VM Cluster Placement Model for Cloud Infrastructure Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In complex cloud environments, automatically determining optimal placement for virtual machine (VM) clusters across heterogeneous hardware infrastructure is challenging due to varying application and customer requirements, as well as infrastructure maintenance needs, while existing methods often rely on static heuristic algorithms that do not account for these factors.

Innovation Solution

A dynamic placement model that selects host nodes based on constraints and optimization criteria, ranking potential combinations to identify the optimal placement for VM clusters, considering factors like resource distribution, NUMA socket affinity, and failure scenarios, to maximize resource utilization and minimize disruption during failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If static heuristic algorithms are used for VM cluster placement, then the placement process is simple, but the placement cannot account for application requirements and administrator goals

Engineering Contradiction:

Improveability to account for application requirements and administrator goalsVSAvoidplacement model complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a dynamic placement model that continuously adapts to changing conditions by incorporating multiple optimization criteria (resource distribution, failure scenarios, application requirements) and re-evaluating placements as infrastructure changes occur. This transforms the static heuristic approach into a dynamic system that responds to real-time conditions.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes multiple parameters simultaneously including resource distribution metrics, failure scenario weights, application requirement priorities, and administrator goal coefficients. By adjusting these parameters, the model can adapt to different situations without requiring complete redesign of the placement algorithm.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If VM clusters are placed to maximize resource utilization, then hardware throughput increases, but disruption during failures increases

Engineering Contradiction:

Improvehardware throughputVSAvoidavailability during failures

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary analysis of failure scenarios during the placement phase, identifying potential single points of failure and pre-positioning VM clusters to avoid them. By anticipating failures before they occur, the system can maintain high resource utilization while preventing catastrophic disruptions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The placement model incorporates cushioning factors that create buffers against failures by distributing VM clusters away from concentrated resource pools and ensuring adequate redundancy. This cushioning effect absorbs the impact of failures while maintaining overall system productivity.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

3Adaptability or versatility

If VMs are distributed across heterogeneous compute nodes, then infrastructure maintenance flexibility improves, but placement optimization difficulty increases

Engineering Contradiction:

Improveinfrastructure maintenance flexibilityVSAvoidplacement optimization complexity

Core Design Contradiction:

Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system segments the heterogeneous infrastructure into manageable categories (compute node types, failure domains, resource pools) and applies specific optimization criteria to each segment. This segmentation reduces the complexity of optimizing across the entire heterogeneous environment while maintaining flexibility for maintenance operations.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20230350733A1Virtual machine cluster placement in a cloud environment

Publication Date: 2023.11.02 ORACLE INT CORP

AI summary

Techniques are described herein for automatically determining optimal placement for VM clusters in multi-device infrastructure. Potential combinations of host nodes for a VM cluster are selected based on applicable constraints on host nodes for the cluster. Further, applicable optimization criteria (OC) for the VM cluster and/or the infrastructure are formally defined and modeled for automatic performance. Application of this placement model to the potential combinations of host nodes results in one or more OC metrics that may be directly compared so that alternate potential host node combinations may be ranked based on the determined OC metrics. The highest-ranked node combination is identified as the optimal VM cluster placement. The placement model can be used to implement initial, incremental, shuffling, or scaling placements of VM clusters. Further, hierarchical decisions may be made based on the determined OC metrics, allowing for application of the placement model to large and complex infrastructures.