Introduction to the Time Series Historian

A NoSQL binary file system designed for efficient time-series data management.

Our mission in developing our time series historian solution is centered around six core goals that directly address the needs of manufacturing environments:

Data Integrity and Security: Securely log and store data without altering original values, ensuring data integrity for long-term storage.


Long-Term Data Retention: Capable of storing data for extended periods, often exceeding seven years, as required by the end user.


Industrial Resilience: Designed to withstand and adapt to industrial conditions like communication failures, power cycling, and hardware issues, minimizing data loss.


Redundancy for Critical Operations: Provide redundancy to maintain continuous data collection and ensure reliability in mission-critical applications.


Trending for Operational Insights and Diagnostics: Include a robust trending tool to support process monitoring and facilitate efficient diagnostics.


Standardized Data Access: Enable data retrieval through standardized methods to ensure compatibility with existing tools and systems.

 

The Timebase Time Series Historian is a high-performance NoSQL binary file system designed for efficient data storage and retrieval. It is specifically built to handle time-series data, making it ideal for manufacturing, automation, and industrial environments. Key features include:

  • NoSQL Binary Storage:
    Data is stored in a NoSQL format, enabling fast read and write operations without the overhead of relational databases.

  • Efficient Data Access:
    This architecture minimizes data retrieval times, especially for large datasets or long historical queries.

  • Scalable and Robust:
    The system can handle high data volumes with ease, supporting up to 25,000 updates per second.

File System Architecture

The historian is organized into datasets, each comprising a series of independent binary files called Timebase Data Files. These files collectively represent all data logged to that dataset, with a new Timebase Data File created every hour to keep data segmented and organized. Each file contains all timestamps, values, and quality information for that specific hour, using a columnar storage format that enables the rapid storage and retrieval of large volumes of data.

The full tag list, along with all properties and metadata, is stored separately in a tag.config file, which is necessary for a complete and minimum backup. Even though the tag.config file is required, the independent nature of each hourly Timebase Data File simplifies management, making it easy to move, back up, or recover files without complex processes. Additionally, old files can be archived or removed as needed without disrupting ongoing operations. By segmenting data hourly, the system ensures faster queries, especially for time-bound requests.

Active Redundancy with Dual Logging Sessions

To ensure high availability and reliability, Timebase supports active redundancy by creating logging sessions that target two separate historians simultaneously. This approach ensures that your data is logged and synchronized across two instances, providing seamless redundancy and minimizing the risk of data loss.

How Active Redundancy Works:

  • A logging session is configured to write identical data to two independent Timebase historians.
  • Both historians operate concurrently, ensuring that all incoming data is stored in both locations in real time.
  • If one historian becomes unavailable, the other continues logging without interruption.