Industrial systems often need to keep data useful in two places at once: close to equipment at the edge and centralized in the cloud for cross-site analysis. A time-series database has to support both low-latency local access and reliable aggregation across networks that may be unstable.
Edge-cloud collaboration requirements in industrial scenarios
Edge devices such as PLCs, sensors, and gateways capture local operating data, but local data alone cannot support fleet-wide analysis. Production-line data often needs to be combined across a factory or enterprise for capacity planning, resource scheduling, and long-term optimization.
Four core challenges:
- Diverse data sources: AVEVA PI System, OPC-UA, OPC-DA, MQTT. How to unify them?
- Unstable network environments: Factory network interruptions affect sync continuity
- Massive data scale: Whether sync mechanisms can handle high throughput
- High operational costs: Traditional approaches require extensive custom coding
Core capabilities of the edge-cloud collaboration architecture
TDengine Enterprise provides a complete solution with five capabilities:
High-efficiency data synchronization: Supports sync rates of millions of data points per second, handling both high-frequency sensor writes and batch historical record uploads.
Multi-source connectivity: Compatible with AVEVA PI System, OPC-UA, OPC-DA, and MQTT. A unified data access layer eliminates the need for separate adapter development.
Flexible sync rule configuration: Users can define sync scope, frequency, and direction based on business needs. For example, the edge retains only the last hour for alerting while the cloud stores full history for trend analysis.
Resume from breakpoint and re-subscription: Built-in mechanisms ensure no data loss and no duplication after network recovery.
Historical data migration: Supports migrating historical data from edge to cloud or vice versa for unified management and long-term archiving.
Data subscription: flexible selective synchronization
Subscription granularity options include: entire database, a single supertable, or query statements with filter conditions. Subscription-based sync is configurable, highly real-time, and avoids unnecessary data transfer.
Implementation advantages
| Advantage | Description |
|---|---|
| Zero-code deployment | No coding required; simple configuration only |
| High degree of automation | Cross-region data sync is largely automated |
| Strong real-time performance | Real-time via subscription; no batch caching |
| Unified data model | Same time-series database on both edge and cloud; no format conversion overhead |
| Simple operations | Unified technology stack reduces learning and maintenance complexity |
From data aggregation to value creation
The goal is not simply to move data to the cloud. The goal is to make edge data available for broader analysis and operational decisions. After data converges in the cloud, AI analysis tools can support:
- Anomaly detection based on historical baselines
- Real-time alerting combining threshold rules and intelligent algorithms
- Production capacity forecasting through trend analysis
- Energy consumption optimization by identifying high-usage areas
Summary
Edge-cloud collaboration gives industrial teams a practical path for local responsiveness and centralized analysis. TDengine supports configurable synchronization, multi-source connectivity, recovery after interruptions, and a unified time-series model across edge and cloud deployments.
