Dynamic Cost Calculation for SRIOV Virtual Functions

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

There is a need for an effective mechanism to implement dynamic cost calculation for Single Root Input/Output Virtualization (SRIOV) virtual functions in cloud environments, allowing for scalable resource allocation and pricing based on usage statistics.

Innovation Solution

A method and system that periodically query SRIOV adapters for activity statistics, build usage heuristics, and calculate dynamic costs, enabling users to scale VF resources up or down based on usage, with both hypervisor and PF device drivers adjusting resources accordingly, and providing different service levels by dynamically allocating and pricing VF resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If static cost pricing is used for SRIOV VFs, then pricing simplicity is maintained, but resource allocation efficiency deteriorates

Engineering Contradiction:

Improvepricing simplicityVSAvoidresource allocation efficiency

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The patent implements dynamic cost calculation for SRIOV VFs by continuously monitoring resource usage statistics (bandwidth, I/O operations, memory usage) and adjusting pricing in real-time. The management function queries usage data periodically and recalculates costs based on actual consumption patterns, transforming the static pricing model into a dynamic one that adapts to changing resource demands.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system establishes a feedback loop where usage statistics are collected from SRIOV adapters, processed by the management function, and used to recalculate dynamic costs. This feedback mechanism ensures that pricing continuously reflects actual resource consumption, enabling efficient resource allocation while maintaining operational simplicity through automated calculations.

Inventive Principle:

Principle #23Feedback

2Ease of operation

If fixed VF resources are allocated, then configuration simplicity is maintained, but resource utilization efficiency deteriorates

Engineering Contradiction:

Improveconfiguration simplicityVSAvoidresource utilization efficiency

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent enables dynamic scaling of VF resources by allowing users to scale VFs up or down based on actual usage patterns. The system monitors resource consumption and permits flexible adjustment of VF capacities without requiring complex reconfiguration procedures, thus maintaining operational simplicity while improving resource utilization efficiency.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system dynamically changes VF resource parameters (bandwidth allocation, I/O capacity, memory resources) based on monitored usage statistics. This allows flexible adjustment of resource allocation to match actual demand patterns while maintaining simple user-facing operations through automated parameter optimization.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If detailed usage monitoring is implemented, then cost calculation accuracy is improved, but system complexity deteriorates

Engineering Contradiction:

Improvecost calculation accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent leverages the existing SRIOV adapter's built-in usage statistics collection capability, which already provides detailed monitoring data for performance management. The management function reuses this existing data infrastructure for cost calculation purposes, avoiding the need for separate monitoring systems and thus maintaining cost calculation accuracy without significantly increasing system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Adaptability or versatility

If dynamic cost calculation is implemented, then resource allocation flexibility is improved, but computational overhead deteriorates

Engineering Contradiction:

Improveresource allocation flexibilityVSAvoidcomputational overhead

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The management function implements periodic querying of usage statistics at predetermined time intervals rather than continuous monitoring. This periodic approach enables dynamic cost calculation and flexible resource allocation while reducing computational overhead by processing updates only when new interval data becomes available, rather than maintaining constant calculation operations.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS9996382B2Implementing dynamic cost calculation for SRIOV virtual function (VF) in cloud environments

Publication Date: 2018.06.12 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A method, system and computer program product are provided for implementing dynamic cost calculation for a Single Root Input/Output Virtualization (SRIOV) virtual function (VF) in cloud environments. A management function periodically queries the SRIOV adapter for activity statistics for each assigned virtual function. The management function builds a usage heuristic based on the resource usage statistics. The management function calculates dynamic cost for the SRIOV VF based on the resource usage statistics. Calculated dynamic costs for the SRIOV VF are provided to a virtual function user and users are enabled to scale their VF resources. The VF resources are selectively scaled-up and scaled-down responsive to user input based upon VF resource usage.