Unified Event Queue for Heterogeneous Message Prioritization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data processing systems face inefficiencies in handling heterogeneous event inputs, requiring separate software and hardware structures for different types of events, leading to high maintenance costs and underutilization of resources.

Innovation Solution

An integrated queue structure that prioritizes events, allows deferred delivery based on event types, and reorders events by origination time, using a single queue to process diverse events efficiently and effectively utilize shared resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If separate software and hardware structures are used for different event types, then appropriate processing for each event type is enabled, but system complexity and maintenance costs increase

Engineering Contradiction:
Improveprocessing capability for different event typesVSAvoidsystem structure complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges multiple separate event queues for different event types into a single unified event queue. This unified queue uses a priority-based ordering mechanism to handle heterogeneous events (real-time, near real-time, and batch events) together, eliminating the need for multiple separate queue structures while maintaining appropriate processing priorities for each event type.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The unified event queue is designed to be universal, handling multiple types of events (real-time, near real-time, and batch events) through a single structure. The queue uses event priority fields and ordering mechanisms that work across all event types, making the system more versatile without requiring separate specialized structures for each event category.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Reliability

If separate queues are maintained for different event types, then event processing priorities are maintained, but resource utilization decreases

Engineering Contradiction:
Improveevent processing priorityVSAvoidresource utilization efficiency
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

By combining multiple event queues into one unified queue, the patent enables more efficient resource utilization. The single queue structure allows processing resources to be dynamically allocated based on event priority rather than being statically partitioned across multiple queues, reducing resource idle time and improving overall system efficiency while maintaining proper priority handling through the ordering mechanism.

Inventive Principle:
Principle #5Merging (Combining)

3Speed

If multiple separate queues are used, then near real-time performance is maintained for specific events, but operational costs increase

Engineering Contradiction:
Improvenear real-time event responseVSAvoidmaintenance cost
Core Design Contradiction:
SpeedVSEase of operation

Solution Approach 1:

The unified event queue maintains near real-time performance for time-sensitive events by using priority-based ordering where real-time events are placed at the head of the queue. This eliminates the need for multiple separate queue structures that would require coordinated maintenance, reducing operational complexity and maintenance costs while preserving the speed requirements for near real-time event processing.

Inventive Principle:
Principle #5Merging (Combining)

4Loss of energy

If a single integrated queue is used, then resource utilization improves, but handling heterogeneous event characteristics becomes more difficult

Engineering Contradiction:
Improveresource utilizationVSAvoidevent characteristic differentiation
Core Design Contradiction:
Loss of energyVSDifficulty of detecting and measuring

Solution Approach 1:

The unified event queue incorporates local quality by associating specific attributes (such as priority indicators and timing characteristics) with individual events based on their type. Real-time events receive high priority placement, near real-time events receive medium priority, and batch events receive low priority. This localized differentiation within the unified structure enables appropriate handling of heterogeneous events while maintaining resource efficiency.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system uses parameter changes to differentiate and handle heterogeneous events within the unified queue. By modifying the priority parameter and timing parameters based on event type, the queue dynamically adjusts the processing order and delivery timing for each event, enabling appropriate handling of real-time, near real-time, and batch events without requiring separate queue structures.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3555747B1Heterogeneous event queue
Publication Date: 2021.12.08 AB INITIO TECHNOLOGY LLC
  • EP3555747B1 patent drawingFigure 1
  • EP3555747B1 patent drawingFigure 2
  • EP3555747B1 patent drawingFigure 3~4

AI summary

Processing multiple kinds of event messages in a computing system includes storing received event messages in a data storage system that includes storage (220) for a first plurality of records (222) associated with event messages. Each event message includes a timestamp. Records (222) associated with event messages include a field (426) indicating a target delivery time for an event result for the event message, the target delivery time for at least some event messages being determined according to a kind of the event message. The stored event messages are processed to deliver event results based on information in the stored event messages and the target delivery times in the records (222) associated with the event messages. Event messages are prioritized to deliver event results according to stored information indicating priority. A target delivery time is computed for each event message in a first set of one or more event messages that store information indicating the same priority for all of the event messages in the first set based on a fixed delay relative to its respective timestamp. Event results are delivered for event messages that are deliverable based on a comparison of their target delivery times to a clock time.