Unified Event Processing System for Stream and Window Operators
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems require multiple implementations and maintenance to process data events using both stateless stream operators and state-based window operators, lacking a versatile method for unified event processing.
Innovation Solution
The method involves using a thread to map stateless and state-based operators to different types of column stores, with a row/transaction identifier to track operations and access specific database versions, allowing for efficient processing and storage of event data across multiple threads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple separate systems are used to process events with stateless stream operators and state-based window operators, then each system can be optimized for its specific operator type, but the overall system complexity increases and maintenance becomes more difficult
Solution Approach 1:
The patent combines stateless stream operators and state-based window operators into a single unified event processing system. The event processing module can dynamically select and execute different operator types within the same system architecture, eliminating the need for multiple separate systems while maintaining the reliability benefits of specialized operators.
Solution Approach 2:
The unified event processing system is designed to handle both stateless stream operators and state-based window operators through a common architecture. The system can adaptively select which operator type to use based on the event characteristics, providing multi-functionality without requiring separate specialized systems.
2Device complexity
If a unified system processes both stateless and state-based operators, then system complexity is reduced, but ensuring data consistency and proper state management becomes more challenging
Solution Approach 1:
The unified event processing system segments the processing logic into distinct operator types (stateless stream operators and state-based window operators) that can be independently managed. Each operator type maintains its own state management mechanisms, ensuring data consistency while allowing the overall system to remain unified and less complex.
Data Source
Figure 1A~1B
Figure 2A
Figure 2B
AI summary
Methods, systems, computer program products, and articles of manufacture for processing events are described. An event is obtained and the event is processed to generate data using a first set of one or more operators. The generated data is stored in a first column store with a first row/transaction identifier and the first row/transaction identifier is stored in one or more first processing queues to enable further processing of the event using a second set of one or more operators.