Amazon IoT SiteWise concepts - Amazon IoT SiteWise
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Amazon IoT SiteWise concepts

The following are the core concepts of Amazon IoT SiteWise:

Aggregate

Aggregates are basic metrics that Amazon IoT SiteWise automatically computes for all time series data. For more information, see Querying asset property aggregates.

Asset

When you ingest data into Amazon IoT SiteWise from your industrial equipment, your devices, equipment, and processes are each represented as assets. Each asset has data associated with it. For example, a piece of equipment might have a serial number, a location, a make and model, and an install date. It might also have time series values for availability, performance, quality, temperature, pressure, and so on. You can organize assets into hierarchies, where assets have access to the data stored in its child assets. For more information, see Modeling industrial assets.

Asset hierarchy

You can define asset hierarchies to create logical representations of your industrial operations. To create a hierarchy, you define a hierarchy definition in an asset model, and then you associate assets created from that model and the model specified in the hierarchy definition. Metrics in parent assets can aggregate data from child assets' properties, so you can calculate metrics that provide insight to your operation or a subset of your operation. For more information, see Defining asset model hierarchies.

Asset model

Every asset is created from an asset model. Asset models are declarative structures that standardize the format of your assets. Asset models enforce consistent information across multiple assets of the same type, so that you can process data in assets that represent groups of devices. In each asset model, you can define attributes, time series inputs (measurements), time series transformations (transforms), time series aggregations (metrics), and asset hierarchies. For more information, see Modeling industrial assets.

You can control where your asset model's properties are processed by configuring your asset model for the edge. Use this feature to process and monitor asset data on your local devices.

Asset property

Asset properties are the structures within each asset that contain industrial data. Each property has a data type and can have a unit. A property can be an attribute, a measurement, a transform, or a metric. For more information, see Defining data properties.

Attribute

Attributes are asset properties that represent information that generally doesn't change, such as device manufacturer or device location. Attributes can have default values. Each asset that you create from an asset model contains the default values of the attributes of that model. For more information, see Defining static data (attributes).

Dashboard

Each project contains a set of dashboards. Dashboards provide a set of visualizations for the values of a set of assets. Project owners create the dashboards and the visualizations that it contains. When a project owner is ready to share the set of dashboards, the owner can invite viewers to the project, which gives them access to all dashboards in the project. If you want a different set of viewers for different dashboards, you must divide the dashboards between projects. When viewers look at dashboards, they can customize time range to look at specific data.

Data stream

You can ingest industrial data to Amazon IoT SiteWise before you create asset models and assets. Amazon IoT SiteWise automatically creates data streams to receive streams of raw data from your equipment.

Data stream alias

Data stream aliases help you easily identify a data stream. For example, the server1-windfarm/3/turbine/7/temperature data stream alias identifies temperature values coming from turbine #7 in wind farm #3. server1 is the data source name that helps identify the OPC-UA server. server1- is a prefix added to all data streams reported from this OPC-UA server.

Data stream association

After you create asset models and assets, you can associate data streams with asset properties defined in your assets to structure your data, then Amazon IoT SiteWise can use asset models and assets to process incoming data from your data streams. You can also disassociate data streams from asset properties. For more information, see Managing data streams.

Formula

Every transform and metric property has a formula that defines how that property transforms or aggregates data. Formulas consist of property inputs, operators, and functions offered by Amazon IoT SiteWise. For more information, see Using formula expressions.

SiteWise Edge gateway

A SiteWise Edge gateway resides on the customer premises to collect, process, and route data. A SiteWise Edge gateway connects to industrial data sources by using OPC-UA, Modbus TCP, or Ethernet/IP protocols to collect data when processing or routing the data to the Amazon cloud. SiteWise Edge gateways use packs to collect data, process data at the edge, and more. For more information about available packs, see Using packs.

You can create a SiteWise Edge gateway on any device or platform that can run Amazon IoT Greengrass. For more information, see Using SiteWise Edge gateways.

Measurement

Measurements are asset properties that represent a device or equipment's raw sensor time series data streams. For more information, see Defining data streams from equipment (measurements).

Metric

Metrics are asset properties that represent aggregated time series data. Every metric has a mathematical expression (formula) that defines how to aggregate data points, and a time interval over which to compute that aggregation. Metrics output a single data point per given time interval. For more information, see Aggregating data from properties and other assets (metrics).

Packs

SiteWise Edge gateways use packs to decide how to collect, process, and route data. Currently, Amazon IoT SiteWise supports the data collection pack and the data processing pack. For more information about the available packs for your SiteWise Edge gateway, see Using packs.

Data collection pack

Use the data collection pack so that your SiteWise Edge gateway can collect your industrial data and route it to the Amazon destination of your choice. This pack is automatically added to your SiteWise Edge gateway and can't be removed.

Data processing pack

Use the data processing pack to process your data at the edge and retain it for 30 days for use in local applications.

Portal

An Amazon IoT SiteWise Monitor portal is a web application that you can use to visualize and share your Amazon IoT SiteWise data. A portal has one or more administrators and contains zero or more projects.

Portal administrator

Each SiteWise Monitor portal has one or more portal administrators. Portal administrators use the portal to create projects that contain collections of assets and dashboards. The portal administrator then assigns assets and owners to each project. By controlling access to the project, portal administrators specify which assets that project owners and viewers can see.

Project

Each SiteWise Monitor portal contains a set of projects. Each project has a subset of your Amazon IoT SiteWise assets associated with it. Project owners create one or more dashboards to provide a consistent way to view the data associated with those assets. Project owners can invite viewers to the project to allow them to view the assets and dashboards in the project. The project is the basic unit of sharing within SiteWise Monitor. Project owners can invite users who were given access to the portal by the Amazon administrator. A user must have access to a portal before a project in that portal can be shared with that user.

Project owner

Each SiteWise Monitor project has owners. Project owners create visualizations in the form of dashboards to represent operational data in a consistent manner. When dashboards are ready to share, the project owner can invite viewers to the project. Project owners can also assign other owners to the project. Project owners can configure thresholds and notification settings for alarms.

Project viewer

Each SiteWise Monitor project has viewers. Project viewers can connect to the portal to view the dashboards that project owners created. In each dashboard, project viewers can adjust the time range to better understand operational data. Project viewers can only view dashboards in the projects to which they have access. Project viewers can acknowledge and snooze alarms.

Property alias

You can define aliases on asset properties to easily identify an asset property when you ingest or retrieve asset data. When you use a SiteWise Edge gateway to ingest data from servers, your property aliases must match the paths of your raw data streams. For more information, see Mapping industrial data streams to asset properties.

Property notification

When you enable property notifications for an asset property, Amazon IoT SiteWise publishes an MQTT message to Amazon IoT Core each time that property receives a new value. The message payload contains information about that property value update. You can use property value notifications to create solutions that connect your industrial data in Amazon IoT SiteWise with other Amazon services. For more information, see Interacting with other Amazon services.

Transform

Transforms are asset properties that represent transformed time series data. Every transform has a mathematical expression (formula) that defines how to transform data points from one form to another. The transformed data points hold a one-to-one relationship with the input data points. For more information, see Transforming data (transforms).

Visualization

In each dashboard, project owners decide how to display the properties and alarms of the assets associated with the project. Availability might be represented as a line chart, while other values might be displayed as bar charts or key performance indicators (KPIs). Alarms are best displayed as status grids and status timelines. Project owners customize each visualization to provide the best understanding of the data for that asset.