Numeric attribute options
When you specify a numeric attribute, you must specify the size, purpose, scale, and range of the attribute. Table 7 contains descriptions of each of these aspects of an attribute.
Numeric aspects | Options and fields | Description |
---|---|---|
Size | 32 bits
64 bits |
The value of 32-bit numbers can range
from -2147483648 to 2147483647 (roughly negative two billion to two
billion).
The value of 64-bit numbers can range from -9223372036854775808 to 9223372036854775807 (roughly negative nine quintillion to nine quintillion). |
Purpose | Gauge | Integer values where the raw values returned are higher or lower than previous values. Negative values are supported. This type is the default type for integers. Data aggregation in the warehouse produces minimum, maximum, and average values. |
Counter | A positive integer value containing raw values
that generally increase over time. Data aggregation in the warehouse
displays the total, high, low, and latest delta values. In the following
example of the Delta-based calculations, where the detailed data values
in one hour are 9, 15, 12, 20, 22, delta-based processing has the
following rules:
|
|
Property | A property of the object that does not frequently change. Data aggregation in the warehouse displays the latest value collected during the period. | |
Delta | An integer value representing the difference between the current value and the previous value for this attribute. Because this attribute is represented as a gauge in the warehouse, data aggregation in the warehouse produces minimum, maximum, and average values. | |
Percent change | An integer value that represents the percent change between the current value and the previous value. This type is calculated as: ((new -old)*100)/old. Because this type is represented as a Gauge in the warehouse, data aggregation in the warehouse produces minimum, maximum, and average values. | |
Rate of change | An integer value representing the difference between the current value and the previous value divided by the number of seconds between the samples. It converts a value (such as bytes) to the value per second (bytes per second). Because this type is represented as a Gauge in the warehouse, data aggregation in the warehouse produces minimum, maximum, and average values. | |
Scale | Decimal adjustment | Scale determines how many decimal
places are in the number. Each decimal place reduces the range mentioned
above by a factor of 10. For example, a decimal adjustment of 2 shows
two decimal places, and in a 32-bit number the allowable range becomes
-21474836.48 to 21474836.47.
When a non-zero decimal adjustment is specified, the number is manipulated internally as a floating point number. Therefore, the precision of large 64-bit numbers might be reduced. |
Range | Minimum
Maximum |
Range gives the expected range of the value. If no minimum or maximum ranges are given, the maximum values described above are used. The range is used to produce a more useful initial view in some graphical IBM Tivoli Monitoring workspace views. |