Polling Broker System for Scalable Client Server Connection Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In client/server environments, scalability is limited when a large number of clients simultaneously connect and send requests to the server, leading to increased chattiness and time delays, which reduces network efficiency due to bottlenecks and resource unavailability.
Innovation Solution
Implementing a polling system that uses a broker system to manage connections, where a node list is sent to the administrative system, and a connect list is multicast to managed nodes, reducing the number of simultaneous connections and optimizing data transfer by pushing data only to nodes that require it, and utilizing a representative client system to stagger requests and volunteer data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple managed nodes simultaneously connect to the administrative system to send requests and receive data, then the system provides comprehensive management coverage, but the time delay for each node increases and network chattiness increases
Solution Approach 1:
The patent segments the simultaneous connection requests from multiple managed nodes into staggered time slots. Each node is assigned a specific time window to connect and exchange data with the administrative system, preventing simultaneous access conflicts. This segmentation reduces the time delay for each node while maintaining comprehensive management coverage across all nodes.
Solution Approach 2:
The system implements periodic polling cycles where managed nodes connect to the administrative system at regular intervals rather than continuously or simultaneously. Each polling cycle includes a designated connection phase and data exchange phase, creating a rhythmic pattern of interactions that reduces network chattiness and response time delays while ensuring all nodes receive timely updates.
2Adaptability or versatility
If multiple managed nodes simultaneously connect to the administrative system, then all nodes can receive updates, but the amount of network chattiness increases and efficiency decreases
Solution Approach 1:
The patent extracts only the necessary data updates from the administrative system and pushes them selectively to specific managed nodes based on their needs and current state. Instead of broadcasting all data to all nodes simultaneously, the system identifies which nodes require which updates and delivers only those specific data packets, reducing network chattiness while ensuring comprehensive update distribution.
Solution Approach 2:
The administrative system acts as an intermediary that manages and coordinates data distribution to multiple managed nodes. It receives data updates, processes them, and then distributes them to appropriate nodes during designated time slots, preventing direct peer-to-peer communication overhead and reducing overall network chattiness while maintaining efficient update propagation.
3Ease of operation
If the broker system manages connections from multiple managed nodes, then connection coordination is provided, but the broker system becomes a bottleneck limiting scalability
Solution Approach 1:
The patent segments the connection management function across multiple administrative systems or distributes it across different time slots and nodes. Instead of relying on a single broker system to handle all connections, the load is divided among multiple components, eliminating the bottleneck and enabling the system to scale to accommodate more managed nodes without degrading connection coordination performance.
Data Source
AI summary
A method and system for polling an administrative system for one or more managed nodes is described. The administrative system is connected to a managed node. The administrative system is polled for data to be downloaded. A node list is sent to the administrative system. A connect list is received from the administrative system. The connect list is multicast to the one or more managed nodes.