Partition Resource Fault Management via Fabric Interconnection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The isolation between partitions in partitionable computer systems, which enhances flexibility and security, complicates resource fault management, making it difficult to handle resource failures and maintain system robustness.
Innovation Solution
The implementation of management logic that includes a partition management tool and a fault management tool, which enables automated resource reassignment, online resource adjustments, and priority-based fault handling, allowing for dynamic resource allocation and failure detection, analysis, and response.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If partition isolation is implemented to enhance flexibility and security, then system security and flexibility are improved, but resource fault management becomes difficult
Solution Approach 1:
The patent introduces a fabric interconnection layer as an intermediary between partitions and resources. This fabric provides standardized interfaces and protocols that enable resource fault management across partition boundaries without compromising isolation. The fabric acts as a mediator that translates between partition-specific resource views and the underlying physical resources, resolving the management difficulty while preserving security.
Solution Approach 2:
The patent implements a universal resource management architecture where the fabric provides multi-functional capabilities including resource allocation, fault detection, and recovery mechanisms that work across all partitions. This universal approach allows the same management logic to handle resource faults in any partition, reducing complexity while maintaining the isolation benefits.
2Adaptability or versatility
If resources are carved into isolated partitions, then flexibility is improved, but resource reassignment during failures becomes difficult
Solution Approach 1:
The patent implements dynamic resource reassignment capabilities through the fabric, allowing resources to be reallocated between partitions based on failure conditions and priority rules. The system can dynamically adjust resource allocation without requiring manual intervention or system reconfiguration, maintaining flexibility while enabling easy resource reassignment during failures through automated management logic.
3Reliability
If automated resource reassignment is implemented, then system stability is improved, but management logic complexity increases
Solution Approach 1:
The patent implements self-service mechanisms where the fabric and management logic automatically detect resource failures, determine affected partitions, and reassign resources based on pre-configured priority rules without requiring external intervention. This automation improves system stability by ensuring rapid failure response while the modular fabric architecture keeps the management logic complexity manageable through standardized interfaces.
Data Source
AI summary
In accordance with at least some embodiments, a system includes a plurality of partitions, each partition having its own operating system (OS) and workload. The system also includes a plurality of resources assignable to the plurality of partitions. The system also includes management logic coupled to the plurality of partitions and the plurality of resources. The management logic is configured to set priority rules for each of the plurality of partitions based on user input. The management logic performs automated resource fault management for the resources assigned to the plurality of partitions based on the priority rules.