Flexible Partition Management for CPU Resource Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current software licensing models for computer systems are inflexible, leading to underutilization of CPU resources when workload managers allow more CPUs than allocated, resulting in unnecessary costs, and risk of compliance issues when fewer CPUs are available than licensed.
Innovation Solution
Implementing flexible partition management systems that allow resource sharing between partitions, monitored by a workload manager to ensure total CPU usage does not exceed licensed limits, with notifications and dynamic adjustments to prevent license violations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If software licensing is based on fixed CPU allocation per partition, then license compliance is ensured, but CPU resource utilization is reduced when workload managers allow dynamic CPU allocation
Solution Approach 1:
The patent implements dynamic license management where the system continuously monitors actual CPU usage across partitions and adjusts license enforcement in real-time. Instead of static CPU allocation, the workload manager dynamically determines CPU usage and the license manager dynamically enforces licensing based on actual usage, allowing partitions to temporarily exceed their allocated CPU limits when workloads require it, while ensuring overall compliance with total licensed CPU capacity.
Solution Approach 2:
The patent establishes a feedback loop where the workload manager reports actual CPU usage to the license manager, which then compares this usage against licensed capacity and enforces appropriate licensing constraints. This continuous monitoring and feedback mechanism enables the system to adapt license enforcement to actual resource consumption patterns, resolving the contradiction between fixed licensing and dynamic resource allocation.
2Adaptability or versatility
If partitions are given maximum CPUs under software license, then resource availability is improved, but cost increases when CPUs are not fully utilized
Solution Approach 1:
The patent merges the CPU resource pools of multiple partitions into a shared capacity that can be dynamically allocated based on actual workload requirements. Instead of each partition having dedicated, fixed CPU allocation, the system combines available CPU capacity across partitions and allocates it dynamically according to demand, allowing underutilized CPUs in one partition to be used by other partitions with higher workload demands.
Solution Approach 2:
The patent changes the parameter of CPU allocation from fixed static values to dynamic values that adjust based on workload conditions. The system monitors actual CPU usage patterns and adjusts the effective CPU allocation for each partition in real-time, allowing the same physical CPU resources to serve different logical partition requirements at different times, thereby improving resource availability without proportionally increasing costs.
Data Source
AI summary
Systems and methods are disclosed in which the software license for server partitions are flexible in that, as between two (or more, if desired) partitions the software licensed resources assigned to one partition can be used by the other partitions, providing a total number of resources do not exceeded the software license limit. In one embodiment, a workload manager monitors workload utilization to be sure that the total number of CPUs working on the application does not exceed the maximum under the software license. Users are notified when the workload's policy is about to be exceeded. In one embodiment, this could be a warning while in another embodiment additional software licenses can be activated, or a limit can be placed on the CPU use.