defined in the stream's SDS type.

The SDS stream is the container that holds the data. When you create a stream, you add it to a specific namespace. Then you define an SDS type for that stream that determines the values and events in the stream, their data type, and the properties used to index the data. Once you create a stream, you cannot change its SDS type.

The SDS stream container uses metadata and tags to define information about the stream container itself. Metadata are key-value pairs that add context to the data, for example, the metadata key, Manufacturer, and a metadata value, Caterpiller. Tags are string values that represent stream attributes, for example, region. Organizing data using metadata tags and tags makes it much easier to retrieve data streams.

PI Server counterpart

An SDS stream is comparable to a PI point in the Data Archive. For example, a float32 PI point might be sent to CONNECT data services as an SDS stream with a type that contains a timestamp index and float32 value. If you use PI to CONNECT to import data into SDS, each PI point in the Data Archive is created in SDS as an individual stream and the data itself is added as values in the stream.