Data metric resolution prediction system and method

Patent metadata

Field	Value
Publication number	US-9245248-B2
Application number	US-201314014707-A
Country	US
Kind code	B2
Filing date	Aug 30, 2013
Priority date	Sep 28, 2012
Publication date	Jan 26, 2016
Grant date	Jan 26, 2016

Abstract

In one embodiment, a method includes receiving an identity of a metric of interest and a future time point. The method further includes retrieving a prediction configuration previously associated with the metric of interest. The prediction configuration comprising a period combination. The period combination comprises a plurality of time periods, each time period comprises one or more segments, and each segment of the one or more segments comprises adapted historical values of the metric of interest incrementally inserted therein. The method also includes, for each time period of the plurality of time periods, identifying, for the future time point, a corresponding segment of the one or more segments, accessing a set of adapted historical values from the corresponding segment, and computing an intermediate predicted value from the set of adapted historical values. Moreover, the method includes calculating a predicted value for the metric of interest based on the computing.

First claim

What is claimed is:

1. A method comprising:
receiving, by a computer system comprising computer hardware, an identity of a metric of interest and a future time point;
retrieving, by the computer system, a prediction configuration previously associated with the metric of interest, the prediction configuration comprising a period combination;
wherein the period combination comprises a plurality of time periods, each time period comprises one or more segments, and each segment of the one or more segments comprises adapted historical values of the metric of interest incrementally inserted therein;
for each time period of the plurality of time periods:
the computer system identifying, for the future time point, a corresponding segment of the one or more segments;
accessing, by the computer system, a set of adapted historical values from the corresponding segment; and
computing, by the computer system, an intermediate predicted value from the set of adapted historical values, the computing comprising:
testing for a trend in the set of adapted historical values;
based on a result of the testing, determining whether the set of adapted historical values satisfies criteria for performing linear regression;
responsive to a determination that the set of adapted historical values satisfies the criteria for performing linear regression, computing the intermediate predicated value using linear regression; and
responsive to a determination that the set of adapted historical values does not satisfy the criteria for performing linear regression, computing the intermediate predicted value based on an average of the set of adapted historical values; and

calculating, by the computer system, a predicted value for the metric of interest based on the computed intermediate predicted value.

2. The method of claim 1 , wherein the calculating comprises calculating a sum of each intermediate predicted value.

3. The method of claim 1 , wherein:
the prediction configuration comprises a prediction algorithm previously associated with the metric of interest; and
the computing comprises utilizing the prediction algorithm.

4. The method claim 3 , wherein the prediction algorithm is selected from the group consisting of: a least square linear regression algorithm, a Theil-Sen estimator, a prediction based on averages, and an exponential prediction algorithm.

5. The method of claim 1 , comprising causing the predicted value to be output.

6. The method of claim 1 , wherein, responsive to a determination that the set of adapted historical values meets specified criteria, the computing of the intermediate predicted value based on an average of the set of adapted historical values comprises calculating an average over a second half of the adapted historical values.

7. The method of claim 1 , wherein the plurality of time periods are selected from the group consisting of: hour, day, week, and month.

8. An information handling system comprising:
at least one computer processor, wherein the at least one computer processor is operable to implement a method comprising:
receiving an identity of a metric of interest and a future time point;
retrieving a prediction configuration previously associated with the metric of interest, the prediction configuration comprising a period combination;
wherein the period combination comprises a plurality of time periods, each time period comprises one or more segments, and each segment of the one or more segments comprises adapted historical values of the metric of interest incrementally inserted therein;
for each time period of the plurality of time periods:
identifying, for the future time point, a corresponding segment of the one or more segments;
accessing a set of adapted historical values from the corresponding segment; and
computing an intermediate predicted value from the set of adapted historical values, the computing comprising:
testing for a trend in the set of adapted historical values;
based on a result of the testing, determining whether the set of adapted historical values satisfies criteria for performing linear regression;
responsive to a determination that the set of adapted historical values satisfies the criteria for performing linear regression, computing the intermediate predicated value using linear regression; and
responsive to a determination that the set of adapted historical values does not satisfy the criteria for performing linear regression, computing the intermediate predicted value based on an average of the set of adapted historical values; and

calculating a predicted value for the metric of interest based on the computed intermediate predicted value.

9. The information handling system of claim 8 , wherein the calculating comprises calculating a sum of each intermediate predicted value.

10. The information handling system of claim 8 , wherein:
the prediction configuration comprises a prediction algorithm previously associated with the metric of interest; and
the computing comprises utilizing the prediction algorithm.

11. The information handling system claim 10 , wherein the prediction algorithm is selected from the group consisting of: a least square linear regression algorithm, a Theil-Sen estimator, a prediction based on averages, and an exponential prediction algorithm.

12. The information handling system of claim 8 , comprising causing the predicted value to be output.

13. The information handling system of claim 8 , wherein, responsive to a determination that the set of adapted historical values meets specified criteria, the computing of the intermediate predicted value based on an average of the set of adapted historical values comprises calculating an average over a second half of the adapted historical values.

14. The information handling system of claim 8 , wherein the plurality of time periods are selected from the group consisting of: hour, day, week, and month.

15. A computer-program product comprising a non-transitory computer-usable medium having computer-readable program code embodied therein, the computer-readable program code adapted to be executed to implement a method comprising:
receiving an identity of a metric of interest and a future time point;
retrieving a prediction configuration previously associated with the metric of interest, the prediction configuration comprising a period combination;
wherein the period combination comprises a plurality of time periods, each time period comprises one or more segments, and each segment of the one or more segments comprises adapted historical values of the metric of interest incrementally inserted therein;
for each time period of the plurality of time periods:
identifying, for the future time point, a corresponding segment of the one or more segments;
accessing a set of adapted historical values from the corresponding segment; and
computing an intermediate predicted value from the set of adapted historical values, the computing comprising:
testing for a trend in the set of adapted historical values;
based on a result of the testing, determining whether the set of adapted historical values satisfies criteria for performing linear regression;
responsive to a determination that the set of adapted historical values satisfies the criteria for performing linear regression, computing the intermediate predicated value using linear regression; and
responsive to a determination that the set of adapted historical values does not satisfy the criteria for performing linear regression, computing the intermediate predicted value based on an average of the set of adapted historical values; and

Assignees

• Dell Software Inc

Inventors

• Solomon Oren Tibi
• Kalush Israel

Classifications

• G06N5/02

  Knowledge representation; Symbolic representation · CPC title

• G06Q10/06312 Primary

  Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling · CPC title