Q:

How does the SACS multirun option work?

 

R:

Within an analysis, let’s say you have different subdivisions. When you put the multirun, the analysis sequence is a parallel type of run. Say for example in the following snap, you can see the first run has moved from "pre processor" mode to "Solve" mode however the second and the third run are still in the pre processor mode. So, eventually, the first run will complete, and the other runs will continue and so on.

 

If you want to utilize the post processor run, then while creating the run file you need to call the SACS common solution file from the previous run. Hence, unless you do the previous analysis, you do not have the common solution file so eventually you cannot create the post processing run file in the analysis generator. So, this type of combination of multi run is not possible.

However, if there are two different types of runs but not interrelated like what is mentioned above, then SACS will follow a series of analysis instead of parallel run. For example, let’s say I have a model file for basic static run and an Eigen run. As these two runs are independent and do not require any output from the first one to be used for the other, here I can create the multi run. However, depending on the order of my multi run file selection, one run will complete and then the other run will start. So basically, for heterogeneous analysis, the multi run option will follow series sequence.

So, as a summary

Case1: Multiple files with similar run> Possible and run will take in parallel synchronization 

Case2: Multiple files with different runs but no interconnectivity> Possible but run will take in series

Case3: Multiple files with different runs with interconnectivity> Not possible.

 

About the report of warning/error, it will happen just the same way SACS does for a single run. If there is any issue, the program will return an error and then will skip to the next run in the multi run sequence.