Introduction to the Cumulocity platform

Device integration

Cumulocity offers multiple methods for connecting devices:

• Pre-integrated IoT gateways: Choose from certified devices offered by Cumulocity certified partners for the easiest integration.

• thin-edge.io: thin-edge.io is recommended for custom device integration. Follow the Getting started with thin-edge.io tutorial for a hands-on example.

• Direct integration: Connect devices via HTTP REST or MQTT interfaces. MQTT interfaces are designed for minimal device-side logic, suitable for microcontroller-based devices.

• LPWAN support: Native LWM2M support and LoRa Network Server integration for Low-Power-Wide-Area-Network devices.

Once a device is connected to Cumulocity, it can start streaming equipment data into the platform. A common practice is to use a gateway device for data acquisition. These gateways are often connected to the different sensors and fieldbus devices of the equipment and responsible for collecting and forwarding relevant data to the IoT platform.

To facilitate the data integration, Cumulocity comes with an OPC UA integration and the Cloud Fieldbus technology. These provide configuration-driven ways to easily integrate OPC UA-enabled equipment and prominent fieldbus protocols like CAN, Profibus or Modbus.

In addition to this, many Cumulocity certified partner devices bring their own software stack with support for many other protocols.

Device management

Effectively managing devices is a critical yet often overlooked aspect of deploying IoT solutions. Without proper device management in place, it will become increasingly difficult to keep your devices healthy and up-to-date over time, leading to security risks, operational failures, and increased maintenance costs. This is especially true once you start scaling up your fleet of devices as well as adding new functionalities and thereby complexity to your deployment.

Connecting equipment to the internet requires secure and robust device management practices. Cumulocity greatly reduces the complexity and time required to manage a heterogeneous fleet of connected devices across its lifecycle through its comprehensive device management features:

• Device onboarding: Register and integrate any number of devices in just one go.
• Over-the-air (OTA) firmware and software updates: Keep your devices secure and up to date by efficiently rolling out the latest available software or firmware version.
• Configuration management: Easily change the settings of your devices by applying new configurations.
• Connection and connectivity monitoring: Quickly identify devices that stopped communicating and identify the underlying issue.
• Remote troubleshooting: Use a set of tools to resolve identified issues and minimize downtimes.
• Device replacement: Replace your physical devices without loosing the data history.

IoT data management

Cumulocity leverages a canonical yet extensible data model to represent data from different device types and protocols consistently. This approach decouples device integration from IoT applications.

To bring IoT sensor data into a meaningful context, Cumulocity offers the Digital Twin Manager (DTM). DTM enables users to model different asset types they are working with and their properties. These models can then be used to create virtual representations of physical assets and their hierarchical relationships. DTM then allows users to link sensor data directly to this asset hierarchy.

By modeling assets and their attributes, organizations can create a comprehensive view of their IoT data that is leveraged by all Cumulocity applications and can also be queried via REST APIs.

Cumulocity also offers the optional DataHub application. Cumulocity DataHub allows for efficient querying of your IoT data using SQL via standard ODBC/JDBC interfaces for an efficient integration of your IoT data with external data lake stores. This capability is particularly valuable for organizations dealing with large volumes of data, as it allows for a seamless integration of the IoT data into existing Business Intelligence, Analytics and AI tooling and workflows.

Application enablement

Cumulocity provides a comprehensive suite of tools and applications to make it easy to use the connected equipment to drive business outcomes.

At the core of this offering is the Cockpit application which offers a range of features that allow users to visualize fleet and equipment Key Performance Indicators (KPIs) through flexible dashboarding, create and manage reports for data analysis and business insights and efficiently manage alarms to ensure prompt response to critical events.

One of Cumulocity’s core design principles is to allow for customization and extension.

For this, Cumulocity provides many self-service customization options such as the following:

1. White-labeling: All applications can be easily branded using the branding manager, allowing organizations to maintain their visual identity.

2. Custom dashboards: Users can create tailored visualizations for their equipment and key performance indicators using the dashboarding feature.

3. Real-time analytics: The Analytics Builder application enables users to define visual real-time rules for data processing and decision-making.

