Optimized Schedule for Distributed Computer Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Designing an optimized schedule for time-controlled distributed computer systems is challenging due to the complexity of assigning and coordinating structural and functional elements, especially when starting from an incomplete or invalid initial solution.
Innovation Solution
The method involves recognizing and classifying dependencies between elements, assigning them to dependence classes, and optimizing the schedule by coordinating elements within these classes, allowing for iterative and flexible optimization without requiring a complete initial schedule, using both deterministic and stochastic methods to evaluate and improve the schedule based on user-defined quality criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a complete initial schedule is required for optimization, then the optimization process can start from a valid solution, but the method cannot handle incomplete or invalid initial solutions and requires complex worst case execution time analysis
Solution Approach 1:
The patent segments the schedule optimization into two distinct phases: coordination phase (for elements with dependencies) and optimization phase (for remaining elements). This segmentation allows the system to handle incomplete initial solutions by first establishing valid coordination relationships among dependent elements, then optimizing the remaining schedule without requiring a complete initial schedule or complex WCET analysis.
Solution Approach 2:
The patent performs preliminary coordination of dependent elements before full optimization. By identifying and coordinating elements with dependencies in advance, the system ensures schedule validity is established early, allowing subsequent optimization to proceed without requiring a complete initial schedule or complex validation mechanisms.
2Productivity
If all elements are coordinated simultaneously for schedule optimization, then comprehensive optimization is achieved, but the computational complexity and processing time increase significantly
Solution Approach 1:
The patent divides the optimization process into two time-efficient phases: a coordination phase that handles only dependent elements, and an optimization phase that handles remaining elements. This segmentation reduces processing time by avoiding simultaneous coordination of all elements, while still achieving comprehensive optimization through the sequential application of both phases.
Solution Approach 2:
The patent applies partial coordination to only those elements that have dependencies, rather than coordinating all elements equally. This partial action approach optimizes processing time by focusing computational resources only where necessary, while still achieving sufficient overall schedule optimization through the subsequent optimization phase.
3Reliability
If strict time scheduling is enforced for all elements, then time behavior is guaranteed, but flexibility in schedule creation and adaptability to different scenarios is reduced
Solution Approach 1:
The patent applies strict time scheduling locally only to elements with dependencies that require coordination, rather than enforcing uniform strict scheduling on all elements. This local application of time guarantees maintains reliability where needed while preserving flexibility for other elements, allowing the schedule to adapt to different scenarios without compromising critical timing requirements.
Solution Approach 2:
The patent creates a dynamic scheduling approach where the strictness of time control varies by element. Elements with dependencies receive strict time guarantees through coordination, while other elements allow more flexible scheduling. This dynamic differentiation maintains time behavior guarantees for critical elements while providing adaptability for the overall schedule.
Data Source
AI summary
A method is described and presented for creation of an optimized schedule for execution of a functionality by means of a time-controlled distributed computer system, in which the distributed computer system and the functionality have a set of (especially structural and functional) elements of at least one element class and the elements are at least partially in a dependence. The method according to the invention, in which the task is solved, is initially and essentially characterized by the fact that the dependences between the elements are recognized, classified and the elements are assigned to corresponding dependence classes, and that optimization of schedule occurs by coordination of elements of at least one dependence class.