Assets, Channels, and Runs
Understand Sift’s core data model, which organizes time-series data into assets, channels, and runs for efficient data management and analysis.
Assets and Channels
An asset represents a physical or virtual entity that generates time-series data through its channels, An asset could be a physical object, such as a vehicle or hardware testbed, or a virtual system like a simulator or CI pipeline.
A channel corresponds to a data stream, typically from a sensor, that produces time-series data. Each channel is associated with a specific data type and can be visualized on the Sift explore page. Here’s an example of how an asset appears in the Sift UI:
Channels can be plotted on the explore page, as shown below with an enum channel and a string channel:
Channel data is streamed into Sift, a process detailed in the Data Ingestion section of this documentation.
Runs
A run represents a discrete period during which data is captured from one or more assets. Runs enable grouping of related data by time, facilitating focused analysis and comparison.
The following image shows the same asset shown above, but displayed within the context of a single run:
