Dell Vizioncore Podręcznik Użytkownika Pobierz pdf (Strona 83)

Queries 83

Creating a Query in vFoglight

Creating an Aggregation

You can create more than one aggregation on the same query. If you do this, multiple

aggregated data objects are returned in the results of the query: one for each

aggregation, in the order specified. Leave the field blank if you do not require an

aggregation on a property.

To add an aggregation:

Click the button under Aggregation.

A new set of fields for an aggregation is added.

In the Calculate drop-down list select how you want the property field

calculated.

In the Property drop-down list, select the type of property used in the

aggregation.

Click the button with the tooltip Add Aggregation if you need to define another

aggregation.

Continue with “Identifying Values” on page 84.

If you aggregate on Metric properties, such as cpuUsage under Hosts, it creates a

data object of the type Host, with a cpuUsage property, and that property has

current, period and history properties. Each of those last three (the MetricValues

within the Metric) have been aggregated, as follows:

•For a max aggregation, it aggregates the max property of the metric values, and

stores it back into the max property in the aggregated MetricValue.

•For a min aggregation, it aggregates the min property of the metric values, and

stores it back into the min property in the aggregated MetricValue.

•For an average aggregation, it aggregates the average property of the m metric

values, and stores it back into the average property in the aggregated

MetricValue.

•For a weighted average aggregation, it aggregates the weighted average property

of the metric values, and stores it back into the average property in the aggregated

MetricValue. The count property of a metric is used for the weight. For example,

if the collected metric is Response Time, its average should be calculated by

weighting the various observed values by how many times each different value

occurred. On the other hand, if the system is monitoring two hosts, a simple

average of CPU usage is appropriate to guard against the case where the sample

rates for collecting data on the two hosts are different. If sample counts were used

1 2 ... 78 79 80 81 82 83 84 85 86 87 88 ... 167 168

Brak uwag

Dell Vizioncore Podręcznik Użytkownika Strona 83