Multi-Tier Application Scheduling with Dynamic Policy Priorities
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 involving a global controller, local controllers, and endpoint controllers, which selects scheduling policies and priorities based on workload information and user preferences to efficiently provision applications across target domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If applications are provisioned in complex distributed multi-tiered computing environments with more devices and data, then the ecosystem capabilities and resource availability increase, but it becomes difficult to determine where to provision applications and how to efficiently schedule them
Solution Approach 1:
The system segments the complex distributed computing environment into multiple target domains, each with its own local controller. The global controller divides application provisioning tasks into domain-specific sub-tasks, making the overall complex provisioning problem manageable through hierarchical decomposition into smaller, independently manageable domain-level scheduling problems
Solution Approach 2:
The global controller acts as an intermediary between application requests and local domain controllers. It receives application deployment information, identifies suitable target domains, selects appropriate scheduling policies, and generates scheduling packages that are then executed by local controllers. This intermediary layer abstracts the complexity from both the application requester and the individual domain controllers
2Productivity
If scheduling policies are applied to manage application provisioning, then resource allocation efficiency improves, but the system requires dynamic selection and management of multiple scheduling policies across different domains
Solution Approach 1:
The system dynamically selects scheduling policies based on workload information and application deployment requirements. Instead of using a fixed scheduling approach, the global controller adapts the scheduling policy selection to match current system conditions and application needs, allowing the scheduling behavior to change dynamically in response to varying workloads and domain states
Solution Approach 2:
The system changes scheduling parameters (policy selection and priorities) based on workload information and application deployment characteristics. By adjusting scheduling policy parameters dynamically according to system state and application requirements, the system optimizes resource allocation efficiency for different scenarios without requiring manual reconfiguration
Data Source
AI summary
Techniques described herein relate to a method for managing a distributed multi-tiered computing (DMC) environment. The method includes obtaining, by a global controller, a request from a user, and the request is associated with scheduling an application in the DMC environment; and in response to obtaining the request: identifying application tasks associated with the request; obtaining application deployment information based on a manifest included in the request; identifying target domains for the application tasks based on the application deployment information; obtaining workload information associated with the target domains; selecting priorities for scheduling policies for the target domains based on the application deployment information and the workload information; generating scheduling packages based on the scheduling policy priorities for each target domain; and providing the scheduling packages to local controllers of the target domains, wherein the local controllers schedule the application tasks using the scheduling packages.