US 2022/0019495 A1 · MACHINE LEARNING-BASED TECHNIQUES FOR PROVIDING FOCUS TO PROBLEMATIC COMPUTE RESOURCES REPRESENTED VIA A DEPENDENCY GRAPH

Abstract

Methods, systems, apparatuses, and computer-readable storage mediums are described for machine learning-based techniques for reducing the visual complexity of a dependency graph that is representative of an application or service. For example, the dependency graph is generated that comprises a plurality of nodes and edges. Each node represents a compute resource (e.g., a microservice) of the application or service. Each edge represents a dependency between nodes coupled thereto. A machine learning-based classification model analyzes each of the nodes to determine a likelihood that each of the nodes is a problematic compute resource. For instance, the classification model may output a score indicative of the likelihood that a particular compute resource is problematic. The nodes and/or edges having a score that exceed a predetermined threshold are provided focus via the dependency graph.

Claims20

Dependent claims

A method for identifying a problematic compute resource via a dependency graph that comprises a plurality of nodes representing a plurality of compute resources, the method comprising:

for each compute resource represented by a node of the dependency graph:
providing features associated with the compute resource to a machine learning model configured to output a score for the compute resource, the score indicating a likelihood that the compute resource is problematic;
and
for each node associated with a score that exceeds a predetermined threshold:
providing focus to at least one of the node or an edge coupling the at least one node to another node of the dependency graph that represents another compute resource.

2 more elements

A system for identifying a problematic compute resource via a dependency graph that comprises a plurality of nodes representing a plurality of compute resources, comprising:

at least one processor circuit;
and
at least one memory that stores program code configured to be executed by the at least one processor circuit, the program code comprising:
a graph analyzer configured to, for each compute resource represented by a node of the dependency graph, provide features associated with the compute resource to a machine learning model configured to output a score for the compute resource, the score indicating a likelihood that the compute resource is problematic;
and
a dependency graph generator configured to, for each node associated with a score that exceeds a predetermined threshold, provide focus to at least one of the node or an edge coupling the at least one node to another node of the dependency graph that represents another compute resource.

3 more elements

A computer-readable storage medium having program instructions recorded thereon that, when executed by at least one processor of a computing device perform a method for identifying a problematic compute resource via a dependency graph that comprises a plurality of nodes representing a plurality of compute resources, the method comprising:

for each compute resource represented by a node of the dependency graph:
providing features associated with the compute resource to a machine learning model configured to output a score for the compute resource, the score indicating a likelihood that the compute resource is problematic;
and
for each node associated with a score that exceeds a predetermined threshold:
providing focus to at least one of the node or an edge coupling the at least one node to another node of the dependency graph that represents another compute resource.

2 more elements

Dependencies are inferred from the claim text.

Machine learning-based techniques for providing focus to problematic compute resources represented via a dependency graph

Life of this patent

Machine learning-based techniques for providing focus to problematic compute resources represented via a dependency graph

Life of this patent

Abstract

Claims20