Stats

  • Locked Locked
  • Replies 1
  • Subscribers 125
  • Views 5015
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Starting simulations when previous one finished

NewScreenName

Dear all,

In order to optimize the use of licenses, it would be very useful to schedule simulations one after the other, i.e. setting the number of jobs, which will determine the maximum number of licenses used at any point in time, but putting in a queue any number of simulations from any number of possible different tests and different maestro views. This way, for example when some simulation finishes, the next queued will automatically start without needing any further action.

This would be especially useful when scheduling simulations from different maestro views, or when having to start new simulations in a maestro view having some simulations already running. In all those cases if I start more simulations I will take more licenses. I would like the overall virtuoso session to consume no more than the given number of jobs from the "job setup" menu.

How can this be done?

Thank you

Best regards

Parents

  • This would normally be done using a DRMS system such as LSF, SGE (or one of the various derivatives of it), NC, Flowtracer, Slurm etc. These often allow a policy that could check for license availability before actually starting the job, and this combined with the safety net of using license queuing with Spectre would mean that no job actually fails - and (in general) a slot on the compute farm won't get taken up until the prerequisites are met.

    Too much to cover in detail in the forums - perhaps you should contact customer support?

    Andrew

Reply

  • This would normally be done using a DRMS system such as LSF, SGE (or one of the various derivatives of it), NC, Flowtracer, Slurm etc. These often allow a policy that could check for license availability before actually starting the job, and this combined with the safety net of using license queuing with Spectre would mean that no job actually fails - and (in general) a slot on the compute farm won't get taken up until the prerequisites are met.

    Too much to cover in detail in the forums - perhaps you should contact customer support?

    Andrew

Children
No Data

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.