Hypervisor Dynamic Processor Replacement in Logical Partitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computer systems with multiple logical partitions, replacing a failing physical processor without unused spare processors is challenging, leading to undesirable outcomes such as partition termination or degraded performance, as existing methods lack efficient mechanisms to dynamically allocate resources and maintain high availability.
Innovation Solution
A method where a hypervisor assigns priorities to dedicated partitions and pools of virtual processors, detects a failing processor, retrieves its state, and replaces it with a physical processor from the partition or pool with the lowest priority, ensuring seamless transition and maintaining system performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a physical processor fails and no unused spare processors are available, then the system must either terminate the partition or run it with fewer processors, but this leads to partition termination or degraded performance
Solution Approach 1:
The patent implements dynamic processor allocation where the hypervisor continuously monitors processor status and dynamically reassigns processors from lower-priority partitions to higher-priority partitions experiencing failures. This dynamic adjustment allows the system to adapt to processor failures without terminating partitions, resolving the contradiction between maintaining system availability and preserving partition performance.
Solution Approach 2:
The system changes the priority parameter of partitions dynamically based on processor availability and failure conditions. When a processor fails, the hypervisor adjusts the priority levels of different partitions and reassigns processors accordingly, allowing critical partitions to maintain performance while less critical partitions temporarily operate with reduced resources.
2Reliability
If unused processors are kept as spares for replacement, then processor failure recovery is enabled, but this results in wasted and expensive unused processors
Solution Approach 1:
The patent makes all processors universal and interchangeable, allowing any processor to serve any partition based on priority needs rather than dedicating specific processors as spares. This multi-functionality enables processors to be dynamically reassigned from lower-priority to higher-priority partitions, eliminating the need for dedicated spare processors while maintaining full recovery capability.
Solution Approach 2:
The system implements self-service through automated hypervisor monitoring and dynamic reassignment. When a processor fails, the hypervisor automatically detects the failure, determines priority levels of affected partitions, and reassigns processors without manual intervention or pre-configured spare pools, eliminating resource waste while ensuring recovery.
3Ease of operation
If random chance selects which partition suffers when a processor checkstops, then simple replacement is possible, but this is not an optimal procedure as it may affect critical partitions
Solution Approach 1:
The hypervisor implements feedback mechanisms by continuously monitoring partition priorities, processor status, and system load. When a processor fails, the system uses this feedback to make informed decisions about which partitions to maintain and which to temporarily deprioritize, ensuring critical partitions receive priority for processor reassignment rather than relying on random selection.
Solution Approach 2:
The system performs preliminary actions by pre-establishing priority levels for different partitions before failures occur. This preliminary configuration enables the hypervisor to immediately act optimally when failures happen, knowing which partitions are critical and require priority protection, thus avoiding random selection and ensuring critical partition availability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Replacing a failing physical processor in a computer supporting multiple logical partitions, where the logical partitions include dedicated partitions and shared processor partitions, the dedicated partitions are supported by virtual processors having assigned physical processors, and the shared processor partitions are supported by pools of virtual processors. The pools of virtual processors have assigned physical processors. Embodiments operate generally by assigning priorities to the dedicated partitions and to the pools of virtual processors; detecting a checkstop of a failing physical processor; retrieving the failing physical processor's state; replacing by a hypervisor the failing physical processor with a replacement physical processor assigned to a dedicated partition or pool, which dedicated partition or pool has the lowest priority among the priorities of the dedicated partitions and pools; and assigning the retrieved state of the failing physical processor as the state of the replacement physical processor.