Predictive Analytics for Information Technology Systems

• US 20130318022A1
• Filed: 05/09/2013
• Published: 11/28/2013
• Est. Priority Date: 05/09/2012
• Status: Active Grant

• Current Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: Yadav, Reena Dayal, Roy, Atanu Kumar, Singh, Udayan, Chisti, Safia, Kumar, Sanjeev, Srivastava, Ankur, Dey, Parthasarathy
• Granted Patent: US 9,367,803 B2
• Time in Patent Office: Days
• US Class Current: 706/46
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 10/00 Administration; Management

Abstract

Disclosed is a method and apparatus for organizing, correlating IT management data in batch mode as well as real time and doing predictive analytics to indicate possible threshold breaches, possible failures and usage bottlenecks in the systems in an information technology (IT) environment. The system and method further predicts the state of the information technology (IT) system components systematically based on current and past system states and system component states, overall systemic behavior, known system behavioral rules and load conditions /usage characteristics.

Claims

1. ) A method for predicting system behavior in an information technology environment in real time, the method comprising:

receiving and storing a stream of data representative of plurality of statistical parameters governing the system behavior, for a period of time;

applying pre-stored rules along with associated threshold values to the plurality of statistical parameters such that a deterministic pattern is obtained of the system behavior;

constructing a statistical model based on the deterministic pattern obtained from the statistical parameters of the stream of data received and stored over the period of time;

iteratively analyzing the plurality of statistical parameters for detecting any deviation from the deterministic pattern of the system behavior;

in response to the detecting of any deviation, dynamically updating the pre-stored rules, the associated threshold values and the statistical model; and

predicting the system behavior based upon the dynamically updated rules, the threshold values and the statistical model.

2. ) The method of claim 1, wherein the pre-stored rules comprises of:

a set of rules governing the system behavior in conformance to industry standards;

a set of rules defined for each of specific target environment;

or

a combination thereof.

3. ) The method of claim 1, wherein the deterministic pattern is reformed periodically based upon:

trend of incoming stream of data and the statistical parameters associated thereto;

updation in pre-stored rules and the associated threshold values in response to the detection of any deviation, or a combination thereof.

4. ) The method of claim 1, wherein the pre-stored rules are applied based on associated threshold values, wherein the threshold values are predetermined for each of the specific target environment, and adaptive to regenerate each time the system behavior is detected to deviate.

5. ) The method of claim 1, wherein the statistical parameters are analyzed for any desirable or undesirable deviation on the system behavior for a predefined period, in order to substantiate updation of the pre-stored rules, the associated threshold values and the statistical model.

6. ) The method of claim 1, wherein the statistical model is constructed using multivariate time series analysis techniques viz Vector Autoregresive (VAR) Models, Vector Error Correction Models (VECM), Structural VAR (SVAR) and the like.

7. ) A system for predicting system behavior in an information technology environment, the system comprising:

a receiving module configured to receive and store a stream of data representative of plurality of statistical parameters governing the system behavior for a period of time;

an analytics engine comprising;

a pattern recognition module configured to apply pre-stored rules along with associated threshold values to the plurality of statistical parameters such that a deterministic pattern is obtained of the system behavior;

a modeler configured to construct a statistical model based on the deterministic pattern obtained from the statistical parameters of the stream of data received and stored over the period of time;

a feedback module configured to iteratively analyze the plurality of statistical parameters for detecting any deviation from the deterministic pattern of the system behavior;

in response to the detecting of any deviation, the feedback module interacting with the analytics engine to dynamically update the pre-stored rules, the associated threshold values and the statistical model; and

a forecasting module to predict the system behavior based upon the dynamically updated rules, the threshold values and the statistical model.

8. ) The system of claim 7, wherein the analytics engine is configured to define a set of pre-stored rules comprising:

a set of rules governing the system behavior in conformance to industry standards;

a set of rules defined for each of specific target environment;

or a combination thereof.

9. ) The system of claim 7, wherein the pattern recognition module reforms the deterministic periodically based upon:

trend of incoming stream of data and the statistical parameters associated thereto;

updation in pre-stored rules and the associated threshold values in response to the detection of any deviation, or a combination thereof.

10. ) The system of claim 7, wherein the modeler employs multivariate time series analysis techniques viz Vector Autoregressive (VAR) Models, Vector Error Correction Models (VECM), Structural VAR (SVAR) and the like to construct the statistical model.

11. ) The system of claim 7, wherein the feedback module analyzes the statistical parameters for any desirable or undesirable deviation on the system behavior for a predefined period, in order to substantiate updation of the pre-stored rules, the associated threshold values upon coordinating with the analytics engine, and the statistical model upon interacting with the modeler.

12. ) The system of claim 7, wherein the feedback module coordinates with the analytics engine to regenerate pre-stored rules and the associated threshold values each time the system behavior is detected to deviate.