Querying tidal tables to get CPU/memory utilisation trend

[septemberhues]

Hi Everyone,

Are there any tables in Tidal database, querying which one can get CPU and memory utilisation of every scheduler currently in execution status?  If not, can this information be retrieved from any other source, to enable the results to be used for reporting purpose? 

[Prakash Hemchand]

Hi September,

By 'scheduler' do you mean agents?  In any case, I don't think TES gathers or stores that information. 

Most likely your company's system administrator can guide you towards the best way to retreive and store that info in your environment.

[Joe Fletcher]

Hi,

This is normally something you might do with a monitoring tool (Tivoli, OpenView, BMC Patrol etc). We have such a tool which will track individual processes and alert on thresholds which we define eg number of processes, resource utilization and so on. We group the monitors together so we have a dashboard for the various process groups.

Its all quite easy to do with the right tool.

Cheers

[Marc Clasby]

Bad News

I don't see any path to get the statistics you want from the database back-end in 5.31 but in 6.1 I see there is a jobrun_cputime but the values are all null for any executed jobs. I also don't see it exposed as a built in variable (job)

Good News

Since the column is there then in theory this on their roadmap for a future release (maybe 6.2 ?)

JAWS might be a potential solution if variable is not exposed (down the road) ... We haven't bought Jaws yet.

Marc