Adaptive Threshold Load Balancing for Server Queues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing client-server architectures face inefficiencies in load balancing due to sub-optimal server selection and workload distribution, leading to low performance and increased latency, particularly in environments with large numbers of clients and servers, where central control methods are costly and prone to failure, and distributed algorithms like random or round-robin selection do not effectively manage server loads.
Innovation Solution
A distributed load balancing system that compares request queue thresholds between servers to dynamically adjust and equalize queue thresholds, allowing clients to reroute requests based on rejection messages, ensuring that requests are processed by servers with manageable loads, thereby optimizing load distribution across servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a central controller is used to manage load balancing, then server load distribution can be optimized, but system complexity increases and a single point of failure is introduced
Solution Approach 1:
The patent extracts the central controller from the load balancing system and replaces it with distributed intelligence embedded in each server. Each server independently makes load balancing decisions based on information exchanged with peers, eliminating the need for centralized control while maintaining optimization capabilities.
Solution Approach 2:
Servers autonomously manage their own load balancing decisions by monitoring their own queue thresholds and exchanging this information with peer servers. Each server serves itself by making intelligent routing decisions without requiring external control, thereby reducing system complexity and eliminating single points of failure.
2Ease of manufacture
If random server selection is used, then implementation is simple, but server load distribution becomes sub-optimal
Solution Approach 1:
The system transitions from static random selection to dynamic threshold-based selection. Servers continuously update their queue thresholds and exchange this information with peers, allowing the load balancing mechanism to adapt dynamically to changing system conditions while maintaining implementation simplicity through standardized peer-to-peer communication protocols.
3Ease of operation
If round-robin algorithm is used, then server assignment is simple, but load distribution is not optimal when servers have different capacities
Solution Approach 1:
The patent applies local quality by allowing each server to have its own customized queue threshold based on its specific capacity and characteristics. Instead of uniform round-robin assignment, each server independently determines its threshold, enabling servers with different capacities to handle appropriately scaled workloads while maintaining simple threshold-comparison logic for request routing.
4Productivity
If weighted probability selection is used, then server selection considers load information, but information distribution among clients becomes complex
Solution Approach 1:
Instead of having clients gather and process complex server load information from multiple sources, the patent inverts the approach by having servers proactively publish their queue threshold information to the network. Clients simply subscribe to this published information and use it for selection, dramatically simplifying the information distribution architecture while maintaining accurate load-aware selection.
Data Source
AI summary
Methods and systems for processing data requests in a client-server computing environment, such as the Web, are disclosed. A client device initially transmits a data request to a randomly selected first server device. The first server device may reject the data request if its request queue threshold is exceeded. The client device retransmits the data request, including the request queue threshold, to a randomly selected second server device. The second server device may reject the data request if the request queue threshold of the first server device is smaller than a request queue threshold of the second server device. The client device transmits the data request back to the first server device, including the request queue threshold of the second server device. The first server device processes the data request and adjusts its request queue threshold based on the request queue thresholds of the first and second server devices.