Network deployment recommendation using machine learning

Patent metadata

Field	Value
Publication number	US-2025385835-A1
Application number	US-202418743286-A
Country	US
Kind code	A1
Filing date	Jun 14, 2024
Priority date	Jun 14, 2024
Publication date	Dec 18, 2025
Grant date	—

Abstract

A method comprises receiving a request to predict a deployment configuration for at least one application, analyzing code of the at least one application to identify one or more additional applications on which the at least one application will depend, identifying a plurality of network paths between the at least one application and the one or more additional applications, and using one or more machine learning algorithms to predict execution times for the at least one application over the plurality of network paths. The predicted execution times for the at least one application over the plurality of network paths are inputted to a network graph model. The network graph model predicts the deployment configuration for the at least one application based at least in part on the predicted execution times, wherein the deployment configuration comprises a subset of the plurality of network paths.

First claim

What is claimed is:

1 . A method comprising:
receiving a request to predict a deployment configuration for at least one application; analyzing code of the at least one application to identify one or more additional applications on which the at least one application will depend; identifying a plurality of network paths between the at least one application and the one or more additional applications; using one or more machine learning algorithms to predict execution times for the at least one application over the plurality of network paths; and inputting the predicted execution times for the at least one application over the plurality of network paths to a network graph model, wherein the network graph model predicts the deployment configuration for the at least one application based at least in part on the predicted execution times for the at least one application over the plurality of network paths, and wherein the deployment configuration comprises a subset of the plurality of network paths; and wherein the steps of the method are executed by a processing device operatively coupled to a memory.

2 . The method of claim 1 wherein the at least one application comprises at least one of a micro-frontend application and a microservice application.

3 . The method of claim 1 wherein the one or more additional applications comprise at least one of one or more micro-frontend applications and one or more microservice applications, and wherein the one or more additional applications are deployed on one or more cloud platforms of a plurality of cloud platforms.

4 . The method of claim 1 wherein analyzing the code of the at least one application comprises identifying one or more protocol patterns in the code corresponding to at least one service call.

5 . The method of claim 1 further comprising collecting execution times for a plurality of applications.

6 . The method of claim 5 wherein the collecting comprises tracing respective network paths of the plurality of applications.

7 . The method of claim 5 further comprising training the one or more machine learning algorithms with the collected execution times for the plurality of applications.

8 . The method of claim 5 wherein:
the one or more machine learning algorithms comprise a regression algorithm; and
the method further comprises using the regression algorithm to predict respective execution times between respective pairs of the plurality of applications.

9 . The method of claim 8 wherein the predicted execution times for the at least one application over the plurality of network paths are based at least in part on one or more of the respective execution times between the respective pairs of the plurality of applications.

10 . The method of claim 8 wherein:
the network graph model graphs one or more of the respective pairs of the plurality of applications as respective node pairs;
the network graph model graphs the one or more of the respective execution times between the respective pairs of the plurality of applications as one or more respective edges between the respective node pairs; and
the one or more respective edges correspond to respective weights representing the one or more of the respective execution times.

11 . The method of claim 10 wherein the network graph model uses a shortest path algorithm to predict the deployment configuration based at least in part on the respective weights.

12 . The method of claim 8 wherein the regression algorithm comprises a random forest algorithm.

13 . The method of claim 1 wherein the network graph model uses a shortest path algorithm to predict the subset of the plurality of network paths.

14 . An apparatus comprising:
a processing device operatively coupled to a memory and configured: to receive a request to predict a deployment configuration for at least one application; to analyze code of the at least one application to identify one or more additional applications on which the at least one application will depend; to identify a plurality of network paths between the at least one application and the one or more additional applications; to use one or more machine learning algorithms to predict execution times for the at least one application over the plurality of network paths; and to input the predicted execution times for the at least one application over the plurality of network paths to a network graph model, wherein the network graph model predicts the deployment configuration for the at least one application based at least in part on the predicted execution times for the at least one application over the plurality of network paths, and wherein the deployment configuration comprises a subset of the plurality of network paths.

15 . The apparatus of claim 14 wherein the processing device is further configured to collect execution times for a plurality of applications.

16 . The apparatus of claim 15 wherein:
the one or more machine learning algorithms comprise a regression algorithm; and
the processing device is further configured to use the regression algorithm to predict respective execution times between respective pairs of the plurality of applications.

17 . The apparatus of claim 16 wherein:
the network graph model graphs one or more of the respective pairs of the plurality of applications as respective node pairs;
the network graph model graphs the one or more of the respective execution times between the respective pairs of the plurality of applications as one or more respective edges between the respective node pairs; and
the one or more respective edges correspond to respective weights representing the one or more of the respective execution times.

18 . An article of manufacture comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes said at least one processing device to perform the steps of:
receiving a request to predict a deployment configuration for at least one application; analyzing code of the at least one application to identify one or more additional applications on which the at least one application will depend; identifying a plurality of network paths between the at least one application and the one or more additional applications; using one or more machine learning algorithms to predict execution times for the at least one application over the plurality of network paths; and inputting the predicted execution times for the at least one application over the plurality of network paths to a network graph model, wherein the network graph model predicts the deployment configuration for the at least one application based at least in part on the predicted execution times for the at least one application over the plurality of network paths, and wherein the deployment configuration comprises a subset of the plurality of network paths.

19 . The article of manufacture of claim 18 wherein the program code further causes said at least one processing device to perform the step of collecting execution times for a plurality of applications.

20 . The article of manufacture of claim 19 wherein:
the one or more machine learning algorithms comprise a regression algorithm; and
the program code further causes said at least one processing device to perform the step of using the regression algorithm to predict respective execution times between respective pairs of the plurality of applic

Assignees

• Dell Products Lp

Inventors

• Mohanty Bijan Kumar
• Linsey David J
• Dinh Hung

Classifications

• H04L41/0806 Primary

  for initial configuration or provisioning, e.g. plug-and-play · CPC title

• G06F8/60

  Software deployment · CPC title

• H04L41/16 Primary

  using machine learning or artificial intelligence · CPC title