Cluster Binary Module Patching via Graph Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computing clusters, especially in Oracle Real Application Clusters environments, patching binary modules often results in downtime due to the need for multiple attempts to apply patches across shared resources, leading to application outages.
Innovation Solution
A method and system that identify resources and binary modules within a computing cluster using a logical graph representation, allowing for the grouping of connected components and simultaneous patching of all modules within a group to minimize downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated patching products perform multiple attempts to patch the same binary module, then patching reliability is improved, but downtime increases
Solution Approach 1:
The patent segments the cluster resources into distinct groups based on shared binary modules. Each group is patched independently in parallel, allowing the system to achieve reliable patching of all modules while reducing total downtime through concurrent operations across segmented groups.
Solution Approach 2:
The patent performs preliminary actions by identifying and grouping resources that share binary modules before the patching process begins. This pre-grouping enables optimized patch execution where all modules in a group are patched together in a single attempt, eliminating the need for multiple retry attempts.
2Manufacturing precision
If resources are shut down for offline patching, then patching completeness is improved, but application availability deteriorates
Solution Approach 1:
The patent divides resources into multiple groups based on their shared binary module dependencies. By patching each group separately and in parallel, the system achieves complete patching of all modules while minimizing the time any single resource group is unavailable, thus maintaining overall application availability.
Solution Approach 2:
The patent implements dynamic patching where resources are patched in parallel across multiple groups rather than sequentially shutting down all resources. This dynamic approach allows the system to maintain service availability while ensuring complete patching coverage across all binary modules.
3Manufacturing precision
If multiple attempts are made to patch binary modules, then patching thoroughness is improved, but productivity deteriorates
Solution Approach 1:
The patent segments binary modules into groups based on resource sharing relationships. By patching each group thoroughly in a single parallel attempt, the system achieves comprehensive patching coverage without the need for multiple sequential retry attempts, thereby improving patching efficiency and productivity.
Solution Approach 2:
The patent ensures continuous useful action by patching multiple groups of binary modules in parallel simultaneously. This approach maintains high productivity while achieving thorough patching coverage, as the system continuously patches different groups without idle retry periods.
Data Source
AI summary
Embodiments of the invention provide systems and methods for applying a patch or a change to a set of binary modules executing on and used by resources of a computing cluster while reducing the downtime of those resources. According to one embodiment, applying a patch to a plurality of binary modules on a computing cluster can comprise identifying resources on one or more of a plurality of nodes of the cluster. Additionally, each of the plurality of binary modules on one or more of the nodes of the cluster can be identified. A graph can be generated logically representing the cluster. Groups within the binary modules and resources can be identified based on the graph. Patches can be applied to the binary modules based on the identified groups. Patching binary modules based on the identified groups can comprise patching all of the binary modules of an identified group together.