Automatic control loop decision variation

Patent metadata

Field	Value
Publication number	US-11586171-B2
Application number	US-201916429736-A
Country	US
Kind code	B2
Filing date	Jun 3, 2019
Priority date	Jun 3, 2019
Publication date	Feb 21, 2023
Grant date	Feb 21, 2023

Abstract

A method includes defining a plurality of variables to modify in a control loop; collecting first data using a first variable of the plurality of variables while executing the control loop, generating a first result based on the collecting first data step, substituting a second variable of the plurality of variables for the first variable, collecting second data using the second variable while executing the control loop, generating a second result based on the collecting second data step, comparing the first result and the second result; and taking an action based on the comparing step.

First claim

The invention claimed is:

1. A method comprising:
defining a plurality of variables to modify in a control loop that is executed in a run-time execution platform in communication with a network;
collecting first performance data of the network by the control loop, the collecting of the first performance data using a first variable of the plurality of variables while executing the control loop, the first performance data characterizing a loading of the network;
generating a first result based on the collecting of the first performance data;
substituting a second variable of the plurality of variables for the first variable;
collecting second performance data of the network by the control loop, the collecting of the second performance data using the second variable while executing the control loop, the second performance data characterizing a latency of the network;
generating a second result based on the collecting of the second performance data;
comparing the first result and the second result; and
taking an action based on the comparing, the action comprising using an optimized variation during operation of the run-time execution platform, the action further comprising facilitating an increase or a decrease in an instantiation of virtual network functions.

2. The method of claim 1 , wherein the taking the action further comprises publishing the first result and the second result.

3. The method of claim 1 , wherein the plurality of variables comprises modifying a processing order of functions executed in the control loop.

4. The method of claim 1 , wherein the plurality of variables comprises two or more threshold values and the taking the action comprises selecting one of the two or more threshold values to use when executing the control loop.

5. The method of claim 1 , wherein the plurality of variables comprises variables that simulate failures and the method further comprises analyzing a reconfigurability of a system based on the failures that are simulated.

6. The method of claim 1 , further comprising repeating the substituting the second variable, the collecting the second performance data, the generating the second result, and the comparing the first result and the second result for additional variables of the plurality of variables.

7. The method of claim 6 , further comprising setting a threshold value and wherein the repeating is halted based on calculations in the control loop exceeding the threshold value.

8. The method of claim 6 , wherein the repeating is performed periodically.

9. The method of claim 6 , wherein the repeating is performed for a set period of time.

10. The method of claim 1 , wherein the increase or the decrease in the instantiation of the virtual network functions scales a cloud-based application.

11. The method of claim 10 , wherein the increase or the decrease in the instantiation of the virtual network functions scales the cloud-based application between a regional level and a national level.

12. A method comprising:
configuring a decision recipe;
executing a control loop that is associated with a run-time execution platform in communication with a network, the executing of the control loop using a first variable from the decision recipe;
collecting first performance data based on the executing of the control loop, the first performance data characterizing a loading of the network;
analyzing the first performance data to form a first result;
substituting an additional variable from the decision recipe for the first variable;
again executing the control loop, the again executing of the control loop using the additional variable from the decision recipe;
collecting second performance data based on the again executing of the control loop, the second performance data characterizing a latency of the network;
analyzing the second performance data to form a second result; and
facilitating a taking of an action based on the first result and the second result, the action comprising an increase or a decrease in an instantiation of virtual network functions.

13. The method of claim 12 , further comprising grading the first result and the second result.

14. The method of claim 13 , wherein the grading is based on performance results.

15. The method of claim 13 , wherein the grading is based on conformance to policies.

16. The method of claim 12 , further comprising comparing the first result and the second result to determine an optimized variation and wherein the action includes using the optimized variation during operation of the run-time execution platform.

17. A device comprising:
a decision recipe design module;
a data collection module configured to collect data from operation of a network;
an input-output interface; and
a processor coupled to the input-output interface, wherein the processor is coupled to a memory, the memory having stored thereon executable instructions that when executed by the processor cause the processor to effectuate operations comprising:
first executing of a control loop that is associated with a run-time execution platform, the first executing of the control loop using a first variable that is retrieved from the decision recipe design module, the first executing of the control loop collecting first performance data characterizing a loading of the network;
analyzing the first performance data to form a first result;
substituting a second variable from the decision recipe design module for the first variable;
second executing of the control loop, the second executing of the control loop using the second variable that is retrieved from the decision recipe design module;
collecting second performance data based on the second executing of the control loop, the second performance data characterizing a latency of the network;
analyzing the second performance data to form a second result; and
facilitating a taking of an action based on the second result, the action comprising an increase or a decrease in an instantiation of virtual network functions.

18. The device of claim 17 , wherein the operations further comprise comparing the first result and the second result to determine an optimized variation and wherein the action that is taken is based on the first result and the second result and includes using the optimized variation during operation of the run-time execution platform.

19. The device of claim 17 , wherein a decision recipe from the decision recipe design module comprises a set of variables based on policies.

20. The device of claim 17 , wherein a decision recipe from the decision recipe design module comprises a set of thresholds and the operations further comprise determining a threshold from the set of thresholds for operation of the run-time execution platform.

Assignees

• At & T Ip I Lp

Inventors

• Aaron Jeffrey
• Fan James

Classifications

• G05B19/4083 Primary

  Adapting program, configuration · CPC title

• G06F2201/81

  Threshold · CPC title

• G06F11/302 Primary

  where the computing system component is a software system · CPC title

• G06F11/3457

  Performance evaluation by simulation · CPC title

• G05B2219/33125

  System configuration, reconfiguration, customization, automatic · CPC title