Monitoring Studio: Calculating the Standard Deviation using the Multi-Parameter Formula

  • kb1082
  • Jul. 29, 2011
  • 2 min read

How to calculate the standard deviation using the Multi-Parameter Formula tool.

Best Practice Multi-Parameter Formula

Related Topics

Objective

This article describes how to create a multi-parameter formula in Monitoring Studio KM to calculate a standard deviation. The standard deviation is a measure of statistical dispersion, i.e. it measures the spread of data around the mean value. The more spread apart the data, the higher the deviation. A distribution will be considered as normal if most of the examples in a set of data are close to the average.

Procedure

To calculate the standard deviation:

  1. Right-click the application/container icon in the PATROL Console and select KM Commands › New › Multi-Parameter Formula...

  2. Click Next. The KMs and their respective parameters currently available are displayed:

  3. Indicate the parameter(s) to be used in the formula:
    • Select the parameter(s)
      IMPORTANT: The maximum number of parameters currently supported by Monitoring Studio KM is 26.
    • If needed, check the Use the objects display name to dynamically determine the PATROL IDs option to use the label instances instead of PATROL IDs to determine a parameter's path. This option may be particularly useful when an object ID is unknown
    • Click Next
  4. Create the formula to apply:
    • To calculate a standard deviation, enter the following formula:
      list_of_parameters = [A, B, C,..., Z]; number_of_parameters = lines(list_of_parameters); foreach current_value (list_of_parameters) { sum_of_parameters_value += current_value; } average_value = sum_of_parameters_value / number_of_parameters; foreach current_value (list_of_parameters) {average_deviation += pow((current_value - average_value), 2);} average_deviation = average_deviation / number_of_parameters; standard_deviation = sqrt(average_deviation); return standard_deviation ;
      Where the [A, B, C,…,Z] variable must be replaced with the actual list of parameters. If you only selected two parameters, replace the variable with [A, B]
    • Check the Do not collect if one or more parameter has no value option if you do not want data to be collected for parameter(s) not returning values
    • Click Next.
  5. Define the display name of the formula result:
    • Enter the object display name
    • Indicate if you want to use default or custom thresholds
    • Click Finish to start monitoring the selected parameters. A new object is automatically created under the application/container icon in the PATROL console.

Note: To edit your parameter formula, right-click the Multi-Parameter Formula icon and select KM Commands › Edit.

Best Practice Multi-Parameter Formula