Logical Partition Resource Allocation for High Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High availability solutions for micro-partitions in computing systems require significant processing resources for failover support, leading to resource inefficiencies and performance overhead, as secondary logical partitions must match primary resources even when not fully utilized.
Innovation Solution
A method and system for managing logical partitions that relocate from a primary to a secondary shared processor pool, allocating a smaller quantity of processing units to the secondary pool, with reserved units being dynamically redistributed among multiple partitions, allowing for reduced resource usage and efficient failover support.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If secondary logical partitions are allocated the same quantity of processing units as primary partitions, then failover support reliability is improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent changes the parameter of processing unit allocation for secondary partitions, allocating fewer processing units to secondary partitions compared to primary partitions. This parameter change enables the system to maintain failover capability while reducing resource consumption, directly resolving the contradiction between reliability and resource efficiency
Solution Approach 2:
The patent implements dynamic resource allocation where processing units can be relocated and reallocated based on system conditions. When a primary partition fails, its processing units are dynamically reassigned to secondary partitions that need them, allowing the system to adapt resource distribution to actual needs rather than maintaining static over-provisioning
2Reliability
If processing units are reserved for multiple secondary logical partitions, then failover support capability is improved, but available resources for other purposes deteriorate
Solution Approach 1:
The patent implements dynamic reservation where processing units are reserved for secondary partitions only when needed for failover. The partition manager continuously monitors system state and adjusts reservations in real-time, allowing reserved units to be released and made available for other purposes when failover is not required, thus resolving the contradiction between reliability and productivity
3Loss of energy
If secondary logical partitions are allocated smaller quantity of processing units, then resource utilization efficiency is improved, but failover performance deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-positioning state information and processing units in secondary partitions before failover occurs. The partition manager prepares secondary partitions with necessary resources and state data in advance, so when failover is needed, the transition can occur quickly without performance degradation, thus resolving the contradiction between resource efficiency and failover performance
Data Source
AI summary
A partition manager relocates a logical partition from a primary shared processor pool to a secondary shared processor pool in response to a predetermined condition, such as a hardware failure. The relocated logical partition is allocated a smaller quantity of processing units from the secondary pool than it was allocated from the primary pool. A quantity of processing units reserved for a second logical partition is identified in the secondary shared processor pool, and a portion of those reserved processing units are allocated to the relocated logical partition. The reserved processing units may be redistributed among multiple relocated logical partitions.