Resource Suitability Rating for Contention Reduction
Solution Overview
Problem
In scenarios where resources are shared among multiple consumers, contention arises due to simultaneous access requests, and existing methods like exclusive access or providing multiple resources are either ineffective or resource-intensive, especially when the number of consumers exceeds the number of resources available.
Innovation Solution
A method and system that calculate a suitability rating for each resource based on recency of access, number of accesses by other consumers, and number of different consumer accesses since the last access by the requesting consumer, allocating the most suitable resource to reduce contention, and generating a new resource if the suitability rating does not meet a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If exclusive access is provided to a resource for a consumer, then contention is reduced, but resource utilization decreases when use is indeterminate
Solution Approach 1:
The patent changes the state parameter of resource allocation from binary (allocated/not allocated) to a spectrum of suitability ratings. Resources are dynamically re-evaluated based on usage patterns, allowing the system to transition resources from one consumer to another based on changing conditions, thus maintaining high utilization while reducing contention through gradual state changes rather than abrupt exclusivity switches.
Solution Approach 2:
The patent implements dynamic resource allocation where resources are not statically assigned but continuously re-evaluated based on suitability ratings. The system adapts to indeterminate usage patterns by allowing consumers to access resources based on current suitability rather than fixed exclusive assignments, enabling the system to respond dynamically to changing resource usage conditions.
2Reliability
If multiple resources are provided per consumer, then contention is eliminated, but resource overhead increases when resources are limited
Solution Approach 1:
The patent makes resources universal by allowing the same resource to serve multiple consumers sequentially based on suitability ratings. Instead of dedicating separate resources to each consumer, a single resource can be shared among multiple consumers dynamically, reducing the total quantity of resources needed while maintaining system reliability through intelligent allocation.
Solution Approach 2:
The patent changes the allocation parameter from fixed one-to-one mapping to dynamic many-to-many relationships based on suitability ratings. This allows the system to accommodate more consumers with fewer resources by continuously adjusting which consumers access which resources based on current suitability metrics, eliminating the need for excessive resource provisioning.
3Productivity
If resources are shared among multiple consumers, then resource utilization increases, but contention increases without proper management
Solution Approach 1:
The patent implements feedback mechanisms where resource usage patterns are continuously monitored and fed back into the suitability rating calculation. This feedback loop allows the system to learn from past usage and adjust allocations to minimize future contention, maintaining high utilization while using the feedback information to proactively manage and reduce contention likelihood before it occurs.
Solution Approach 2:
The patent performs preliminary evaluation of resource suitability before allocation occurs. By calculating suitability ratings in advance based on historical usage patterns and current system state, the system proactively identifies the best resource-consumer matches before contention can arise, preventing simultaneous access conflicts while maintaining high resource utilization through pre-planned allocations.
Data Source
AI summary
In a computer system including a set of resources for allocation to, and access by, a plurality of consumers, a method for allocating a resource to a requesting consumer comprising: calculating a suitability rating for each resource in the set of resources as a value in a range from least suitable to most suitable, the suitability rating being calculated from: a) a measure of recency of access to the resource; b) a measure of a number of times the resource has been accessed by a consumer other than the requesting consumer since a previous access to the resource by the requesting consumer; and c) a measure of a number of times the resource is accessed by a different consumer since the previous access to the resource by the requesting consumer; identifying a resource having a suitability rating which is closest to most suitable; and allocating the identified resource to the requesting consumer.