I am having a problem where I have a several parallel groups put together and waits with the minutes set to 0 between them unfortunately, it is still waiting almost a minute to start the next parallel group. For example the first group starts at 14:00:02 processes the data flow inside of it. The data flow inside of it takes 10 seconds and is finished at 14:10:02. I have a wait after the parallel group and it is set to 0 seconds which means start the next group as soon as the first one is finished. The next group will not start until 15:00:02. Does anyone know what setting will change it to not wait a minute before the next group will start? We are running 7.6.0.

Thanks!

Deanna,

Are you talking about Report Caster jobs? It sounds like it as the RC process performs a sweep every minute to identify and run processes that are due.

If you are talking RC then we need more detail to understand what you are on about.

Are these tasks all in the same schedule job?

If not, do you set the trigger after a previous job finishes?

If they are all in the same schedule job then someone like Susannah or Leah might be able to offer some assistance on timings etc. although I would imagine that if the processes are able to be run in parallel then a seperate schedule job for each task triggered at the same time might be in order as the jobs should queue OK.

More detail please.

T

T, it may have something to do with this:

http://www.freepatentsonline.com/3962706.html



(Sorry for the red herring!)

This message has been edited. Last edited by: Francis Mariani,

Tony,

Sorry, this is in the Data Migrator and not the Report Castor. We are using a scheduled job to kick off several smaller jobs and each one stringed to the last one to start after the last one finishes.

Because of timing we really don't want to submit several jobs because at some point some of the jobs are dependant on the others finishing and we are not sure during the night what time that will be.

I was able to reduce some of the wait time that we were experiencing by changing the sched_interval from 60 sec down to 15 seconds in the configuration settings. This changed the time that the process seemed to stay within the parallel group so it reduced my processing time down by half. But there is still some efficiencies that I see can be gained....I am just not sure where to look.

Deanna,

I am going to have to leave this to someone who has used DM recently - I haven't touched it for many a year

Come on folks, who uses DM at the moment and can help Deanna?

T

Francis, I see your wit is still as sharp

We found a the settings for this to improve time and I am posting for others:

Change the configuration settings on the dfm_int_max to a shorter time.

Thanks Tony for your responses and for trying to rally support.