Dependency Filter Object for Server Cluster Resource Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current server cluster systems lack flexibility in managing dependencies between resource objects, particularly in handling failures and redundancy, as they do not support dependent resources specifying redundant dependencies on multiple provider resources.
Innovation Solution
The introduction of a dependency filter object that allows for the description of combined effects of dependencies, enabling dependent resources to be brought online if one or more provider resources are available, and specifying actions based on the state changes of provider resources, such as bringing resources online or offline.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a dependent resource is configured to depend on a single provider resource, then the dependency relationship is simple and easy to manage, but the system lacks fault tolerance and redundancy capabilities
Solution Approach 1:
The patent segments the dependency relationship by introducing a dependency filter object that independently manages each dependency condition. This allows the dependent resource to have multiple provider resources with different dependency filters, enabling fault tolerance while keeping each individual dependency relationship simple and manageable through modular filtering mechanisms.
Solution Approach 2:
The dependency filter object acts as an intermediary between the dependent resource and multiple provider resources. It mediates the complexity by consolidating multiple dependency conditions into a single filter object that evaluates whether to allow the dependent resource to start or continue based on the availability of provider resources, thus improving fault tolerance without overwhelming complexity.
2Productivity
If a dependent resource must wait for all provider resources to be online before coming online, then resource initialization is reliable and ordered, but the time to bring resources online increases and productivity decreases
Solution Approach 1:
The patent applies partial action by allowing the dependent resource to start when only a subset of provider resources are available, rather than requiring all providers to be online. The dependency filter object evaluates which provider resources are necessary and allows the dependent resource to initialize with partial provider availability, reducing startup time while maintaining reliability through selective dependency satisfaction.
3Adaptability or versatility
If the cluster automatically terminates dependent resources when provider resources fail, then failure propagation is handled automatically, but the system lacks flexibility and adaptability in handling failures
Solution Approach 1:
The patent introduces dynamics by allowing dependency filters to be configurable and adaptable rather than fixed. The dependency filter object can be dynamically configured to specify different failure propagation behaviors, such as allowing the dependent resource to continue operating with reduced functionality or to terminate based on specific conditions, thus providing flexibility while maintaining automated failure handling capabilities.
Data Source
AI summary
A dependency filter object for use in a server cluster is described. In an implementation, a method includes bringing online one or more resource objects included in a dependency tree. Each of the resource objects corresponds to a resource available in a server cluster. The dependency tree includes a dependency filter object that specifies a dependency of a first one of the resource objects on a second one of the resource objects. The bringing online of the first resource object is managed according to the specified dependency on the second resource object.


