Glossary
This glossary provides definitions for key terms and concepts used in the WEDA platform and edge AI application development.
Terms and Definitions
A
ACC (Advantech Container Catalog)
A comprehensive collection of ready-to-use, containerized software packages designed to accelerate the development and deployment of Edge AI applications, pre-validated and optimized for Advantech hardware.
Access Token
A short-lived credential used to authenticate API requests, obtained after successful authentication through OAuth 2.0 or other authentication mechanisms.
Activation
The process of registering and connecting an edge device to WEDA Core, establishing authentication and initial configuration for device management.
Alert Event
A record of an Alert Policy firing against a specific DataStream, capturing when the condition was met, its current lifecycle status, and when it was resolved. Alert Events are read-only history: they are generated by the platform, not created by users.
Alert Policy
An independent, Org-level resource defining conditions and thresholds that trigger notifications when a monitored DataStream's value exceeds defined limits. A single Alert Policy can attach to multiple DataStreams, enabling proactive system monitoring.
Analyzer
A cloud-side computation in WEDA Core that reads one or more existing DataStreams and produces an Output DataStream. The TransferFunction type applies a custom arithmetic expression; other types are reserved for future releases.
API (Application Programming Interface)
A set of protocols and tools that allows different software applications to communicate and exchange data with the WEDA system.
Auto-Provisioning
Zero-touch device onboarding capability where edge devices automatically discover and register to WEDA Core upon startup using MAC address-based identification, eliminating manual configuration requirements.
B
BMC (Baseboard Management Controller)
A specialized microcontroller embedded on the motherboard that enables out-of-band management, allowing remote power control, BIOS configuration, and OS updates even when the primary system is powered off.
BMS (Building Management System)
An automated system for monitoring and controlling a building's electrical and mechanical equipment, including HVAC, lighting, and security systems.
C
Client Credentials
An OAuth 2.0 authentication flow designed for service-to-service communication where applications authenticate using client ID and secret instead of user credentials.
Container Orchestration
The automated management of containerized application lifecycles, including deployment, scaling, updates, and rollback operations across distributed edge devices using formats like Helm Chart or Docker Compose.
Container Registry
A storage and distribution system for container images, supporting both private and public registries with credential management for secure access to Docker images.
CRUD
The four basic operations of persistent storage: Create, Read, Update, Delete — fundamental database and API operations supported across WEDA Core APIs.
D
Data Bridge
A one-way forwarding rule that publishes DataStream values from WEDA Core to a customer-hosted MQTT broker. A Data Bridge is defined by two resources: an MQTT Broker Config holding the connection profile and credentials, and the Bridge itself, which binds specific DataStreams to outbound topics and sets the message assembly policy.
DataStream
The queryable, time-series data unit that all WEDA features build on — either auto-synced from a device's reported sensor (managedBy: System) or created directly, e.g. as an Analyzer's output (managedBy: User). Replaces the earlier Data Source and Data Object concepts.
Desired State
The target configuration defined by users in the back-office cloud system, representing the intended operational parameters for edge devices in the Digital Twin framework.
Device Credential
A device-specific encrypted authentication file (.bin) for single device provisioning with pre-configured access rights and security credentials.
Device State Synchronization
A mechanism that maintains a cloud-side record of each device's desired and last-reported configuration, enabling asynchronous synchronization of configuration and status information between WEDA Core and edge devices, with offline tolerance support.
Digital Twin
A virtual representation of a physical device in digital space, enabling real-time state synchronization, simulation, and remote control through standardized protocols.
Docker Compose
A tool and configuration file format for defining and running multi-container Docker applications, used in WEDA for container stack deployment.
DTDL (Digital Twin Definition Language)
A standardized language for describing models of physical devices, ensuring interoperability with platforms like Azure Digital Twins and consistent device representation across systems.
E
Edge AI
Artificial Intelligence algorithms and models deployed on edge devices rather than cloud servers, enabling real-time processing with reduced latency and improved privacy.
Edge Computing
A distributed computing paradigm that brings computation and data storage closer to the location where it is needed, improving response times and saving bandwidth.
Edge Device
A physical computing device at the edge of a network that collects, processes, and analyzes data locally before sending relevant information to central systems.
G
GPU (Graphics Processing Unit)
A specialized processor designed to accelerate parallel computing tasks, commonly used in WEDA environments for AI inference and model training acceleration.
H
Helm Chart
A package management format for Kubernetes applications that defines, installs, and upgrades containerized applications in a standardized way.
I
IPMI (Intelligent Platform Management Interface)
A standardized interface for out-of-band management and monitoring of computer hardware, enabling remote monitoring of system health, temperature, and power status.
J
JWT (JSON Web Token)
A compact, self-contained token format for secure information transmission and authentication between parties, used in WEDA for user and service authentication.
M
MAC Address
A unique hardware identifier assigned to network interfaces, used as the primary device identifier in WEDA for auto-provisioning and device registration.
Machine Learning (ML)
A subset of artificial intelligence that enables systems to learn from data patterns and improve performance without explicit programming, used for demand forecasting and behavior prediction in edge applications.
Master Key
An encrypted authentication file (.bin) enabling auto-provisioning for multiple devices in mass deployment scenarios, bundled with WEDA installer utilities.
MDC (Micro Data Center)
A localized, high-performance computing environment for on-premises deployment, operating in the WEDA Cloud layer alongside cloud deployments.
MLOps
A set of practices that combines machine learning model development, deployment, monitoring, and continuous improvement, integrated with WEDA Core for edge AI lifecycle management.
Modbus
An industrial communication protocol widely used for connecting PLCs, sensors, and other industrial automation equipment, supported by WEDA SubNode framework.
MQTT (Message Queuing Telemetry Transport)
A lightweight publish/subscribe messaging protocol commonly used for IoT device communication, supported by WEDA for telemetry data transmission.