Queries used by Data Collectors


Queries used by Data Collectors

Hi All,

I know that the Data Collectors in Viewpoint get the data from the DBQL tables, is there any way to get the queries that are actually used so that I can have a better understanding of what is going on?
Also I want to be able to manipulate the data to make my own graphs for specific time periods instead of having to rely on Viewpoint or rewinding to particular time and date.
Once I have the queries, I think I would be able to make the graphs.
Teradata Employee

Re: Queries used by Data Collectors


Actually there are several different strategies for Viewpoint data collections. These are a combination of PM/API calls, access into ResUsage tables, some SQL, and DBQL access. Many of the collectors leverage PM/API as it proves to be a better strategy for performance and reliability. Check out the Viewpoint Configuration Guide and the section for "Enabling the Data Collectors" which will identify each data collector collection strategy / mechanism. The Configuration Guide is located at:


However we would recommend a different strategy for creating your own visulizations / graphs. As of Viewpoint 14.01, just about every portlet now offers the "export" capability where you can export the reported data out of Viewpoint. Viewpoint 14.01 added export to the metrics portlets (Metrics Graph, Capacity Heatmap, Metrics Analysis). This will prove much easier than trying to duplicate the various colletions on your own. 

Hope that helps.


Re: Queries used by Data Collectors


I think you can have a look at populating QCD tables and look further ahead. If this is what you require, I can help you.



Re: Queries used by Data Collectors

Hi, hopefully this query is the correct forum.

I have found that the Metrics Analysis portlet inlcudes our FICON Nodes in calculating the NODE CPU Usage graphs and I want to remove these from the data, so as to provide our business users a better view of how the system is performing, especially when all the TPA nodes are running at 100%.

Can you tell me if it is possible to edit the SQL run against the Postgress data to remove these nodes or if there is a way we can configure the portlet to only look at Nodes with AMPS, like in the Node Resource tab?

The graph displayed in Metrics Analysis is so much easier to understand than the collection of blocks that represent the nodes in the Node Resource Portlet.