Dynamic VM Resource Allocation for Cloud Server Recovery

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Cloud computing systems face performance degradation due to abnormal events like hardware failures, software failures, and scheduled updates, which require additional computing resources to mitigate downtime and restore performance efficiently.

Innovation Solution

A resource management module dynamically allocates additional computing resources to affected partitions by activating dormant processor cores, increasing processing capacity, and reallocating resources from other systems within a server cluster to ensure uninterrupted service and rapid recovery.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If additional computing resources are allocated to a partition experiencing performance degradation, then the recovery speed and mitigation of downtime improve, but the device complexity and resource management overhead increase

Engineering Contradiction:

Improveservice continuityVSAvoidresource management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system dynamically adjusts computing resource allocation based on real-time performance monitoring. When performance degradation is detected, the system automatically allocates additional computing resources to the affected partition. This dynamic adaptation allows the system to respond to changing conditions without manual intervention, maintaining service continuity while managing complexity through automation.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements a feedback mechanism that continuously monitors partition performance and triggers resource allocation decisions based on detected performance degradation. The performance monitoring system provides feedback to the resource management module, which then adjusts resource allocation accordingly. This closed-loop control enables automatic recovery while managing system complexity through structured feedback processing.

Inventive Principle:

Principle #23Feedback

2Productivity

If computing resources are dynamically reallocated from other systems within the server cluster, then the productivity and recovery speed improve, but the reliability of other partitions may be compromised

Engineering Contradiction:

Improverecovery speedVSAvoidservice stability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system applies partial resource reallocation rather than complete resource transfer. When performance degradation is detected, the system allocates additional computing resources to the affected partition by activating dormant processor cores or reallocating a portion of resources from other partitions. This partial action approach accelerates recovery while maintaining sufficient resources in other partitions to preserve their service stability.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system maintains a pool of dormant or underutilized computing resources that can be quickly activated when performance degradation occurs. This beforehand preparation ensures that resources are available for rapid reallocation without requiring complete resource transfer from operational partitions, thus maintaining both productivity and reliability simultaneously.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

3Quantity of substance

If dormant processor cores are activated to increase processing capacity, then the quantity of computing resources improves, but the energy consumption increases

Engineering Contradiction:

Improvecomputing resourcesVSAvoidprocessor energy consumption

Core Design Contradiction:

Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

The system pre-configures dormant processor cores in a low-power state rather than completely shutting them down. These cores can be quickly activated when additional computing resources are needed, avoiding the energy-intensive process of cold booting. This preliminary preparation enables rapid resource scaling while minimizing energy consumption during idle periods.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes the operational state of processor cores between active and dormant modes based on demand. When performance degradation is detected, the system transitions dormant cores to active state, changing their operational parameters from low-power to full-power mode. This parameter change approach allows flexible resource scaling while managing energy consumption through state transitions rather than continuous high-power operation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10884818B2Increasing processing capacity of virtual machines

Publication Date: 2021.01.05 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

According to one or more embodiments of the present invention a computer-implemented method includes allocating, for a partition from a plurality of partitions of a computer server, a first set of computing resources of the computer server. The method further includes receiving a request to increase processing capacity for the partition. The method further includes in response, allocating a second set of computer resources, in addition to the first set of computer resources, for the partition. The method further includes operating the partition using the first set of computing resources and the second set of computing resources.