Grid Message Distribution via Identifier Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current message distribution methods, particularly in data grids, rely on point-to-point approaches that result in network overload due to large message replication and require custom filtering policies and recipient registration, limiting efficiency and scalability.
Innovation Solution
Implementing a data grid architecture where a source node stores messages in a message grid, sending only message identifiers to recipient nodes, which then retrieve the messages, leveraging data co-location policies for efficient message delivery and implementing filtering policies in a registration grid to reduce load on recipient nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If point-to-point message distribution is used, then message delivery is achieved, but network load increases due to message replication
Solution Approach 1:
The patent extracts the message content from the distribution process and stores it centrally in a message repository. Only message identifiers (not the actual messages) are replicated and sent to recipient nodes. This separates the notification function from the message transfer function, reducing network load while maintaining reliable delivery.
Solution Approach 2:
The patent introduces a message repository as an intermediary component between the source node and recipient nodes. The repository stores messages centrally and manages their distribution, acting as a mediator that eliminates the need for direct point-to-point message replication across the network.
2Reliability
If point-to-point message distribution is used, then message delivery is achieved, but message transfer speed decreases
Solution Approach 1:
The patent extracts only the essential notification element (message identifier) from the full message and transmits it rapidly to recipient nodes. The actual message retrieval happens asynchronously from the central repository, enabling fast notification without being constrained by large message transfer speeds.
Solution Approach 2:
The patent implements preliminary action by pre-storing messages in the message repository before distribution. Recipient nodes can quickly retrieve messages using pre-computed identifiers without waiting for complex real-time message generation or filtering at each node.
3Adaptability or versatility
If custom filtering policies are implemented in each recipient node, then message filtering is achieved, but device complexity increases
Solution Approach 1:
The patent merges the filtering functionality into a centralized message repository or broker rather than distributing it across all recipient nodes. The centralized component handles message filtering based on subscription criteria, while recipient nodes only need simple subscription management, significantly reducing their complexity.
Solution Approach 2:
The patent creates a universal message repository that handles multiple functions including message storage, filtering, routing, and distribution. This single multi-functional component replaces the need for each recipient node to implement its own filtering logic, reducing overall system complexity while maintaining adaptability.
Data Source
AI summary
A method, system, and a computer program product for transferring a message from a source node to recipient nodes connected by a grid architecture. According to an aspect, a source node stores a message in a message grid. An identifier and a message-type of the message are then received in a message manager node. If the type of message to be sent to the recipient nodes is the same as the message-type of the message stored in the message grid, the message manager node sends the identifier to the recipient nodes. The recipient nodes then retrieve the message from the message grid using the message identifier. Recipient nodes register with a registration grid specifying the type of message desired to be received.