ScaleOut’s straightforward software framework makes it fast and easy to build and deploy digital twin applications for real-time analytics and simulation. The ScaleOut Digital Twin Builder™ software toolkit provides APIs in Java and C# for constructing digital twin models, which you can then deploy to the ScaleOut Digital Twins service with just a few clicks using its web-based UI. In addition, the ScaleOut Model Development Tool™ gives non-programmers the ability to develop digital twins with easy-to-use business rules and machine learning algorithms instead of code.
Once deployed, digital twin models can run simulations or connect to live data sources for real-time analytics. The ScaleOut Digital Twins service connects to numerous popular messaging hubs, including Microsoft Azure IoT Hub, Amazon AWS IoT Core, Kafka, and a REST web service. It delivers incoming messages to digital twins and sends responses back to data sources as needed.
With the ScaleOut Model Development Tool, you can incorporate machine learning algorithms from Microsoft's ML.NET library into digital twin models. No coding is required.
Beyond just using digital twins to model physical entities, they can be organized in a hierarchy that implements interacting subsystems, such as supervisory layers, at multiple levels.
Because of their ability to encapsulate state information and application logic, ScaleOut’s digital twins can transparently migrate functionality from the cloud to the edge for maximum performance.
Digital twin models can take many forms. To simplify application design and scale for very large systems, ScaleOut’s digital twin model uses a simple, object-oriented software design that supports both real-time analytics and simulation. Each digital twin is represented by an in-memory object that either models a single device during simulation or analyzes messages from a single data source for real-time analytics.
A digital twin hosts dynamic state information about its corresponding physical entity and an application-defined, message-processing method that uses and updates this information. During simulation, a digital twin uses a second method that runs at every time step to implement its simulation model, and twins can exchange messages.