Resource Allocation for Virtual Infrastructure Isolation and Affinity

Solution Overview

Problem

Current infrastructure management systems fail to provide a deployment policy that balances both isolation and affinity requirements for virtual resources, which are crucial for ensuring the reliability and performance of application programs, particularly in telecom systems with advanced telecom computing architecture (ATCA) standards.

Innovation Solution

A method and apparatus for resource allocation by an infrastructure management node that ranks physical resources by isolation and affinity levels, determining suitable resources to create virtual resources that meet the deployment requirements of service applications, including options for host, chassis, rack, site, and geographic disaster recovery, ensuring higher isolation and affinity levels are met.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If virtual resources are deployed on the same physical resource to achieve high affinity, then service performance is improved, but reliability deteriorates due to lack of isolation

Engineering Contradiction:

Improveservice performanceVSAvoidservice reliability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent segments the physical resource hierarchy into multiple levels (host, chassis, rack, site, geographic disaster recovery) and assigns different segments to different virtual resources. This allows affinity to be maintained at lower levels (same host) while isolation is guaranteed at higher levels (different chassis/rack/site), simultaneously achieving high performance and reliability

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a multi-dimensional resource deployment model by adding hierarchical levels beyond the traditional single-host dimension. Virtual resources can be deployed across multiple dimensions (host→chassis→rack→site→geographic), enabling the system to satisfy both affinity (same dimension) and isolation (different dimensions) requirements

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Reliability

If virtual resources are deployed on different physical resources to achieve high isolation, then reliability is improved, but service performance deteriorates due to reduced affinity

Engineering Contradiction:

Improveservice reliabilityVSAvoidservice performance

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent segments the physical resource hierarchy into multiple levels (host, chassis, rack, site, geographic disaster recovery) and assigns different segments to different virtual resources. This allows affinity to be maintained at lower levels (same host) while isolation is guaranteed at higher levels (different chassis/rack/site), simultaneously achieving high performance and reliability

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a multi-dimensional resource deployment model by adding hierarchical levels beyond the traditional single-host dimension. Virtual resources can be deployed across multiple dimensions (host→chassis→rack→site→geographic), enabling the system to satisfy both affinity (same dimension) and isolation (different dimensions) requirements

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Reliability

If the infrastructure management layer provides only separate deployment policy, then reliability is improved, but adaptability deteriorates as it cannot satisfy applications requiring both isolation and affinity

Engineering Contradiction:

Improveservice reliabilityVSAvoiddeployment policy flexibility

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic deployment policy that can adapt to different application requirements. The system dynamically selects the appropriate deployment level (host, chassis, rack, site, or geographic) based on the specific isolation and affinity needs of each service application, making the infrastructure management layer both reliable and adaptable

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the deployment parameter from a binary choice (same or different host) to a multi-level hierarchical parameter (host, chassis, rack, site, geographic disaster recovery). This parameter expansion enables the system to provide differentiated deployment policies that satisfy diverse application requirements while maintaining reliability

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10104010B2Method and apparatus for allocating resources

Publication Date: 2018.10.16 HUAWEI TECH CO LTD

AI summary

A method for allocating resources performed by an infrastructure management node is provided, the method comprising obtaining deployment requirements of virtual resources of a service application at an application layer, wherein the deployment requirements include requirements of isolation physical resources and requirements of affinity physical resources, wherein physical resources of levels are ranked by isolation from low to high and wherein the physical resources of the levels are ranked by affinity from low to high, wherein a level of the isolation physical resources is higher than a level of the affinity physical resources in the deployment requirements of the virtual resources, determining the physical resources which satisfy the requirements of isolation physical resources and requirements of affinity physical resources at a physical resources layer, and creating the virtual resources on the determined physical resources and providing the virtual resources to the service application.