Unified Compute Storage Provisioning in Cloud Networks

Solution Overview

Problem

Conventional virtualization schemes fail to optimally provision virtual machines and virtual storage in cloud environments as they only consider compute resources for virtual machine provisioning and storage resources for virtual storage provisioning, neglecting the importance of both compute and storage characteristics, which results in suboptimal system performance.

Innovation Solution

A method and apparatus that analyze attributes of both compute and storage resources to select the best-fit resources for provisioning virtual machines and virtual storage, using network devices to identify resource information, apply best-fit rules, and ensure optimal performance and redundancy by considering relative and absolute attributes such as locality, performance, and redundancy factors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional virtualization schemes provision virtual machines considering only compute resources, then the provisioning process is simple, but system performance is suboptimal

Engineering Contradiction:

Improvesystem performanceVSAvoidprovisioning process complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent merges compute resource provisioning and storage resource provisioning into a unified process. The system simultaneously evaluates both compute and storage resources when provisioning virtual machines, rather than handling them separately. This integration ensures that both resource types are optimized together, improving overall system performance while managing complexity through a coordinated approach.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The provisioning system is designed to handle multiple resource types (compute and storage) through a single universal mechanism. The same provisioning logic and resource selection algorithms are applied to both compute and storage resources, allowing the system to optimize allocations across different resource dimensions without requiring separate specialized processes for each resource type.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Productivity

If conventional virtualization schemes provision virtual storage considering only storage resources, then the provisioning process is simple, but system performance is suboptimal

Engineering Contradiction:

Improvesystem performanceVSAvoidprovisioning process complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent merges compute resource provisioning and storage resource provisioning into a unified process. The system simultaneously evaluates both compute and storage resources when provisioning virtual machines, rather than handling them separately. This integration ensures that both resource types are optimized together, improving overall system performance while managing complexity through a coordinated approach.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The provisioning system is designed to handle multiple resource types (compute and storage) through a single universal mechanism. The same provisioning logic and resource selection algorithms are applied to both compute and storage resources, allowing the system to optimize allocations across different resource dimensions without requiring separate specialized processes for each resource type.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Manufacturing precision

If the system analyzes both compute and storage resource attributes for provisioning, then resource allocation optimality is improved, but processing complexity increases

Engineering Contradiction:

Improveresource allocation precisionVSAvoidanalysis process complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the resource analysis process into distinct phases: first analyzing compute resource attributes, then analyzing storage resource attributes, and finally integrating the results for provisioning decisions. This segmentation allows the complex multi-attribute analysis to be broken down into manageable steps, improving allocation precision while controlling processing complexity through structured analysis.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary analysis of resource attributes by pre-evaluating and storing compute and storage resource characteristics before provisioning is needed. This preliminary action includes assessing resource capacities, performance metrics, and compatibility attributes in advance, so that when provisioning decisions are required, the system can quickly reference pre-analyzed data rather than performing complex analysis in real-time.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9229749B2Compute and storage provisioning in a cloud environment

Publication Date: 2016.01.05 CISCO TECHNOLOGY INC

AI summary

In one embodiment, a method includes receiving at a network device, resource information comprising attributes for compute and storage resources in a network, identifying a need for provisioning a virtual element, and selecting one of the compute and storage resources for use in provisioning the virtual element. Selection of the compute or storage resource includes analyzing the resource information for the compute resources and the storage resources in the network. An apparatus is also disclosed.