Granularity versus Poll Interval

Technote (troubleshooting)


Problem(Abstract)

Reports do not show continuous data but the report shows data spikes

Cause

Poll interval is more than the Granularity

Diagnosing the problem

The CSV report shows data identifier is 1

Resolving the problem

Poll interval should always be less than the Granularity. Otherwise, the same data should be displayed at different data points.

Example 1:
Poll Interval: 15 min and the Granularity is: 5 mins.

Then the data at the following intervals are:
5 ==> no data
10 ==> no data
15 ==> data collected at the first poll interval (some time it gets delayed and the same data may displayed at the next Granularity point).
20 ==> 15 min data
25 ==> 15 min data
30 ==> second poll interval


Hence the Poll interval should be less than or equal to the Granularity.


Example 2:

Poll Interval: 5 min and the Granularity is : 15 mins.

Then the data at the following intervals are:
15 ==> aggregated data of first three poll intervals and it will be done by dataview
30 ==> aggregated data of first six poll intervals and it will be done by dataview.
45 ==> aggregated data of first nine poll intervals and it will be done by dataview.



In the CSV report, data identifier 1 means that the granularity is set to less than the polling period 


Rate this page:

(0 users)Average rating

Add comments

Document information


More support for:

Netcool/Proviso
DataView

Software version:

4.4.3

Operating system(s):

AIX, Solaris

Reference #:

1402719

Modified date:

2009-09-14

Translate my page

Machine Translation

Content navigation