Building Control Brokering Architecture for Bump-Less Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing web services platforms face delays in data presentation due to the need for applications to retrieve and process raw timeseries data from relational databases, which can lead to inefficient data processing and delayed control of building systems.
Innovation Solution
The implementation of isolated compute environments that perform calculations related to building control, with a failover mechanism to ensure continuous operation even if one environment fails, and the use of a second isolated compute environment to store results and facilitate data recovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single isolated compute environment is used to perform building control calculations, then the system structure is simple, but the reliability is low due to potential failures causing control interruptions
Solution Approach 1:
The system divides the compute environment into multiple isolated instances (first isolated compute environment, second isolated compute environment, and additional isolated compute environments). Each instance can independently perform building control calculations, allowing failover when one instance fails without affecting overall system reliability.
Solution Approach 2:
The system changes the operational state parameters of compute environments by transitioning between active, standby, and failed states. When the first isolated compute environment fails, the system activates a second isolated compute environment or additional environments to maintain control continuity, effectively using state parameter changes to ensure reliability.
2Reliability
If multiple isolated compute environments are deployed for failover capability, then the reliability improves, but the device complexity increases
Solution Approach 1:
The system creates copies of the isolated compute environment (second isolated compute environment as a copy of the first, and additional isolated compute environments as copies). These copies can take over when the primary environment fails, providing reliability without requiring fundamentally different architectural components.
Solution Approach 2:
Each isolated compute environment is designed to be universal and capable of performing the same building control calculations independently. The first, second, and additional environments all can execute the same control logic, allowing any of them to take over based on operational needs, simplifying the management of multiple environments.
3Reliability
If the first isolated compute environment encounters a failure and needs to transfer calculations, then the reliability is maintained through failover, but the loss of time occurs during the transition period
Solution Approach 1:
The system prepares the second isolated compute environment and additional isolated compute environments in advance as standby configurations. These environments are pre-configured with the necessary capabilities to perform building control calculations, so when a failure occurs, the transition to failover is minimized because the backup environments are already ready to execute.
Solution Approach 2:
The system ensures continuous building control calculations by having multiple isolated compute environments capable of performing the same functions simultaneously or in hot-standby mode. The failover mechanism maintains the continuity of useful action (building control) by seamlessly transferring calculations from the failed environment to a standby environment without significant interruption.
4Reliability
If additional isolated compute environments are used to store calculation results, then the data recovery capability improves, but the loss of time increases during environment recreation
Solution Approach 1:
The second isolated compute environment and additional isolated compute environments store copies of calculation results from the first isolated compute environment. These copied data sets enable rapid recovery without requiring time-consuming data reconstruction, as the backup environments already contain the necessary calculation results.
Solution Approach 2:
The system performs preliminary data storage by having the second and additional isolated compute environments continuously or periodically store calculation results in advance. This preliminary action ensures that when a failure occurs, the data is already prepared and available for immediate recovery, minimizing the time required to restore operational capability.
Data Source
AI summary
A distributed control system with bump-less failover and replay capability is disclosed. The distributed control system stores the state of each control application in a separate storage system. Each time the control applications runs it can store the results of the calculations required for a subsequent calculation. When a compute environment running the control applications fails another environment can begin performing the same operations. Replay capability is provided to facilitate the investigation of the control application's execution.