Filtering Results According to Time Dimensions
It is very important to understand how aggregated data is stored in the TIBCO Service Performance Manager hierarchy tables (and, in turn, in LiveView tables) across time dimensions. Based on the time dimensions defined in the SPM schema file, the measurements are computed across every level of time dimension.
For example, if the time dimensions are weeks, days, hours, and minutes, then all the measurements are computed and stored for every week, every day of the week, every hour of the day, and every minute of the hour. In the LiveView table, datetime value is stored against every time dimension for every computed row. If a time dimension value is not present for a row, then null is stored against that cell.
Consider an example of the SPM Server with AMX_3_0_SPM_Schema.xml integrated with the LiveView project sample. To represent measurement value, such as AvgJVMTotalMemory computed for the week level, only the week value in the row has a datetime value. Other time dimensions, such as days, hours, and minutes have null values.
Note the weeks dimension has a datetime value, but days, hours, and minutes have null values. Similarly, for identifying a day, week and day values must have datetime values, but hours, minutes must be null, and so on.