How TDengine Enables Advanced Analytics for Industrial Data

With traditional industries increasingly moving toward Industry 4.0, the need for better analytics is more apparent than ever. Only through modern, advanced analytics is it possible to implement predictive maintenance and data-driven decision making. Furthermore, the recent advances in AI technology have sparked interest among industry leaders who are looking for ways to make their operations more efficient.

However, traditional data historians are not equipped to handle the requirements of modern analytics, much less AI. They typically lack the ability to centralize data from multiple sites, preventing analytics tools from obtaining a global view of company operations. More importantly, they are closed systems that are difficult to integrate with cutting-edge analytics tools.

TDengine is a time-series database specifically designed to resolve these issues for industrial data. It is an open system that can seamlessly integrate with the latest AI tools and models, finally bringing the power of AI to traditional industries. At the same time, TDengine is familiar to existing users, supporting standard SQL like a relational database. TDengine provides powerful and easy data analytics capabilities as follows:

1. Built-in AI agent for time-series analytics: TDgpt provides time-series data forecasting and anomaly detection, supporting AI and ML, including time-series foundation models and large language models, as well as traditional statistical algorithms, all in a single SQL statement. In addition to the algorithms and models included with TDgpt, you can also add your own in-house or open-source algorithms to TDgpt through its open SDK and use them in your applications.

2. Efficient aggregation between multiple data collection points: TDengine introduces the innovative concept of the supertable, which is a template for a type of data collection point based on the characteristics of time series data. By specifying filtering conditions when querying a supertable, you can efficiently aggregate data collection points of the same type, making it easier to organize and find data and eliminating the need for costly JOIN operations. In addition, TDengine allows you to add up to 128 labels to each data collection point, which you can delete and update later.

3. Separation of storage and compute: TDengine supports the separation of storage and compute resources. The system can start one or more query nodes as needed to increase computing resources, speed up complex queries, and reduce latency. The compute node can be a container, which can be started or stopped quickly and take full advantage of the elasticity of cloud platforms.

4. Unified analysis of historical and real-time data: TDengine automatically partitions data according to time interval. As much as 10 years of data can be stored in a database or table, and there is no concept of archived data in the TDengine system. This remains true even when tiered storage is used: the only difference between querying the latest data and data from 10 years ago is the start and end times in your SQL statement.

5. Extended functions for time series data analytics: TDengine extends standard SQL to meet the needs of of time series data analytics, providing essential functions such as cumulative summation, time-weighted average, and moving average as built-in SQL functions. Through time windows and interpolation, the timestamps of data from different data collection points can be aligned at fixed time intervals to facilitate further analysis. To learn more, see SQL Manual.

6. Real-time data analytics: TDengine provides time-driven and event-driven stream processing. Stream processing can be performed not only on a data stream generated by a single data collection point, but also on the aggregated data streams of multiple collection points. Support for user-defined functions (UDF) enables stream processing to easily provide preprocessing, transformation, and other complex compute functions.

7. Python support: TDengine provides a Python client library and supports pandas and data frames, so that data analysts can easily use familiar Python libraries to do their time-series data analytics.

8. Other convenient means of data access and analysis: TDengine provides a command line interface (CLI) with which you can run ad hoc queries and import or export data. TDengine also provides R and Matlab connectors and seamless integration with Grafana and Power BI.

9. Easy integration with third-party analytics tools: TDengine is an open system that provides standard interfaces, making it easy to integrate and test the latest analytics software with your data.

In typical Industrial IoT scenarios, TDengine can be used as a time-series data warehouse; it is no longer necessary to import time series data into a special data warehouse or data lake for processing and analysis. And its open ecosystem enabling seamless integration with AI tools brings industrial data analytics into the modern age. All in all, TDengine greatly reduces the cost of industrial data processing while opening the door to advanced, AI-enabled analytics.