Granted Patent B2

US 12,602,362 · App. 18/084,649 · Granted Apr 14, 2026

Inventors: Jayachandu Bandlamudi (Guntur, IN); Sreenivasa Rao Pamidala (McLean, VA); Subil Mathew Abraham (Lewisville, TX)

Assignee: International Business Machines Corporation

G06F16/2237G06F16/24553G06N20/00

Abstract

Claims (64)

1 . A computer-implemented method for indexing microservices for optimized querying based on microservice attributes, the method comprising:

generating a plurality of application graph data structures, wherein each node in an application graph data structure specifies a corresponding microservice and has corresponding microservice attribute information, and wherein each edge between nodes in the application graph data structure specifies a microservice functionality between microservices of nodes connected by the edge;

executing a data transformation operation on the application graph data structures to generate, for each node in each application graph data structure, a corresponding microservice document specifying microservice attributes of the corresponding microservice;

executing a machine learning training operation on an embedding computer model based on a plurality of the microservice documents, wherein the machine learning training operation trains the embedding computer model to learn a representation vector space for representing microservices as vector representations;

executing the trained embedding computer model on the plurality of microservice documents to generate corresponding vector representations;

compiling the vector representations into entries of a microservice index data structure; and

processing queries for microservices based on the vector representations in the entries of the microservice index data structure.

2 . The computer-implemented method of claim 1 , wherein processing queries for microservices based on the vector representations in the entries of the microservice index data structure comprises:

receiving a microservice query comprising a microservice specification;

executing the embedding computer model on the microservice specification to generate a microservice specification vector representation;

executing a similarity based search of the microservice index data structure based on a similarity analysis between the microservice specification vector representation and vector representations in entries of the microservice index data structure; and

returning results of the similarity based search of the microservice index data structure to a source of the microservice query.

3 . The computer-implemented method of claim 2 , wherein the similarity analysis comprises, for each pairing of the microservice specification vector representation and a vector representation of an entry in the microservice index data structure, determining a similarity metric based on a similarity metric calculation function.

4 . The computer-implemented method of claim 2 , wherein returning results of the similarity based search of the microservice index data structure comprises:

ranking microservices corresponding to the entries in the microservice index data structure relative to one another based on their corresponding similarity metrics to generate a ranked listing;

selecting a subset of microservices from the ranked listing;

executing an analysis of microservice metadata information stored in entries of the microservice index data structure corresponding to the subset of microservices to determine, for each microservice in the subset of microservices, whether the microservice implements specified best practices and avoids specified inefficiencies;

re-ranking the microservices in the subset of microservices based on results of the analysis; and

generating a recommendation output that recommends one or more of the microservices in the subset of microservices as a result of the received microservice query.

5 . The computer-implemented method of claim 4 , wherein the recommendation output is a user interface having user selectable user interface elements that, when selected by a user, retrieve a corresponding microservice container from a microservice repository into a development computer environment for developing a composite application.

6 . The computer-implemented method of claim 1 , wherein the embedding computer model is a neural network.

7 . The computer-implemented method of claim 1 , wherein the embedding computer model is an autoencoder.

8 . The computer-implemented method of claim 1 , wherein generating the graph data structures comprises, for each composite application in a plurality of composite applications:

receiving input data comprising code repository information, deployment environment information, and application information;

extracting microservice metadata information for the microservices implemented in the composite application from the input data; and

generating a graph of the composite application based on the extracted microservice metadata information.

9 . The computer-implemented method of claim 8 , wherein the microservice metadata information comprises at least one of an owner of the microservice, application information specifying an application type with which the microservice operates, compatibility information, container runtime information, connectivity information to other microservices, deployment environment information, or integration pattern information.

10 . The computer-implemented method of claim 8 , wherein each entry in the entries in the microservice index include, for a corresponding microservice, a microservice identifier, microservice metadata information, and a vector representation for the corresponding microservice.

