Cloud System Management for Job Deadline Compliance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud-native systems lack the ability to select countermeasures for performance failures in a way that considers the influence on ongoing jobs, potentially causing batch jobs to miss their termination deadlines.
Innovation Solution
A system management device and method that evaluates and selects countermeasures for performance failures in a cloud environment, ensuring that the chosen countermeasure allows jobs to complete by their associated execution deadlines, by creating and assessing the impact of countermeasures on job execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional technology automatically executes selected countermeasures for performance failures, then the system can respond automatically to performance degradation, but the batch job may fail to meet the termination deadline due to lack of consideration for job impact
Solution Approach 1:
The system performs preliminary evaluation of countermeasure impact on jobs before automatic execution. The evaluation unit assesses each countermeasure's potential effect on job execution time and deadline compliance in advance, allowing the selection unit to choose countermeasures that maintain deadline compliance while enabling automatic response.
Solution Approach 2:
The system incorporates feedback mechanisms where the evaluation unit continuously monitors job status and countermeasure impact, providing feedback to the selection unit. This feedback loop ensures that automatically executed countermeasures are those that maintain job deadline compliance, resolving the contradiction between automation and reliability.
2Adaptability or versatility
If multiple countermeasures are created for performance failures, then the system has more options to address performance issues, but the complexity of selecting the appropriate countermeasure increases
Solution Approach 1:
The evaluation unit automatically assesses each countermeasure's impact on jobs without manual intervention. The system self-evaluates countermeasure feasibility and impact, reducing selection complexity while maintaining multiple countermeasure options. The evaluation process is performed automatically based on predefined criteria and job status.
Solution Approach 2:
The system changes evaluation parameters dynamically based on current job status and performance failure conditions. The evaluation unit adjusts assessment criteria according to real-time system state, making the selection process simpler while maintaining adaptability. This parameter adaptation allows the system to handle multiple countermeasure options without increasing operational complexity.
Data Source
AI summary
A system management device manages a monitored system. When a performance failure occurs in the monitored system, the system management device creates a plurality of countermeasures for the performance failure. The system management device evaluates, for each of the plurality of countermeasures, the influence of the countermeasure on the execution of a job having an execution deadline associated with the countermeasure. The system management device selects the countermeasure to be executed on the monitored system from out of the plurality of countermeasures based on the evaluation results.