Motif-Based Subgraph Matching for Workload Placement

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional subgraph matching techniques face challenges due to their high complexity and resource requirements, especially when dealing with large graphs and partially observed networks, which limits their efficiency in applications like workload placement and image processing.

Innovation Solution

The approach involves creating motifs from subgraphs to reduce the complexity of subgraph matching. This is achieved by decomposing workload graphs into smaller subgraphs (graphlets) and then constructing motifs from these graphlets, which are semantically richer and easier to match against target graphs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional subgraph matching techniques are used to find all nodes in a graph, then complete matching accuracy is achieved, but computational complexity and resource requirements increase significantly

Engineering Contradiction:

Improvematching accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the target graph into multiple partitions or blocks, and divides the query subgraph matching task into smaller sub-tasks that can be executed independently on each partition. This segmentation reduces the computational complexity by avoiding exhaustive search over the entire graph while maintaining matching accuracy through coordinated processing of partitions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary indexing and preprocessing of the target graph structure before executing subgraph matching queries. By pre-computing and storing graph structural information in an optimized format, the system reduces the complexity of the actual matching process while ensuring complete and accurate results.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If exhaustive search is performed for all nodes in the graph, then all matching subgraphs are identified, but execution time increases significantly

Engineering Contradiction:

Improvematching completenessVSAvoidexecution time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The graph is divided into partitions that can be processed in parallel, significantly reducing execution time while maintaining matching completeness. Each partition is searched independently for matching subgraphs, and results are aggregated to ensure all matches are found.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent employs iterative refinement processes where initial matching results are periodically updated and refined. This allows the system to quickly identify potential matches and then systematically verify them, reducing overall execution time while ensuring completeness.

Inventive Principle:

Principle #19Periodic action

3Quantity of substance

If the size of the target graph increases, then more nodes are available for matching, but the resources required to perform subgraph matching increase significantly

Engineering Contradiction:

Improvenumber of available nodesVSAvoidcomputational resources

Core Design Contradiction:

Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

Large graphs are partitioned into smaller manageable blocks that can be processed with limited computational resources. This allows the system to handle graphs of any size by distributing the processing load across multiple partitions without requiring proportional increases in total computational resources.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies filtering techniques that perform partial matching operations first to eliminate non-matching candidates before conducting full verification. This reduces the effective search space and computational resources required while still examining all necessary nodes for potential matches.

Inventive Principle:

Principle #16Partial or excessive action

4Productivity

If conventional subgraph matching is applied to workload placement, then optimal task allocation is achieved, but network strain and processing overhead increase

Engineering Contradiction:

Improveworkload placement efficiencyVSAvoidnetwork strain

Core Design Contradiction:

ProductivityVSObject-generated harmful factors

Solution Approach 1:

The workload placement process is divided into independent phases: query processing, matching execution, and result aggregation. This segmentation allows each phase to be optimized independently and reduces network strain by minimizing communication overhead between processing stages.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces intermediate data structures and caching mechanisms that act as mediators between the query processor and the target graph. These intermediaries reduce network strain by localizing data access and minimizing repeated queries to the underlying graph storage.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250131038A1Motif-based subgraph matching in partially observed graphs

Publication Date: 2025.04.24 DELL PROD LP

AI summary

A motif based approach for subgraph matching in partially observed graphs is disclosed. Graphlets are extracted from a query graph, such as a graph of a workload. Motifs are built from the graphlets and the motifs are matched to a target graph, such as an infrastructure graph. Once the motifs are matched to nodes in the target graph, tasks of the workload, which correspond to nodes in the query graph, are placed in the infrastructure for execution.