Working with correlation maps


The correlation analyses are built on the correlation maps. When you create any of the correlation analyses, the correlation map tab is displayed. 

A correlation map shows the association between business driver metrics or performance metrics inspected in a single analysis. The association between different business driver metrics (performance metrics) is used to evaluate the correlation between the metrics. This correlation helps with understanding the dynamics of the application. For details, see Correlating-resource-utilization-for-systems-and-business-drivers.

To configure a correlation map

  1. In Workspace, select the domain for which you want to analyze data.
  2. Click Works to access the domain's Works folder.
  3. Create any of the correlation analysis, as described in Creating-an-analysis.
  4. On the analysis detail page, select the Correlation Map tab to edit the associated correlation map.

At the first access, the map is empty.

You can perform the following actions in a map:

  • Add associations:  Add associations manually. For details, see add associations.
  • Edit associations: add or remove existing associations from the map.
  • Analyze: Performs an analysis on the defined associations, calculating the correlations and other parameters. 
  • Copy: Import a set of associations from another analysis for the same domain.

To add associations to the map 

  1. Click Add associations.
  2. On the Add associations to the correlation map page, perform these steps:
    1. Provide the following details for business drivers:
      • Specify whether you want to use the selected time series metrics or metrics from other time series
      • Workload name
      • Metric and subresource
      • Namespace (only available option is automatic, which indicates that BMC Helix Continuous Optimization selects the appropriate namespace.)
    2. Provide the following details for systems:
      • Specify whether you want to use the selected time series metrics or metrics from other time series
      • System name
      • Metric and subresource
      • Select one of the following namespaces:

        22.3_namespace_correlation.png

        • Automatic - BMC Helix Continuous Optimization selects the appropriate namespace. This is the default selection.
        • opt:meas: Used for the metrics that are collected from the hypervisors. For example, metrics collected by the VMware ETL.
        • opt:meas:vn: Used for the metrics that are collected from the operating systems. For example, metrics collected by the Continuous Optimization Agents.
        • opt:der: Used for the derived metrics.

        The namespace option is available for the following types of analysis: Performance versus load, performance versus performance, and load versus load

  3. Click Add. The correlation map is automatically updated on the page.
  4. When you are done adding all associations, click Cancel or close the box.

After adding initial associations, the correlation map helps you understand system behavior by means of the automatically calculated indicators. For details, see Correlating-resource-utilization-for-systems-and-business-drivers.

To edit or remove associations from the map

  1. Click Edit associations.
    The correlation map is displayed in edit mode.
  2. Make the required changes.
  3. Click Save.

To analyze a map

Perform an analysis of the defined associations, calculating the correlations and other parameters. 

  1. Click Analyze.
    A preview of the results can be accessed from the indicators, such as Correlations and Samples size.
  2. To permanently store the analysis results, click Save.

On the top right corner of the map you can find information such as:

  • The date and time of the last refresh of the map.
  • The status of the estimations.

The following table provides additional information about the displayed results.

Example of a chart obtained from the analysis of a correlation map

pla_example.png

Note for advanced users

In a multivariate case, the solution of the correlation map analysis is a hyperplane, that cannot be represented by a X-Y 2D chart. The chart of a multivariate case is the projection of the N-dimensional U/W cloud on a U/W1 2-D plane. As a consequence, the linear regression expressed by the multivariate model cannot fit visually with the service demand line for this projection; however, the mathematical estimation still remains correct.

To copy the correlation map

Import a set of associations from another analysis for the same domain.

  1. Click Copy.
    Analyses available for the same domain are displayed in a list.
  2. Select the analysis from which you want to copy the correlation map, and click Copy.
    A set of associations are imported.

Related topics

Analyzing-domain-data

Correlating-resource-utilization-for-systems-and-business-drivers

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*