Selective Candidate-Server Pruning for Faster Task Execution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing load-balancing methods face inefficiencies due to overburdening of queuing instances in vertical scaling and indiscriminate task assignment leading to overloaded or underutilized servers, which can increase execution times and resource wastage.

Innovation Solution

Implementing a system that selectively prunes candidate servers based on recent task completion failures, CPU utilization thresholds, and computational task counts to identify qualified servers for task assignment, enabling horizontal scaling and optimizing resource allocation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If indiscriminate task assignment is used in load-balancing, then task distribution is simplified, but server overload or underutilization occurs increasing execution times and resource wastage

Engineering Contradiction:

Improvetask assignment simplicityVSAvoidexecution time

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The system performs preliminary evaluation of candidate servers by monitoring their recent task completion history and current workload status before task assignment. This preliminary action filters out servers that have recently failed tasks or are currently overloaded, ensuring that only qualified servers are selected for task execution, thereby reducing execution times while maintaining operational simplicity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The load-balancing system implements feedback mechanisms by continuously monitoring server performance metrics including task completion status, failure rates, and current workload. This feedback information is used to dynamically adjust server selection, preventing assignment to overloaded or malfunctioning servers, thus optimizing execution time without complicating the assignment process

Inventive Principle:

Principle #23Feedback

2Quantity of substance

If vertical scaling is used to handle increased task load, then system capacity increases, but queuing instances become overburdened

Engineering Contradiction:

Improvesystem capacityVSAvoidqueuing instance burden

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The system segments the load-balancing function by introducing separate worker instances that handle task execution, distinct from the queuing instances that manage task distribution. This segmentation allows queuing instances to focus solely on task management while worker instances handle execution, preventing overburdening of queuing instances while maintaining increased system capacity through horizontal scaling

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system transitions from vertical scaling (adding more capabilities to existing queuing instances) to horizontal scaling by introducing a new dimension of worker instances. This dimensional change allows the system to handle increased task loads by adding more worker nodes rather than overloading existing queuing instances, thereby maintaining system capacity while reducing complexity

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Productivity

If all candidate servers are considered for task assignment, then server utilization is maximized, but malfunctioning servers receive tasks reducing overall reliability

Engineering Contradiction:

Improveserver utilizationVSAvoidtask completion reliability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary filtering of candidate servers by monitoring their recent task completion history and identifying servers that have recently failed tasks. This preliminary action removes malfunctioning servers from the candidate pool before task assignment, ensuring that only reliable servers are selected, thereby maintaining high server utilization while improving task completion reliability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The load-balancing system implements feedback mechanisms by continuously monitoring server performance and identifying patterns of task completion failures. This feedback information is used to dynamically adjust the candidate server pool, excluding servers showing signs of malfunction while maintaining inclusion of healthy servers, thus optimizing both utilization and reliability

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250280051A1Selective pruning of candidate load-balancing servers

Publication Date: 2025.09.04 CAPITAL ONE SERVICES LLC

AI summary

In some implementations, a load-balancing system may identify one or more computational tasks. The load-balancing system may identify a plurality of candidate servers in a load-balancing server pool. The load-balancing system may identify one or more servers by selectively pruning the plurality of candidate servers. The load-balancing system may assign the one or more computational tasks to the one or more servers.