Hierarchical Domain Scheduling for Multi-Tier Computing Workloads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex distributed multi-tiered computing environments, determining efficient application provisioning across diverse devices and domains is challenging due to increased complexity and scale, necessitating improved management and scheduling strategies.
Innovation Solution
A hierarchical management approach is implemented, comprising global, domain, and device-level controllers that normalize task resource demands, classify tasks based on priority, and generate scheduling assignments using resource demand vectors and critical path analysis to optimize provisioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional scheduling methods are used in complex distributed environments, then implementation is simpler, but scheduling efficiency and productivity deteriorate due to increased complexity and scale
Solution Approach 1:
The scheduling system is divided into multiple independent controllers (global controller, domain controllers, device controllers) that operate at different hierarchical levels. Each controller manages specific aspects of scheduling, allowing the system to handle complexity through modular segmentation while maintaining overall efficiency through coordinated operation across levels.
Solution Approach 2:
The system performs preliminary actions by pre-calculating resource demand vectors, classifying tasks by priority before scheduling, and determining critical paths in advance. This allows the actual scheduling decisions to be made more efficiently based on pre-processed information, improving productivity despite environmental complexity.
2Measurement precision
If more comprehensive scheduling considerations are included, then scheduling accuracy improves, but processing time increases
Solution Approach 1:
Resource demand vectors are pre-calculated and stored for each task, critical paths are determined in advance, and tasks are classified by priority before the actual scheduling occurs. This preliminary processing allows the scheduling algorithm to access pre-computed information quickly, achieving high accuracy without excessive processing time during the actual scheduling decision.
Solution Approach 2:
The scheduling system operates continuously by maintaining updated information about resource demands, task priorities, and device capacities in the hierarchical controllers. This continuous availability of processed information allows rapid scheduling decisions to be made without repeated computation, balancing accuracy with processing time.
Data Source
AI summary
Techniques described herein relate to a method for managing a distributed multi-tiered computing (DMC) environment. The method includes normalizing, by a local controller associated with an DMC domain, task resource demand dimensions for each task associated with a scheduling job; summing the resource demand dimension for each task to generate resource demand vectors; classifying tasks based on priority; sorting tasks based on associated resource demand vectors; obtaining critical path, earliest start time, and latest start time associated with each task; sorting tasks based on critical path and earliest start time; and generating scheduling assignments based on the priority, capacity of devices in a final candidate list, resource demand vectors, earliest start time, and the critical path.