Backup Job Scheduling via Dynamic Load Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Backup systems face inefficiencies due to non-uniform utilization of media resources, leading to performance bottlenecks and scalability issues, particularly during peak user periods when backups interfere with application performance.
Innovation Solution
A predictive approach that calculates the current load of media resources and schedules new backup jobs to optimize resource utilization, allowing for the adjustment of start times and re-scheduling of existing jobs to maintain peak usage below 100%, ensuring uniform resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If backup jobs are scheduled during peak user periods, then backup completion time is reduced, but application performance and user response times deteriorate
Solution Approach 1:
The system dynamically adjusts backup scheduling based on real-time media resource load conditions. Instead of fixed scheduling, the backup job start time is determined by calculating current load and predicting future load, allowing the system to adapt to changing conditions and find optimal windows that balance backup speed with application performance requirements
Solution Approach 2:
The system performs preliminary calculation of current media resource load and prediction of future load before scheduling backup jobs. This advance analysis allows the system to proactively identify optimal scheduling windows that prevent performance degradation before it occurs, rather than reactively adjusting after problems arise
2Productivity
If media resources are utilized at 100% load, then backup efficiency is maximized, but system reliability and scalability deteriorate due to bottlenecks
Solution Approach 1:
The system implements a feedback mechanism by continuously monitoring media resource load and using this information to adjust backup scheduling decisions. The load calculation and prediction results feed back into the scheduling algorithm, creating a closed-loop system that maintains optimal utilization without exceeding capacity thresholds, thereby preventing bottlenecks while maximizing efficiency
Solution Approach 2:
The system changes the utilization parameter from fixed 100% capacity to a dynamic value based on calculated current load and predicted load. By adjusting the effective utilization threshold according to real-time conditions, the system optimizes backup efficiency while maintaining reliability and preventing system bottlenecks
3Productivity
If uniform utilization of media resources is achieved through experimentation, then resource allocation is optimized, but system complexity and administrative burden increase
Solution Approach 1:
The system performs self-service by automatically calculating media resource load, predicting future load, and determining optimal backup scheduling without requiring administrator intervention. The automated load-based scheduling eliminates the need for administrators to conduct time-consuming experimentation to achieve uniform resource utilization, while the system independently optimizes its own operation
Data Source
AI summary
Method for scheduling a new backup job within a backup application to optimize a utilization of a media resource of said backup application. The backup application includes one or more previously scheduled backup jobs. The backup application calculates a current load of the media resource as a function of the previously scheduled backup jobs and the media resource and predicts a load value for the new backup job as a function of job parameters associated with the new backup job. Then, the backup application schedules the new backup job as a function of the calculated current load and the predicted load value such that the resulting load on the media resource will yield a minimum peak percentage utilization of the media resource. Alternatively, the backup application schedules the new backup job and previously scheduled backup jobs as function of the calculated current load and the predicted load value such that the resulting load on the media resource will yield a minimum peak percentage utilization of the media resource.