Iterative Workload Loops With Moving-Average Task Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing iterative workloads face challenges in maintaining consistent responsiveness due to varying times required for mandatory and preferred processing tasks, leading to inconsistent user experiences.
Innovation Solution
Implementing a workload processing loop that adjusts the time for the preferred processing task based on a moving average of mandatory processing task times, ensuring the preferred task is completed within a defined timeframe.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the preferred processing task is performed with a fixed time allocation, then the workload processing loop is simple to implement, but wide variations in user responsiveness occur when mandatory processing task times vary significantly
Solution Approach 1:
The preferred processing task time allocation is made dynamic by adjusting it based on the moving average of mandatory processing task times. When mandatory tasks take longer, the preferred task time is reduced proportionally, and when mandatory tasks are faster, more time is allocated to preferred tasks. This dynamic adjustment eliminates wide variations in user responsiveness while maintaining a relatively simple implementation through a straightforward time adjustment mechanism.
2Productivity
If the preferred processing task time is increased to improve processing quality, then processing completeness improves, but the total processing time exceeds the target timeframe
Solution Approach 1:
The time allocation parameter for the preferred processing task is changed dynamically based on the moving average of mandatory processing task times. The system adjusts the preferred task time parameter within the target timeframe, allocating more time when mandatory tasks are fast and less time when mandatory tasks are slow. This ensures processing completeness is maintained without exceeding the target timeframe, as the parameter adjustment is constrained by the target timeframe boundary.
Solution Approach 2:
The system uses feedback from the moving average of mandatory processing task times to adjust the preferred processing task time allocation. This feedback mechanism ensures that the total processing time remains within the target timeframe while maintaining processing completeness, as the adjustment is based on actual performance data from previous iterations.
3Loss of time
If the preferred processing task is performed with minimal time allocation, then the total processing time stays within target timeframe, but processing quality deteriorates
Solution Approach 1:
The preferred processing task time allocation dynamically adjusts based on the moving average of mandatory processing task times. When mandatory tasks complete quickly, the system allocates more time to preferred tasks to maintain processing quality. When mandatory tasks take longer, the system reduces preferred task time to stay within the target timeframe. This dynamic adjustment ensures processing quality is maintained without sacrificing time constraints.
Data Source
AI summary
Iterative workload processing having a mandatory processing task and a preferred processing task. A processor device iteratively performs a processing workload that comprises a mandatory processing task and a preferred processing task. The mandatory processing task includes accessing a plurality of input messages that have not yet been processed, and the preferred processing task has a target timeframe within which to be performed. For each iteration, a maximum preferred processing task amount of time to perform the preferred processing task is determined based on a moving average of mandatory processing task times of previous iterations and based on the target timeframe. The preferred processing task is performed for a period of time no greater than the maximum preferred processing task amount of time.