RESOURCE CAPACITY MONITORING AND REPORTING

• US 20110270795A1
• Filed: 04/29/2010
• Published: 11/03/2011
• Est. Priority Date: 04/29/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Shah, Smiral, Mitchell, Christopher Michael, Smith, Gregory Lincoln
• Granted Patent: US 8,712,950 B2
• Time in Patent Office: Days
• US Class Current: 706/52
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals

Abstract

Embodiments provide capacity estimates for components of a computing environment, but are not so limited. In an embodiment, a computer-implemented method includes using residual capacity estimates and smooth averages to provide point-in-time capacity estimates for resources of a computing environment. In one embodiment, a computing environment includes a capacity manager that can operate under processor control to provide capacity estimates for a collection of resources as part of a capacity monitoring and planning process. Other embodiments are included and available.

Claims

1. A method comprising:

monitoring pods of a computing environment, each pod including a number of pod resources;

collecting performance data associated with the pod resources;

calculating average values of the performance data;

calculating smooth average values of the performance data;

calculating residual capacity values based in part on the average values and the smooth average values at coincident times over a given time period;

providing capacity estimates for each pod over the given time period based in part on an average of the smooth average values and a probabilistic determination using the residual capacity values; and

storing the capacity estimates in computer readable storage.

2. The method of claim 1, further comprising collecting the performance data for each pod role including processor resource consumption amounts, free disk space, and disk input/output operations per second.

3. The method of claim 2, further comprising collecting the performance data including pages per second and available memory metrics.

4. The method of claim 1, further comprising producing a trend over a trending period using the capacity estimates.

5. The method of claim 4, further comprising predicting future capacity estimates using regression and worst case parameters.

6. The method of claim 1, further comprising calculating the residual capacity values using differences between the average values and the smooth average values at each point in time over the given time period.

7. The method of claim 6, further comprising providing the capacity estimates for each pod over the given time period based in part on the average of the smooth average values and a standard deviation of the residual capacity values.

8. The method of claim 7, further comprising providing the capacity estimates for each pod over the given time period based in part on the average of the smooth average values and a ninety fifth percentile measure of the standard deviation of the residual capacity values.

9. The method of claim 7, further comprising providing the capacity estimates for each pod of the datacenter over the given time period based in part on a peak value that corresponds to a sum of the average of the smooth average values and the standard deviation of the residual capacity values, divided by pre-determined capacity maximums.

10. The method of claim 1, further comprising providing capacity trend data for the datacenter, pods, and pod roles.

11. The method of claim 10, further comprising providing future capacity data for the datacenter, pods, and pod roles.

12. The method of claim 10, further comprising providing optimistic and pessimistic capacity estimates.

13. A system comprising:

a plurality of monitored systems, each monitored system including a number of system resources;

a collector component to collect operational information associated with a select number of the monitored systems; and

a capacity manager that uses collected operational information as part an average capacity determination and a smooth average determination to provide residual capacity values that are used in conjunction with an average of the smooth average determination to provide capacity estimates for the select number of monitored systems.

14. The system of claim 13, the capacity monitor further to use a predictor to predict future capacity estimates based in part on a number of server roles including one of a hub transport role, a unified communications server role, and a client access server role.

15. The system of claim 14, the capacity monitor further to use fitting operations to determine projected capacity information.

16. The system of claim 14, further comprising a trend generator to use the number of server roles and performance data corresponding to processing capacities, memory capacities, and a number of input-output operations to provide capacity trends.

17. A computer readable medium including operational instructions to perform estimations comprising:

calculating average capacity values for a number of pods having respective pod resources;

calculating moving average capacity values for the pod resources of each pod;

calculating residuals for the pod resources of each pod using the average capacity values and the moving average capacity values; and

providing capacity estimates for the number of pods using resource capacity estimates that are based in part on probabilistic operations using corresponding residuals and an averaging of the moving average capacity values associated with the resources of each pod.

18. The computer readable medium of claim 17 including operational instructions to perform estimations further comprising providing trend data for the number of pods.

19. The computer readable medium of claim 17 including the operational instructions to perform estimations further by generating capacity projection interfaces indicating projected capacity estimates.

20. The computer readable medium of claim 17 including operational instructions to perform estimations further by storing capacity estimates as part of trending future capacity estimates.