Resource provisioning using predictive modeling in a networked computing environment

• US 9,860,134 B2
• Filed: 08/25/2015
• Issued: 01/02/2018
• Est. Priority Date: 08/24/2012
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Bogdany, Richard J., Harrison, Thomas B., McAvoy, Cameron Niall-Kerr, O'Connell, Brian M., Pearthree, Herbert D., Sun, Shengzhi, Upton, Clay T.
• Primary Examiner(s): Wang, Jin-Cheng
• Application Number:
  US14/834,723
• Publication Number:
  US 20150365297A1
• Time in Patent Office: 861 Days
• Field of Search: 345440, 709226
• CPC Class Codes: G06T 11/203 Drawing of straight lines or curves
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs
• CPC Class Codes: H04L 41/142 using statistical or mathematical methods
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 41/147 for predicting network behaviour

Abstract

An approach is provided for allowing a network computing (e.g., cloud computing) infrastructure to modify its resource allocation plan (e.g., an instance count) by using a Kth derivative vector plot, which may be generated using historical logs. Among other things, this approach enables an infrastructure to project an allocation forecast for a specified duration and adapt to changes in network traffic.

Claims

1. A computer-implemented method for provisioning computing resources using predictive modeling in a networked computing environment, comprising:

segmenting a set of graphical curves of network data traffic versus time generated from historical logs of network traffic into a set of predetermined time intervals to yield a set of time interval curves;

overlaying and fitting the set of time interval curves to yield a set of best fit overlaying curves;

generating a derivative vector plot based on a set of data points of the set of best fit overlaying curves;

forecasting network traffic in the networked computing environment based on the derivative vector plot; and

modifying an allocation of a set of computing resources in the networked computing environment based on the forecasted network traffic.

2. The computer-implemented method of claim 1, further comprising provisioning a set of computing resources in the networked computing environment based on the forecasted network traffic.

3. The computer-implemented method of claim 1, the derivative vector plot comprising a K^th derivative vector plot being generated by transforming the set of data points using a mathematical algorithm.

4. The computer-implemented method of claim 3, further comprising outputting a network traffic projection based on the forecasting.

5. The computer-implemented method of claim 4, the network traffic projection being generated by integrating K initial conditions a K quantity of times using a predetermined numerical technique.

6. The computer-implemented method of claim 3, the mathematical algorithm comprising a fourth order Runge-Kutta method.

7. The computer-implemented method of claim 1, the networked computing environment comprising a cloud computing environment.

8. A system for provisioning computing resources using predictive modeling in a networked computing environment, comprising:

a memory medium comprising instructions;

a bus coupled to the memory medium; and

a processor coupled to the bus that when executing the instructions causes the system to;

segment a set of graphical curves of network data traffic versus time generated from historical logs of network traffic into a set of predetermined time intervals to yield a set of time interval curves;

overlay and fit the set of time interval curves to yield a set of best fit overlaying curves;

generate a derivative vector plot based on a set of data points of the set of best fit overlaying curves; and

forecast network traffic in the networked computing environment based on the derivative vector plot; and

modify an allocation of a set of computing resources in the networked computing environment based on the forecasted network traffic.

9. The system of claim 8, the memory medium further comprising instructions for causing the system to provision a set of computing resources in the networked computing environment based on the forecasted network traffic.

10. The system of claim 8, the derivative vector plot comprising a K^th derivative vector plot, the memory medium further comprising instructions for causing the system to transforming the set of data points using a mathematical algorithm to yield the K^th derivative vector plot.

11. The system of claim 10, the memory medium further comprising instructions for causing the system to output a network traffic projection based on the forecasting.

12. The system of claim 11, the network traffic projection being generated by integrating K initial conditions of a K quantity of times using a predetermined numerical technique.

13. The system of claim 10, the mathematical algorithm comprising a fourth order Runge-Kutta method.

14. The system of claim 8, the networked computing environment comprising a cloud computing environment.

15. A computer program product for provisioning computing resources using predictive modeling in a networked computing environment, the computer program product comprising a computer readable storage device, and program instructions stored on the computer readable storage device, to:

segment a set of graphical curves of network data traffic versus time generated from historical logs of network traffic into a set of predetermined time intervals to yield a set of time interval curves;

overlay and fit the set of time interval curves to yield a set of best fit overlaying curves;

generate a derivative vector plot based on a set of data points of the set of best fit overlaying curves; and

forecast network traffic in the networked computing environment based on the derivative vector plot; and

modify an allocation of a set of computing resources in the networked computing environment based on the forecasted network traffic.

16. The computer program product of claim 15, the computer readable storage device further comprising instructions to provision a set of computing resources in the networked computing environment based on the forecasted network traffic.

17. The computer program product of claim 15, the derivative vector plot comprising a K^th derivative vector plot, the computer readable storage device further comprising instructions to transforming the set of data points using a mathematical algorithm to yield the K^th derivative vector plot.

18. The computer program product of claim 17, the computer readable storage device further comprising instructions to output a network traffic projection based on the forecasting.

19. The computer program product of claim 18, the network traffic projection being generated by integrating K initial conditions a K quantity of times using a predetermined numerical technique.

20. The computer program product of claim 17, the mathematical algorithm comprising a fourth order Runge-Kutta method.

21. The computer program product of claim 15, the networked computing environment comprising a cloud computing environment.

22. A method for deploying a system for provisioning computing resources using predictive modeling in a networked computing environment, comprising:

providing a computer infrastructure being operable to;

segment a set of graphical curves of network data traffic versus time generated from historical logs of network traffic into a set of predetermined time intervals to yield a set of time interval curves;

overlay and fit the set of time interval curves to yield a set of best fit overlaying curves;

generate a derivative vector plot based on a set of data points of the set of best fit overlaying curves; and

forecast network traffic in the networked computing environment based on the derivative vector plot; and

modify an allocation of a set of computing resources in the networked computing environment based on the forecasted network traffic.