Techniques and system for optimization driven by dynamic resilience

Patent metadata

Field	Value
Publication number	US-10936458-B2
Application number	US-201916297098-A
Country	US
Kind code	B2
Filing date	Mar 8, 2019
Priority date	Nov 27, 2018
Publication date	Mar 2, 2021
Grant date	Mar 2, 2021

Abstract

Disclosed are hardware and techniques for testing computer processes in a network system by simulating computer process faults and identifying risk associated with correcting the simulated fault and identifying computer processes that may depend on the corrected computer process. The interdependent computer processes in a network may be determined by evaluating a risk matrix having a risk score and non-functional requirement score. An analysis of the risk score and non-functional requirement score accounts for interdependencies between computer processes and identified corrective actions that may be used to determine an optimal network environment. The optimal network environment may be updated dynamically based on changing computer process interdependencies and the determined risk and robustness scores.

First claim

What is claimed is:

1. A method, comprising:
monitoring, by a monitoring component, a simulation instance of a plurality of computer-implemented processes operating in a network environment, wherein the simulation instance of the plurality of computer implemented processes includes a test computer-implemented process;
generating a pre-breakage snapshot of the plurality of computer-implemented processes, wherein the pre-breakage snapshot includes:
a robustness score indicating a process health of each computer-implemented process of the plurality of computer-implemented processes, and
a risk score indicating a threshold between automated correction and manual correction of a degrading system;

generating a simulated process volatility in the test computer-implemented process of a plurality of computer-implemented processes;
generating, by the monitoring component in response to the simulated process volatility, a simulation result snapshot of process health of each of the plurality of computer-implemented processes, wherein the simulation result snapshot includes an updated robustness score and an updated risk score for each computer-implemented process including the test computer-implemented process;
accessing a library of runbooks, wherein:
each runbook in the library of runbooks addresses a respective computer-implemented process volatility of the plurality of computer-implemented processes operating in the network environment, and
each respective runbook includes a plurality of response strategies, wherein each final response strategy of the plurality of response strategies is implementable to cure specific process volatilities of the respective computer-implemented process addressed by the respective runbook;

identifying, based on the simulated process volatility, a specific runbook in the library of runbooks that addresses process volatilities of the test computer-implemented process;
selecting a final response to cure the simulated process volatility of the test computer-implemented process to be implemented in the simulation instance of the network environment;
generating, by the monitoring component in response to a simulated implementation of the final response strategy, a cure result snapshot of process health of each of the plurality of computer-implemented processes;
evaluating the pre-breakage snapshot, the simulation result snapshot, and the cure result snapshot with reference to one another; and
based on results of the evaluation, identifying a network environment architecture as an optimal network architecture that cures the simulated process volatility.

2. The method of claim 1 , wherein:
the optimal network architecture has a below-threshold risk score for each of the plurality of computer-implemented processes and an above-threshold robustness score for each of the plurality of computer-implemented processes.

3. The method of claim 1 , wherein generating the pre-breakage snapshot comprises:
identifying in a list of break event flags for each computer-implemented process of the plurality of computer-implemented processes respective break event symptoms for each of the break event flags in the list of break event flags,
generating, by a rules engine, a robustness score for each respective computer-implemented process, wherein the robustness score for each respective computer-implemented process is based on at least one of an identified break event symptom, a computing environment indicator, a code environment indicator, the respective break event symptom corresponding to the respective computer-implemented process, and the response strategy corresponding to the respective break event symptom of the respective computer-implemented process;
generating, by the rules engine, a risk score for each respective computer-implemented process based on the identified break event symptom, and the response strategy corresponding to the respective break event symptom of the respective computer-implemented process; and
storing the generated robustness and risk scores of each computer-implemented process with a timestamp of when the pre-breakage snapshot was taken in a data structure.

4. The method of claim 1 , wherein generating the simulation result snapshot, comprises:
in response to application of the selected final response to the simulation instance, generating, based on inputs received from the monitoring component, a list of break event flags for each computer-implemented process of the plurality of computer-implemented processes;
identifying respective break event symptoms for the break event flags in the list of break event flags,
determining, for each of the identified respective break event symptoms, a corresponding computing environment indicator, a corresponding code environment indicator, and a corresponding final response strategy;
storing each respective break event symptom of the identified respective break event symptoms, the determined corresponding computing environment indicator, the determined corresponding code environment indicator and determined corresponding final response strategy into a data structure;
generating, by a rules engine, a simulation robustness score for each respective computer-implemented process of the plurality of computer-implemented processes based on at least one break event symptom of the identified respective break event symptom, the determined corresponding computing environment indicator, the determined corresponding code environment indicator, the break event symptom corresponding to the respective computer-implemented process and the final response strategy corresponding to the break event symptom of the respective computer-implemented process;
generating, by the rules engine, a risk score for each computer-implemented process of the plurality of computer-implemented processes based on the identified break event symptom and the final response strategy corresponding to the break event symptom of the respective computer-implemented process; and
storing the generated robustness and risk scores of each computer-implemented process with a timestamp indicating when the simulation result snapshot was taken in the data structure.

5. The method of claim 1 , further comprising:
in response to applying the selected final response to the simulation instance of the network environment, generating a modified robustness score of the updated robustness score and a modified risk score of the updated risk score for each computer-implemented process of the plurality of computer-implemented processes; and
storing each modified robustness score and each modified risk score in the cure result snapshot.

6. The method of claim 1 , wherein evaluating the pre-breakage snapshot, the simulation result snapshot, and the cure result snapshot with reference to one another; comprises:
identifying score changes of robustness scores and risk scores between the pre-breakage snapshot and the simulation result snapshot and changes in robustness scores and risk scores between the simulation result snapshot and the cure result snapshot for each of the computer-implemented processes;
recognizing, by evaluating the identified score changes by a rules engine, interdependencies between each computer-implemented process in the plurality of computer-implemented processes; and
in response to the recognized interdependencies, identifying an operating state of each respective computer-implemented process in the plurality of computer-implemented processes that provides a below threshold risk score for each respective computer-implemented process and an above threshold robustness score for each respective computer-implemented process.

7. The method of claim 1 , further comprising:
modifying the runbooks in the library of runbooks based on

Assignees

• Capital One Services Llc

Inventors

• Gudka Bhavik
• Barnum Eric

Classifications

• H04L41/145 Primary

  involving simulating, designing, planning or modelling of a network · CPC title

• G06F11/0793

  Remedial or corrective actions (recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40) · CPC title

• G06F11/3457

  Performance evaluation by simulation · CPC title

• H04L43/50

  Testing arrangements · CPC title

• H04L41/0631

  using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis · CPC title