VM Deployment in Hyperconverged Infrastructure via Workload Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information handling systems with hyperconverged infrastructure (HCI) lack coordination between virtualized workloads and hardware platforms, leading to inefficient resource utilization, as they do not consider the hardware platform when balancing virtual machine (VM) workloads.
Innovation Solution
The system employs machine learning to classify VM workloads and optimize their deployment by matching them with like-provisioned nodes, using a predictive model to determine workload types based on compute, storage, and network metrics, ensuring that VMs execute on nodes with similar resource configurations, thereby aligning VMs with the most suitable hardware resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If VMs are balanced across hosts without considering hardware platform, then resource contention is reduced, but hardware utilization efficiency deteriorates
Solution Approach 1:
The patent applies local quality by characterizing both VMs and hosts with specific workload type labels (e.g., compute-intensive, storage-intensive, network-intensive, performance-intensive) and matching VMs to hosts with compatible characteristics. This ensures that each host receives workloads appropriate to its hardware configuration, optimizing local resource utilization while maintaining overall system balance
Solution Approach 2:
The system implements feedback by continuously monitoring host resource usage and VM workload characteristics, then using this information to make intelligent placement decisions. The workload type characterization and matching process creates a closed-loop system that adapts to changing conditions and optimizes hardware utilization based on actual system state
2Device complexity
If VM deployment is managed without workload-hardware coordination, then system simplicity is maintained, but resource utilization efficiency deteriorates
Solution Approach 1:
The system applies self-service by enabling VMs to effectively 'self-identify' their workload type through characterization, and by allowing hosts to 'self-describe' their capabilities through hardware profiling. The automated matching process eliminates the need for manual intervention while achieving optimized resource allocation, maintaining operational simplicity despite the sophisticated matching logic
3Loss of energy
If hardware-specific VM placement is implemented, then hardware utilization is maximized, but system complexity increases
Solution Approach 1:
The patent applies parameter changes by transforming the complex problem of hardware-VM matching into a simpler parameter-based classification system. By characterizing workloads and hosts using discrete workload type parameters (compute-intensive, storage-intensive, etc.), the system reduces the complexity of deployment management while achieving optimized hardware utilization through parameter-based matching
Data Source
AI summary
Disclosed systems and methods align virtual machines (VMs) with hyperconverged infrastructure (HCl)-based hardware based on a determination of the VM's characteristic type. Each cluster node may be provisioned with a specific combination of hardware. In some embodiments, disclosed methods include determining, for each of the plurality of nodes, a target workload type based at least in part on the combination of information handling resources provisioned on each node. Disclosed methods may manage deployment of the VMs among the cluster nodes based on one or more factors including a workload compatibility factor determined in accordance with a workload type of each VM and a target workload type of each node, where the target workload type may reflect a hardware characteristic of the node.