Cloud Resource Allocator Using Predictive Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing environments face complexity in selecting optimal resource configurations and scheduling tasks, leading to long run times and significant costs due to sub-optimal resource allocation and scheduling.
Innovation Solution
A system and method for automatically allocating resources and scheduling tasks in cloud computing environments, using a predictor to generate operating constraints and a scheduler to evaluate possible operating points based on a cost function, thereby determining optimized resource allocation and scheduling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional resource allocation and scheduling techniques are used, then device complexity is reduced, but productivity deteriorates due to long run times and sub-optimal resource selection
Solution Approach 1:
The system performs preliminary actions by pre-generating multiple candidate resource configurations and scheduling options before actual workflow execution. The resource configurator creates a library of pre-evaluated resource allocations, and the scheduler pre-computes multiple scheduling scenarios, allowing the system to quickly select from pre-prepared options during runtime rather than computing everything from scratch.
Solution Approach 2:
The patent introduces intermediary components including a resource configurator that acts as a mediator between user requirements and cloud resources, a cost estimator that mediates between resource allocation and cost constraints, and a scheduler that mediates between task dependencies and resource availability. These intermediaries break down the complex optimization problem into manageable stages.
2Loss of time
If manual resource configuration and scheduling is performed, then device complexity is minimized, but loss of time increases due to significant time required for resource allocation
Solution Approach 1:
The system implements self-service automation where the resource configurator automatically generates resource configurations based on workflow characteristics without manual intervention. The scheduler autonomously evaluates candidate schedules, estimates costs, and selects optimal allocations. The system serves itself by using historical data and performance metrics to automatically refine future resource allocations, reducing both time and manual complexity.
Solution Approach 2:
The patent dynamically changes multiple parameters including resource types, allocation amounts, scheduling priorities, and cost weights based on workflow characteristics and constraints. The system adjusts these parameters automatically through the configurator and scheduler, allowing rapid adaptation to different workflow scenarios without manual reconfiguration, thereby reducing allocation time while managing complexity through parameterized control.
3Loss of energy
If sub-optimal resource selection is made, then device complexity is reduced, but cost increases due to significant costs incurred for resources used
Solution Approach 1:
The system implements feedback mechanisms where the cost estimator continuously evaluates the cost implications of resource allocations and feeds this information back to the resource configurator and scheduler. Performance metrics from executed workflows are fed back to refine future resource selections. This closed-loop feedback allows the system to learn from past decisions and continuously improve cost efficiency while managing optimization complexity through iterative refinement.
Solution Approach 2:
The patent employs parameter changes by dynamically adjusting resource allocation parameters, scheduling parameters, and cost weightings based on workflow characteristics and constraints. The system transforms the complex cost optimization problem into a parameter selection problem, where the configurator and scheduler adjust parameters to find optimal cost-performance tradeoffs without requiring exhaustive search of all possible configurations.
4Manufacturing precision
If comprehensive resource evaluation is performed, then manufacturing precision is improved for optimal resource selection, but difficulty of detecting and measuring increases
Solution Approach 1:
The patent segments the comprehensive resource evaluation into distinct modular components: the resource configurator handles resource configuration generation, the cost estimator handles cost analysis, and the scheduler handles scheduling optimization. Each component evaluates specific aspects independently, producing intermediate results that are integrated to form the final optimal allocation. This segmentation improves selection accuracy while managing evaluation complexity through modular, divide-and-conquer evaluation.
Data Source
AI summary
A method of efficiently executing a plurality of processes is described. The method generates, using a predictor, operating constraints for the processes. An operating constraint of the operating constraints is for each process of the processes. The method evaluates possible operating points for each process consistent with the operating constraints and according to a cost function for the processes. The method also determines an operating point for each process based on the evaluating.