[FJSJ]

Hi all,

I would like to ask you for some question about running fluent under LSF. I am a beginner with this topic, and I would like to know some aspects regarding to settup of a case.

The thing is I am preparing the .cas file on my local computer as well as I initialize the solution and, therefore, I have the .dat file. Obviously, I want to run the job on the cluster and I would like to use autosave option to save every 1000 iteration, for example. However, as I am making the settup on my local computer, the path for the autosave option, as welll as, for different monitors (drag/lift coefficient, mass flow rates, etc) are in terms of my local computer. So, my first question is, should I write down on my computer (on the autosave option) the path that it would be use on the cluster to my local directory (on the cluster)? Or it doesn´t matter because we have the standar error and standar output files.

For other hand, I suppose that If I make the settup on my computer activating autosave option every x iteration and then I save it (before copy on the cluster), when I run this job on the cluster that option would be activate. Am I right?

I know these are easy questions but as I told at the begining I am a beginner with LSF and I would like to be sure about it.

[LuckyTran]

Originally Posted by FJSJ

So, my first question is, should I write down on my computer (on the autosave option) the path that it would be use on the cluster to my local directory (on the cluster)? Or it doesn´t matter because we have the standar error and standar output files.

For other hand, I suppose that If I make the settup on my computer activating autosave option every x iteration and then I save it (before copy on the cluster), when I run this job on the cluster that option would be activate. Am I right?

It does matter how you write it. There is no standard error output.

There are two ways to enter the path. 1) is the absolute path. 2) is the relative path. If you choose the first option, you should put the entire absolute path of the cluster. Relative paths is probably the easiest, just don't put any address and just enter the filename and the file will be written into the local working directory (i.e. just put \filename.cas instead of C:\dir1\dir2\dir3\filename.cas).

Originally Posted by FJSJ

For other hand, I suppose that If I make the settup on my computer activating autosave option every x iteration and then I save it (before copy on the cluster), when I run this job on the cluster that option would be activate. Am I right?

Right, it will work that way once you put it on the cluster.

[FJSJ]

Originally Posted by LuckyTran

It does matter how you write it. There is no standard error output.

There are two ways to enter the path. 1) is the absolute path. 2) is the relative path. If you choose the first option, you should put the entire absolute path of the cluster. Relative paths is probably the easiest, just don't put any address and just enter the filename and the file will be written into the local working directory (i.e. just put \filename.cas instead of C:\dir1\dir2\dir3\filename.cas).

Hi Lucky,

Ok. So, I understand that I need to specify cluster path on my computer when I do the case settup or just write down the filename as you said in the second option.

Right, it will work that way once you put it on the cluster.

Ok. So, I don´t need to specify on the journal file autosave option as, for example:

; Autosave every 1000 iterations
auto-save/data-frequency 1000

because that option was activated when I save the .cas and .dat files.

Thank you for your help

[FJSJ]

Hi all,

I am trying to make run a journal file but I cannot get it.

The thing is, as I asked on this thread before, that I don't know what happen when you settup the case data on your own laptop (before send it to the cluster) and set autosave option to 200 iterations and then you set your number of iteration to 1000, for example. Then, save you .cas and .dat file (after initialized it). So, when you have to write down the journal file, is it necessary write down this line again on it?

; Calculate iterations
it 1000

or this one:

; Autosave
auto-save data-frequency 200

Thank you in advance!

[LuckyTran]

nothing happens to the cas file just because it got transferred from one drive to another. Your settings will be identical. It is not necessary to set the autosave frequency again, however it is not wrong to do so. In fact, it is a good habit to always set your autosave frequency in your journal file. Either way you do it, it will work. I recommend to always include autosave frequency in the journal file.