Meta-Scheduler Package Deployment Across Distributed Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional software lifecycle management methods are inadequate for distributed systems executing on multiple computers, complicating tasks such as reliably distributing artifacts, launching, and monitoring software processes.
Innovation Solution
A system utilizing a metascheduler and subschedulers to manage software package distribution across a cluster of computers, incorporating multi-tiered fault detection and correction, and performance monitoring with remedial actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional software lifecycle management methods are used, then management processes are simple, but they become inadequate for distributed systems with multiple computers
Solution Approach 1:
The patent segments the package management system into multiple independent components: a metascheduler that coordinates across the distributed system and individual subschedulers that manage packages on specific machines. This segmentation allows the system to handle distributed environments while keeping each component's logic relatively simple and manageable.
Solution Approach 2:
The metascheduler acts as an intermediary between the package management system and individual subschedulers distributed across multiple computers. It coordinates package installation, monitoring, and relaunching across the distributed system without requiring direct complex interactions between all system components, thereby reducing overall system complexity while improving adaptability.
2Reliability
If distributed package management is implemented, then reliability of software distribution improves, but fault detection and correction becomes more complex
Solution Approach 1:
The fault detection system is segmented into multiple independent monitoring points: the metascheduler monitors package status across distributed systems, while individual subschedulers monitor local package installations. Each component independently detects and reports failures, simplifying the overall fault detection architecture while maintaining high reliability through distributed monitoring.
Solution Approach 2:
The system implements multi-tiered feedback mechanisms where subschedulers provide feedback to the metascheduler about package installation status and failures. The metascheduler uses this feedback to coordinate remedial actions across the distributed system, enabling reliable fault correction without requiring complex centralized monitoring and control.
3Reliability
If automated remedial actions are added, then system reliability improves, but automation extent increases complexity
Solution Approach 1:
Automated remedial actions are segmented and executed by individual subschedulers rather than requiring complex centralized automation. Each subscheduler independently executes remedial actions for failures detected on its local machine, reducing automation complexity while maintaining system reliability through distributed automated response.
Solution Approach 2:
The system implements self-service automation where subschedulers automatically detect and execute remedial actions for package failures without requiring manual intervention or complex centralized control. The metascheduler coordinates these automated actions across the distributed system, improving reliability while keeping automation mechanisms simple and maintainable.
Data Source
AI summary
A system for package management includes an interface and a processor. The interface is to receive an indication to install a package. The processor is to determine a configured package using a set local configuration properties and using the package and to launch, using a metascheduler, a set of subschedulers to install a plurality of applications of the configured package.