This page describes how automatically calculates the baseline performance for your applications. After it establishes these prevailing performance characteristics, it can detect anomalous conditions for your application. uses baselines to:

How Baselines Work

calculates baselines by using the underlying hourly data. A baseline can be calculated for every metric. Baselines have two main variables:

Use Baselines to Identify Trends

Performance expectations can differ between hours of the day, days of the week, or even months of the year. For example:

To account for this variation, you can use a rolling time period as the baseline context. A rolling time period establishes the baseline against data from the current hour taken at a daily, weekly, or monthly interval over the course of the time period.

For example, you can have a baseline that uses a daily trend with a time period configured for 30 days or a weekly trend with a time period of 90 days.

A daily trend calculates the baseline from the data accumulated at the same time each day over the last 30 days.

30-day baseline diagram

A weekly trend calculates the baseline from the data accumulated on the same hour and day of the week over the last 90 days. This is one of the available default configurations.

Weekly 90-day Baseline Diagram

A monthly trend calculates the baseline from data accumulated at the same hour and same day of the month. So, for example, on January 5th at 10:30 AM, the baseline is established based on data accumulated at the same hour on the 5th of each month for the prior 365 days: 

Monthly 365-day baseline diagram

Baselines are not available immediately upon startup. It takes time and application load for the  to collect data and create its initial baselines. The amount of time depends on the baseline type being used: daily, weekly, monthly, or none. Using none requires several hours before a baseline is available, daily takes several days, weekly takes several weeks, and so on.

By default, baselines are not computed for metrics if there are fewer than 20 calls per minute. To configure the calls per minute threshold, contact customer support.

How are Standard Deviations Calculated

This standard formula is used to calculate the standard deviation:

standard deviation = sqrt [(B - A^2/N)/N]

where: 
A = sum of the data values
B = sum of the squared data values
N = number of data values

The standard deviation calculations differ based on the type of metric used:

This example shows a standard deviation using metric types of:

Metric - Calls Per Minute (Average, Collective)
(Hourly Metric Table)
TimeSumCountWeight-ValueWeight-Value-Squared
09:00
1802180300
10:00
3002300780
11:00
1502150195

For these metric values, the baseline of type = All Data lasts three hours (at 12:00) using the above values for calculations.

With daily, weekly, or monthly seasonality, the data points considered are values for the same hour for all days, the same day of the week, or the same day of the year, respectively.

In this case, the baseline calculations will be:

Value = (180+300+150)/360 = 1.75

Standard deviation = sqrt(((300+780+195) - (((180+300+150)^2)/360))/360) = 0.69

Configure and View Baselines

A baseline deviation is the standard deviation from a baseline at a point in time, represented as an integer value. You can set health rule conditions based on baseline deviations. For example, you can configure a warning condition as two standard deviations from the baseline and a critical condition as four standard deviations from the baseline. 

To configure baselines:

  1. Navigate to Configuration > Baselines.

  2. Click Set as Default to choose another baseline in the baseline list as default.

    Changing the default baseline changes the actual baseline for all existing health rules that rely on this default and do not specify another baseline. Be watchful of your existing baselines and health rule definitions before you select this option.

  3. Select the trend type to use.

  4. Set the trend time period and the trend. The base time period is available in these two ranges: 

Daylight Saving Time for Baseline Calculation

The baseline calculation takes into account the daylight saving time. If your region has a daylight saving, contact your Accounts team to configure the baseline.read.time.zone property for your account.

On the http://<controller host>:<port>/controller/admin.jsp page, the Accounts team need to add the baseline.read.time.zone property and one of the following values based on your region: