Publish/Subscribe Engine QoS Control via Predicate Probability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing publish/subscribe systems face challenges in ensuring quality of service (QoS) with real-time requirements, particularly in scenarios involving the Internet of Things, where complex topic trees are difficult to build and content-based filtering is resource-intensive for sensors with limited capabilities.
Innovation Solution
A method and apparatus for controlling subscription requests in a publish/subscribe engine by parsing requests into predicate expressions, determining occurrence probabilities, estimating message increments based on publishing rates, and adjusting QoS accordingly, allowing for improved message handling without requiring complex topic trees or extensive computations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If topic-based pub/sub model is used to ensure real-time requirements, then real-time performance is improved, but device complexity increases due to complex topic trees
Solution Approach 1:
The patent extracts the complex topic tree structure from the system and replaces it with a simplified subscription request management approach. Instead of requiring publishers and subscribers to navigate complex topic hierarchies, the system directly manages subscription requests with QoS parameters, removing the unnecessary intermediary complexity while preserving real-time delivery guarantees.
Solution Approach 2:
The patent inverts the traditional approach by not requiring predefined topic trees for real-time performance. Instead, it dynamically evaluates subscription requests against actual message publishing patterns and QoS requirements, determining real-time feasibility on-demand rather than through static complex topic structure planning.
2Adaptability or versatility
If content-based pub/sub systems are used to increase flexibility, then adaptability is improved, but quality of service control deteriorates due to dynamic data consumption
Solution Approach 1:
The patent applies preliminary action by evaluating QoS requirements and estimating message increments before subscription requests are fully processed. The system pre-calculates whether a subscription can be satisfied given current system load and message publishing rates, making QoS control decisions in advance rather than reacting to dynamic consumption patterns.
Solution Approach 2:
The patent implements feedback mechanisms that continuously monitor message publishing rates and subscription request patterns. The system uses this feedback to dynamically adjust QoS parameter evaluations and make informed decisions about subscription request acceptance, ensuring reliable QoS control while maintaining flexibility.
3Reliability
If complex operations are performed to ensure QoS in content-based systems, then quality of service is improved, but energy consumption increases beyond sensor capabilities
Solution Approach 1:
The patent introduces an intermediary pub/sub engine that performs the complex QoS evaluation and message increment estimation operations. Sensors publish messages with minimal processing, while the engine mediates between publishers and subscribers, handling the computationally intensive QoS control tasks centrally rather than at resource-constrained sensor nodes.
Solution Approach 2:
The patent enables sensors to self-service by publishing messages with basic QoS parameters without requiring them to perform complex computations. The sensors simply publish data with associated metadata, and the system automatically handles the complex QoS evaluation and subscription management, allowing sensors to operate with minimal energy expenditure.
4Reliability
If subscription requests are controlled to ensure real-time QoS, then reliability is improved, but productivity decreases due to additional control overhead
Solution Approach 1:
The patent applies partial action by evaluating only the necessary QoS parameters for each subscription request rather than performing exhaustive analysis. The system estimates message increments using simplified models and makes acceptance decisions based on key thresholds, achieving reliable QoS control without excessive processing overhead that would reduce productivity.
Data Source
AI summary
The present invention provides a method and apparatus of controlling subscription requests, which can be used in a publish/subscribe engine. The method includes parsing a received subscription request into a predicate expression; determining an occurrence probability of the subscription request based on occurrence probabilities of predicates in the predicate expression; estimating a message increment to be caused by the subscription request based on a message publishing rate associated with the subscription request and on the occurrence probability of the subscription request; and controlling a quality of service of messages based on the estimated message increment.