Jobs and schedules - how do they work with each other?

Question

could please someone enlighten me?

I have a data set and I need this data set to be refreshed on the 5th and 15th of the month. But there can only be one schedule per Cognos object, right?

Is this the optimal solution?

I setup a schedule on the data set itself for the 5th of Month.

Then I create a job and add that data set as a job step.

I schedule the job containing that data set (which includes its own schedule) for the 15th.

What happens on the 15th and how these two schedules work with each other?

(This could also apply to reports etc).

Many thanks

I went through the documentation and some online articles, but there is no definitive answer if a data set (or a report..) can have multiple schedules set.

Answer 1

Lots here about reports and report views explaining how I handle schedules. Substitute "dataset" and "job" and I think you'll get what you want.

One schedule per object leads to limitations. I have found for reports that it is better to create a report view to handle the schedule. I very rarely schedule a report anymore because I often end up needing more than one schedule or set of parameters.

Using a report view provides a means to not only create multiple schedules per report, by creating multiple report views, but also have those schedules send the reports based on independent sets of parameters. For example, the regional VPs want reports showing their sales revenue for the month, but only for their regions. I would create one report, then create a report view for the U.S. and a report view for Europe. Both get sent on the 5th of the month.

Likewise, I could set up 2 report views to work around the scheduling limitations. Say I want to send the same report on the 5th and the 15th. I would create the report, then create 2 report views to handle the schedules.

So, to translate this to your need...
To refresh the same dataset on the 5th and the 15th, create 2 jobs that have the dataset as a step -- one job for each schedule.

Using jobs looks like the right thing to do for your use case. I don't see another scheduling option for datasets.