Shared Peak Resource Prediction for Cloud Workload Density
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based computing platforms face inefficiencies in resource allocation, leading to low utilization and increased costs due to overestimation of resource needs for computing workloads, resulting in unused resources and reduced density.
Innovation Solution
A shared estimator approach that predicts peak resource usage for the entire computing environment, allowing for dynamic resource allocation and reallocation based on actual needs, rather than peak workload requirements, to optimize resource utilization and density.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If resource requests are configured based on manual estimation of peak workload requirements, then reliability is improved to prevent resource starvation, but resource utilization deteriorates due to overestimation
Solution Approach 1:
The system continuously monitors actual resource usage patterns of computing workloads and uses this feedback to dynamically adjust resource allocations. The resource allocation module receives real-time usage data and modifies allocations accordingly, transitioning from static manual estimates to dynamic, data-driven decisions that balance reliability and efficiency.
Solution Approach 2:
The patent changes the parameter of resource allocation from fixed manual estimates to dynamic values based on actual workload behavior. By monitoring and adapting allocations based on real usage patterns, the system optimizes the balance between ensuring sufficient resources (reliability) and avoiding excess allocation (resource wastage).
2Reliability
If resource requests are manually configured with overestimation to ensure smooth operation, then reliability is improved, but productivity deteriorates due to low resource utilization
Solution Approach 1:
The system implements continuous monitoring of actual resource consumption and uses this feedback to adjust allocations dynamically. This closed-loop approach ensures that resources are allocated efficiently based on real usage patterns, improving both productivity through better utilization and reliability through adaptive resource management.
Solution Approach 2:
The patent transitions from static manual resource configuration to dynamic, automated allocation that adapts to changing workload conditions. The resource allocation module continuously adjusts resource distribution based on actual usage patterns, enabling the system to optimize both productivity and reliability in response to real-time conditions.
3Reliability
If resource requests are based on peak workload requirements, then reliability is improved to prevent resource starvation, but device complexity increases due to manual configuration needs
Solution Approach 1:
The system performs self-adjustment by automatically monitoring its own resource usage patterns and allocating resources accordingly. The resource allocation module uses built-in monitoring data to autonomously optimize allocations, eliminating the need for complex manual configuration while maintaining reliable resource sufficiency.
Solution Approach 2:
The patent implements a feedback mechanism where the system continuously monitors actual resource consumption and automatically adjusts allocations based on this data. This self-service approach with feedback loops simplifies configuration complexity by replacing manual tuning with automated, data-driven decision-making that maintains reliability.
Data Source
AI summary
The techniques disclosed herein enable systems to efficiently allocate computing resources for various computing workloads through a shared peak resource usage prediction. To achieve this, a predictive model analyzes a historical dataset defining resource usage of a computing environment for a past timeframe, and calculates a peak environment resource usage for a future or current timeframe. In addition, the predictive model estimates a peak number of computing workloads for the computing environment. Using the peak resource usage and/or the peak number of computing workloads, the system derives resource requests for allocating computing resources to a plurality of computing workloads. The computing workloads are subsequently assigned to computing nodes within the computing environment for execution. Furthermore, computing workloads within a computing node are configured to share computing resources to accommodate sudden surges in demand. In this way, the system can reduce computing resource wastage for computing environments.