Rules Engine Dynamic Update via Pub-Sub In-Memory Loading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Updating rules in distributed rules engines often requires downtime, increasing lead-time and making it difficult to implement urgent rules, especially in mission-critical systems like financial services where rapid changes are needed to address dynamic issues such as fraud.
Innovation Solution
A distributed system comprising a rules manager and a pub/sub service that allows rules engines to subscribe to new rules, enabling them to download and execute new rules in RAM without downtime, using a pub/sub service like Apache Kafka for asynchronous communication, allowing for real-time updates without disrupting existing operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional rule updating methods are used in distributed rules engines, then rules can be updated, but system downtime occurs and lead-time increases
Solution Approach 1:
The patent implements preliminary action by pre-compiling rules into bytecode format and pre-configuring the pub/sub service infrastructure before rule updates are needed. This allows the rules engine to rapidly deploy new rules by simply downloading pre-compiled bytecode through the established pub/sub channel, eliminating the need for time-consuming compilation and installation processes during updates.
Solution Approach 2:
The patent introduces a pub/sub service as an intermediary between the rule management system and distributed rules engines. This mediator enables asynchronous rule distribution where new rules are published to topics and automatically subscribed to by relevant engines, eliminating direct update mechanisms that cause downtime and enabling continuous operation.
2Adaptability or versatility
If rules are updated in distributed systems, then new rules can be deployed, but updating becomes extremely difficult and complex
Solution Approach 1:
The patent implements universality by creating a standardized pub/sub service infrastructure that serves multiple rules engines across different systems simultaneously. The same publishing mechanism and subscription model work for various rule types and engines, providing a universal update pathway that simplifies deployment across diverse distributed systems.
Solution Approach 2:
The patent uses copying by distributing rule bytecode copies through the pub/sub service to multiple rules engines. Instead of complex installation processes, each engine receives and stores local copies of rule bytecode, enabling rapid deployment through simple data replication rather than complex update procedures.
3Productivity
If frequent rule changes are implemented, then the system can respond to changing regulations and policies, but system downtime increases
Solution Approach 1:
The patent ensures continuity of useful action by enabling rules engines to continue processing requests while downloading and loading new rule bytecode in the background through the pub/sub service. The asynchronous update mechanism allows the engine to switch to new rules without interrupting service, maintaining continuous operational capability during frequent rule changes.
Solution Approach 2:
The patent applies preliminary action by pre-compiling rules into bytecode format before deployment. This advance preparation eliminates compilation time during updates, allowing rapid rule changes to be implemented by simply downloading pre-compiled bytecode through the pub/sub service without interrupting engine operation.
Data Source
AI summary
Computer-implemented systems and methods update rules engines in a distributed computer system with new rules in a lightweight, non-intrusive, real-time manner. Rules engines are subscribed to a pub/sub service for new rules pertaining to a topic. A rules manager publishes a new rule to the pub/sub service with an associated topic. The rules engines download the new rule and store the new rule in-memory for execution without downtime.