Diagonal Scaling Prioritization in Distributed Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed computing systems face inefficiencies in resource allocation and scaling, as they often rely on predefined targets and separate resource and application management, leading to misalignment with actual application requirements and increased costs.
Innovation Solution
Implementing a unified algorithmic mechanism for automatic diagonal scaling that combines vertical and horizontal scaling, prioritizing applications based on significance and dependencies, and dynamically adjusting resources according to actual workload and availability, without pre-defined targets or additional technologies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If existing distributed computing systems use predefined targets and separate resource and application management, then resource allocation can be simplified, but resource utilization efficiency deteriorates due to misalignment with actual application requirements
Solution Approach 1:
The patent merges separate resource management and application management into a unified diagonal scaling mechanism. The system simultaneously scales both resources and application instances based on actual workload requirements, eliminating the misalignment caused by separate management approaches while maintaining operational simplicity through automated priority-based decision making.
Solution Approach 2:
The system implements dynamic scaling by continuously monitoring actual application workload and automatically adjusting both resource allocation and application instance count in real-time. This dynamic approach allows the system to adapt to changing requirements without predefined targets, optimizing resource utilization while maintaining simplicity through automated control.
2Productivity
If existing systems focus on horizontal scaling by adding application instances, then application capacity can be increased, but resource efficiency deteriorates due to lack of vertical scaling optimization
Solution Approach 1:
The patent combines vertical scaling (resource optimization) and horizontal scaling (application instance management) into a unified diagonal scaling approach. The system first optimizes resource allocation vertically for existing instances, then determines whether additional instances are needed, ensuring both capacity increase and resource efficiency without wasting resources on suboptimally configured instances.
Solution Approach 2:
The system dynamically selects between vertical and horizontal scaling actions based on real-time workload analysis. When resources can be better utilized through vertical scaling, the system adjusts resource allocation; when additional capacity is needed, it horizontally scales. This dynamic decision-making optimizes both application capacity and resource efficiency.
3Extent of automation
If systems use pre-defined utilization targets for scaling, then scaling decisions can be made automatically, but alignment with actual quality of service requirements deteriorates due to arbitrary target selection
Solution Approach 1:
Instead of setting predefined utilization targets and scaling to meet them, the system inverts the approach by directly analyzing actual application workload and quality of service requirements to determine scaling actions. This inversion eliminates the arbitrary target selection problem while maintaining automated decision-making through workload-based algorithms that directly prioritize applications based on their actual needs.
4Device complexity
If systems manage resources and applications separately, then management complexity can be reduced, but overall system optimization deteriorates due to lack of coordinated scaling
Solution Approach 1:
The patent merges resource management and application management into a single coordinated diagonal scaling system. The unified mechanism simultaneously optimizes both resource allocation and application instance deployment based on workload requirements, achieving system-wide optimization without significantly increasing management complexity through automated priority-based decision making and integrated control logic.
Data Source
AI summary
Embodiments for prioritizing applications for diagonal scaling operations in a distributed computing environment. A significance value of an application of a plurality of applications is defined, the significance value representing an importance of the application or the functionality performed by the application, and dependencies between the plurality of applications are configured. A significance value of a dependency of a dependent application on an antecedent one of the plurality of applications is defined, and priorities for each of the plurality of applications are computed based on the significance values of each of the plurality of applications and respective dependencies therebetween for performing the diagonal scaling operations of resources allocated to each of the applications in the distributed computing environment.