Compute Resource Grouping for Cloud Migration Readiness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of enterprise networks makes it difficult for IT administrators to assess and manage compute resources effectively, especially during cloud migration, as they lack a clear picture of network utilization and resource allocation between on-premise and cloud-based infrastructure.
Innovation Solution
A system that processes a list of compute resource names to assess and remove irrelevant data, generate feature values, determine distances using metrics, group similar resources, and provide recommendations for resource grouping and cloud migration readiness, including workload discovery and cost analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If compute resources are managed individually without grouping, then resource tracking is simple, but resource management efficiency and cloud migration readiness assessment deteriorate
Solution Approach 1:
The patent merges multiple compute resources into application groups based on hostname similarity analysis. By automatically grouping resources that share common applications or functions, the system improves management efficiency while reducing the perceived complexity through organized categorization rather than individual resource tracking.
Solution Approach 2:
The patent introduces an intermediary analysis layer that processes hostnames and extracts meaningful features to determine resource groupings. This intermediary processing layer enables intelligent resource organization without requiring direct complex interactions between all resources, thereby improving management efficiency while maintaining system tractability.
2Measurement precision
If all compute resource names are retained with full detail, then resource identification precision is high, but data processing complexity and time increase
Solution Approach 1:
The patent extracts meaningful features from compute resource hostnames by removing irrelevant components such as sequential numbers, deployment identifiers, and other non-essential suffixes. This extraction process retains the critical identifying information needed for precise resource identification while eliminating data redundancy that would increase processing time.
Solution Approach 2:
The patent segments hostname data into meaningful components (prefix, application identifier, suffix) and processes each segment differently. By segmenting the data, the system preserves essential identification information while removing redundant elements, thereby maintaining identification precision while reducing overall data processing requirements.
3Measurement precision
If compute resources are grouped by multiple methodologies, then grouping accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies multiple grouping methodologies (hierarchical clustering and auto-optimization) to compute resources, using more analysis than a single method would provide. This partial application of multiple methods improves grouping accuracy by cross-validating results, while the system manages computational complexity by implementing the methods sequentially and using early results to inform subsequent analysis.
Data Source
AI summary
A method comprises receiving a list of names of compute resources of an enterprise network, the names of compute resources including at least one name of a virtual machine and at least one name of a host, for each of the list of names: assessing a particular name of the list of names for removable data and removing the removable data to generate a feature for that particular name, for each feature, determining distances based on at least one metric for every other feature, grouping features based on distances, each feature being in only one group, recommending groups of compute resources based on groups of features, each feature of a group of features being associated with a different compute resource, the group of compute resources corresponding to a particular group of features, and providing a report of recommended groups.