4. Plug-ins and extensions: A wide array of plug-ins and extensions is available to enhance platform functionality.

Cumulocity has also been designed with easy extensibility by developers in mind. This is enabled through well-documented open APIs exposing the complete platform functionality along with a Command Line Interface (CLI) for efficient development workflows and an active developer community providing a platform for knowledge sharing and problem-solving. All aspects of the platform are extensible:

1. Device-side logic: Developers can use the thin-edge.io framework for integrating device-side logic.

2. UI applications: The Web SDK allows developers to seamlessly extend any UI application with new functionality or leverage any Cumulocity UI component for your entirely own UI application.

3. Backend services: The managed microservices hosting together with the Microservice SDK makes it easy to develop, deploy and operate custom backend services.

Analytics

Cumulocity offers powerful analytics capabilities that enable businesses to extract valuable insights from their data as well as to operationalize derived insights using real-time rules. 

For this, Cumulocity comes with real-time data streaming and processing capabilities, allowing businesses to process the incoming data and turn it into insights or actions. The predefined smart rules offer a wizard-driven approach to get started with data analytics, whereas the Streaming Analytics application offers flexibility to create even more rules, ranging from simple data transformations to more advanced business metric calculations like uptime or utilization rate, either for individual devices & assets or aggregated over a subset of the entire connected fleet. Real-time rules also allow for both triggering external workflows (such as the creation of a service ticket) as well as sending operations back to the equipment, for example, to optimize machine operations based on analytical insights.

Not all metrics and insights are solely relying on real-time incoming data. Metrics like “remaining useful life” of a device or asset include some kind of benchmark or learning based on historical data from that same or from similar devices or assets, and that is where AI and Machine Learning (ML) come in. The Cumulocity DataHub makes it easy for data analysts and data scientists to access the equipment data which is required to create and/or train any kind of AI model; the support for SQL via standard ODBC/JDBC interfaces allow for a seamless integration with the data science workbench or open source tooling of choice.

Bear in mind that the real value of AI/ML comes with the operationalization of the created models. You need to close the loop onto the incoming IoT data and switch from historical insight generation to real-time insight generation. Here again does our Streaming Analytics application come into play to support you in orchestrating the flow of data towards the deployed model and afterwards working with the model output to, for example, raise an alarm of trigger a workflow. For more information on machine learning capabilities, refer to the Machine Learning introduction.

Platform management

Cumulocity provides a platform management solution that allows to efficiently control and secure IoT deployments, whether provided as Software-as-a-Service (SaaS) in the cloud or installed on-premises. Through a dedicated Administration application, the platform allows administrators to manage key areas such as tenant and application management, user roles and permissions, and secure data sharing.

Key Capabilities

• Multi-tenancy support: With Enterprise tenants, Cumulocity supports full multi-tenancy, providing clear data separation and enabling organizations to securely manage multiple tenants from a single instance.

• Application management: Fully customize the look and feel of your tenants and applications using white-labeling and control, deploy, subscribe, and monitor your custom web applications, extensions and microservices.

• Advanced security and access control: The platform offers fine-grained Role-Based Access Control (RBAC) to define precise permissions for devices, users, and services. Integration with existing Identity and Access Management (IAM) systems is supported through Single Sign-On (SSO) for streamlined access management, while certificate-based authentication ensures device-level security to prevent unauthorized access.

• Data sharing and synchronization: Using the data broker, administrators can enable secure data exchange between different tenants and efficiently manage data sharing policies. This feature allows for safe and controlled device data sharing across different tenant environments.

Deployment options

Cumulocity offers flexible deployment options to accommodate various business and regulatory requirements:

• Cloud-based (SaaS) deployment: Businesses can leverage the platform as a fully managed SaaS solution in the cloud, benefiting from seamless updates, scalability, and reduced infrastructure management overhead.

• On-premises deployment with Cumulocity Edge: For environments requiring local installations, Cumulocity Edge provides the full capabilities of the platform on a single industrial PC or a locally managed Kubernetes environment. This option allows businesses to run the IoT platform independently from the cloud, making it ideal for environments with limited connectivity or stringent security requirements. The platform can operate in air-gapped networks and supports strict security measures, including STIG compliance for enhanced cybersecurity.

With its consistent architecture, Cumulocity enables organizations to develop once and deploy seamlessly across both cloud and edge environments, ensuring unified management and consistent functionality. The data broker further supports this flexibility by enabling secure data synchronization between on-premises and cloud-based deployments, allowing for efficient data exchange while maintaining complete control over data security.