Predictive Autoscaling via Historical Utilization Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprise cloud-based computing environments face challenges in flexible and proactive resource scaling, leading to potential service interruptions or high costs due to insufficient resource provisioning, as existing methods either require expensive over-provisioning or reactive scaling.
Innovation Solution
A predictive scaling system that gathers historical utilization patterns, uses algorithms like CHAID to generate predictive usage data, and applies business rules to automatically scale computing resources proactively, enabling horizontal and vertical scaling based on predefined ranges and risk thresholds, thus avoiding reactive provisioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computing resources are over-provisioned to ensure adequate capacity, then service reliability is improved, but resource cost increases
Solution Approach 1:
The system performs preliminary scaling actions by predicting future resource usage patterns and provisioning additional computing resources before demand actually increases. This proactive approach ensures service reliability is maintained during demand spikes while avoiding the need for permanent over-provisioning, thus reducing overall resource costs.
2Quantity of substance
If computing resources are scaled reactively when demand increases, then resource cost is reduced, but service quality deteriorates due to provisioning delays
Solution Approach 1:
The system predicts future resource usage patterns and initiates scaling actions before demand actually increases, eliminating the delay inherent in reactive scaling. This ensures service quality is maintained during demand transitions while avoiding unnecessary provisioning, thus optimizing resource costs.
3Productivity
If provisioning steps are executed to accommodate increased usage demands, then service capacity is improved, but service interruption occurs during provisioning
Solution Approach 1:
The system provisions additional computing resources in advance during low-demand periods before scaling events occur. This preliminary provisioning ensures that capacity is already available when demand increases, eliminating service interruptions that would otherwise occur during emergency provisioning operations.
4Productivity
If scaling is performed quickly to meet demand, then productivity is improved, but resource allocation precision deteriorates
Solution Approach 1:
The system uses predictive analytics to determine the optimal amount of resources needed for future demand scenarios. By provisioning the precise amount of resources predicted to be needed rather than using generic scaling rules, the system achieves both fast scaling response and precise resource allocation, avoiding both over-provisioning and under-provisioning.
Data Source
AI summary
Embodiments may enable cloud based computing infrastructure to automatically scale in response to changing service demands. Auto-scaling may be enabled by automatically provisioning computing resources as they may be needed by hosted computing services. Historical utilization patterns may be tracked enabling the generation of models that may be employed to predict future computing resource requirements. The automatic scaling system may comprise one or more models that may be trainable using business rules that may be applied to determine to if and how computing resources are scaled. Further, business rules may be arranged to determine provisioning and scaling of computing resources based in part on the historical usage patterns of the computing services.