11 . A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed in a data processing system, causes the data processing system to:

generate a plurality of application graph data structures, wherein each node in an application graph data structure specifies a corresponding microservice and has corresponding microservice attribute information, and wherein each edge between nodes in the application graph data structure specifies a microservice functionality between microservices of nodes connected by the edge;

execute a data transformation operation on the application graph data structures to generate, for each node in each application graph data structure, a corresponding microservice document specifying microservice attributes of the corresponding microservice;

execute a machine learning training operation on an embedding computer model based on a plurality of the microservice documents, wherein the machine learning training operation trains the embedding computer model to learn a representation vector space for representing microservices as vector representations;

execute the trained embedding computer model on the plurality of microservice documents to generate corresponding vector representations;

compile the vector representations into entries of a microservice index data structure; and

process queries for microservices based on the vector representations in the entries of the microservice index data structure.

12 . The computer program product of claim 11 , wherein processing queries for microservices based on the vector representations in the entries of the microservice index data structure comprises:

receiving a microservice query comprising a microservice specification;

executing the embedding computer model on the microservice specification to generate a microservice specification vector representation;

executing a similarity based search of the microservice index data structure based on a similarity analysis between the microservice specification vector representation and vector representations in entries of the microservice index data structure; and

returning results of the similarity based search of the microservice index data structure to a source of the microservice query.

13 . The computer program product of claim 12 , wherein the similarity analysis comprises, for each pairing of the microservice specification vector representation and a vector representation of an entry in the microservice index data structure, determining a similarity metric based on a similarity metric calculation function.

14 . The computer program product of claim 12 , wherein returning results of the similarity based search of the microservice index data structure comprises:

ranking microservices corresponding to the entries in the microservice index data structure relative to one another based on their corresponding similarity metrics to generate a ranked listing;

selecting a subset of microservices from the ranked listing;

executing an analysis of microservice metadata information stored in entries of the microservice index data structure corresponding to the subset of microservices to determine, for each microservice in the subset of microservices, whether the microservice implements specified best practices and avoids specified inefficiencies;

re-ranking the microservices in the subset of microservices based on results of the analysis; and

generating a recommendation output that recommends one or more of the microservices in the subset of microservices as a result of the received microservice query.

15 . The computer program product of claim 14 , wherein the recommendation output is a user interface having user selectable user interface elements that, when selected by a user, retrieve a corresponding microservice container from a microservice repository into a development computer environment for developing a composite application.

16 . The computer program product of claim 11 , wherein the embedding computer model is a neural network.

17 . The computer program product of claim 11 , wherein the embedding computer model is an autoencoder.

18 . The computer program product of claim 11 , wherein generating the graph data structures comprises, for each composite application in a plurality of composite applications:

receiving input data comprising code repository information, deployment environment information, and application information;

extracting microservice metadata information for the microservices implemented in the composite application from the input data; and

generating a graph of the composite application based on the extracted microservice metadata information.

19 . The computer program product of claim 18 , wherein the microservice metadata information comprises at least one of an owner of the microservice, application information specifying an application type with which the microservice operates, compatibility information, container runtime information, connectivity information to other microservices, deployment environment information, or integration pattern information.

20 . An apparatus comprising:

at least one processor; and

at least one memory coupled to the at least one processor, wherein the at least one memory comprises instructions which, when executed by the at least one processor, cause the at least one processor to:

generate a plurality of application graph data structures, wherein each node in an application graph data structure specifies a corresponding microservice and has corresponding microservice attribute information, and wherein each edge between nodes in the application graph data structure specifies a microservice functionality between microservices of nodes connected by the edge;

execute a data transformation operation on the application graph data structures to generate, for each node in each application graph data structure, a corresponding microservice document specifying microservice attributes of the corresponding microservice;

execute a machine learning training operation on an embedding computer model based on a plurality of the microservice documents, wherein the machine learning training operation trains the embedding computer model to learn a representation vector space for representing microservices as vector representations;

execute the trained embedding computer model on the plurality of microservice documents to generate corresponding vector representations;

compile the vector representations into entries of a microservice index data structure; and

process queries for microservices based on the vector representations in the entries of the microservice index data structure.

Assignments (1)

Continuity (1)

References Cited (28)