Method for measuring web site performance

• US 8,090,679 B2
• Filed: 06/30/2008
• Issued: 01/03/2012
• Est. Priority Date: 06/30/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Zhuge, Yue, Wu, Kun
• Primary Examiner(s): LE, THU NGUYET T
• Application Number:
  US12/165,600
• Publication Number:
  US 20090327353A1
• Time in Patent Office: 1,282 Days
• Field of Search: 707/600, 707/999.01, 707/603, 707/736
• US Class Current: 707/603
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet

Abstract

An approach for systematically and objectively assessing the health of a web site, or other complex system, using two layers of information. A scoring method is used to provide an objective qualitative judgment, together with detailed areas of analysis if desired. A set of metrics is identified, and a health score for each metric is calculated based upon assessment parameters or angles, based upon the actual data and upon the expectations. The overall health score of the system is then calculated as a composite metric based upon the individual metric health scores. A total score, together with a visual representation of the individual metric scores, is presented to the end user. The two layers of information utilized in the general framework include the domain (which determines the set of metrics and assessment parameters, e.g., web site analysis) and the model (or processing desired, e.g., a scoring model or alert model).

Claims

1. A method for determining a score indicating an assessment of web traffic for a selected website over a selected timeframe, the method comprising:

identifying a plurality of independent metrics with regard to web traffic, each metric indicating a performance indicator of the selected website;

receiving user input including an expected metric value for each metric of the plurality of metrics;

obtaining, from a data store, historical data including web traffic information for the selected website and web traffic information for a peer website;

evaluating the web traffic information for the selected website to determine an actual metric value for each of the plurality of metrics with regard to the selected website;

evaluating the web traffic information for the peer website to determine a corresponding metric value for each of the plurality of metrics with regard to the peer website;

calculating, in accordance with a model, a parameter health score for each of a plurality of assessment parameters for each metric, wherein the assessment parameters provide different assessments of each metric based upon the actual metric value for each metric, the different assessments including;

a value assessment of the metric that compares the actual metric value for the metric against the expected value for the metric,

a historical assessment of the metric that compares the actual metric value for the metric against historical data for the metric, and

a peer assessment of the metric that compares the actual metric value for the metric to a corresponding metric value with regard to the peer website;

calculating, in accordance with the model, a metric health score for each metric of the plurality of metrics based upon averaging the parameter health scores calculated for the plurality of assessment parameters for each metric; and

determining, in accordance with the model, a composite overall health score assessing web traffic for the selected website based upon the metric health scores calculated for the plurality of metrics.

2. The method of claim 1, wherein:

the model is a scoring model having a weighted average function, and

the composite overall health score is determined by using the metric health scores calculated for the plurality of metrics as inputs to the weighted average function of the scoring model.

3. The method of claim 1, wherein said metrics include page views, unique users, session counts, and time spent on the selected website.

4. The method of claim 1, wherein said historical assessment measures growth of number of users.

5. The method of claim 4, wherein:

the peer assessment compares growth of number of users for the selected website to growth of number of users for the peer website.

6. The method of claim 1, wherein:

the model is an alert model having a rule based function, and

the composite overall health score is determined by using the metric health scores calculated for the plurality of metrics as inputs to the rule based function of the alert model.

7. The method of claim 1, wherein the parameter health score for each assessment parameter for each metric is calculated based upon the actual metric value for the metric and the expected metric value for each metric.

8. The method of claim 1, further comprising:

providing a visual graph including the composite overall health score for the selected website to a display.

9. The method of claim 8, wherein the metric health scores calculated for each of the plurality of metrics are included in the visual graph.

10. The method of claim 8, wherein the composite overall health score for the selected website is determined according to a scoring model when the selected website is a static website and determined according to an alert model when the selected website is a dynamic website.

11. The method of claim 8, wherein the composite overall health score, the metric health scores, the parameter health scores are graphically displayed in a tabular form to an end user.

12. A computer storage medium storing computer-executable instructions that, when executed by a processor, implement a framework for assessing performance of a server system over a period of time, the framework comprising:

a domain dictating a set of independent metrics and assessment parameters for each metric of the set of metrics; and

a model for interpreting the set of metrics and assessment parameters in the domain, wherein;

the framework includes the ability to define the metrics and the ability to control contributions from each metric and from each assessment parameter,

the model is used to calculate a parameter health score for each assessment parameter, a metric health score for each metric, and a composite overall health score for the server system,

the assessment parameters provide different assessments of each metric based upon an actual metric value for each metric,

the different assessments include a value assessment of the metric that compares the actual metric value for the metric against an expected value for the metric, a historical assessment of the metric that compares the actual metric value for the metric against historical data for the metric, and a peer assessment of the metric that compares the actual metric value for the metric to a corresponding metric value with regard to a peer server system,

the metric health score for each metric is based upon averaging the parameter health scores calculated for the assessment parameters for each metric, and

the composite overall health score is calculated based upon the metric health scores calculated for the set metrics.

13. The computer storage medium of claim 12, wherein said domain comprises web site analysis.

14. The computer storage medium of claim 12, wherein said model comprises one or more of a scoring model and an alert model.

15. The computer storage medium of claim 14, wherein when said model comprises a scoring model, different parameters are used to produce different sub-models.

16. A system architecture for assessing performance of a server system over a selected timeframe, the system architecture comprising:

a front end for receiving inputs and presenting final results to end users, the inputs including a domain description, model definitions of a model, model parameters of the model, expected metric values, and libraries to support computations;

data stores for storing data including input received from the front end, historical data, data to compute metrics, and data to support assessment of assessment parameters for each metric; and

processing modules for;

obtaining actual metric values for a plurality of independent metrics,

calculating, in accordance with the model, a parameter health score for each assessment parameter for each metric, wherein the assessment parameters provide different assessments of each metric based upon the actual metric value of each metric, the different assessments including a value assessment of the metric that compares the actual metric value for the metric against an expected value for the metric, a historical assessment of the metric that compares the actual metric value for the metric against historical data for the metric, and a peer assessment of the metric that compares the actual metric value for the metric to a corresponding metric value with regard to a peer server system,

calculating, in accordance with the model, a metric health score for each metric based upon averaging the parameter health scores calculated for the assessment parameters for each metric, and

determining, in accordance with the model, a composite overall health score for the server system based upon the metric health scores calculated for the metrics.

17. The system architecture of claim 16, wherein said front end includes a display mechanism on which the final results are presented.

18. The system architecture of claim 16, wherein the domain description determines a set of metrics and a set of assessment parameters for each metric.

19. The system architecture of claim 16, wherein the domain comprises web site analysis, and the model comprises a scoring model.

20. The system architecture of claim 16, wherein the domain comprises web site analysis, and the model comprises an alert model.