[FLASH-USERS] Checkpoint files vs. Plot files

[John ZuHone]

Hi Marissa,

The main reason why one would want to dump plotfiles more frequently and use them in analysis is disk space. Checkpoint files dump _everything_—every grid variable, all the particles, and all of the metadata. Though you may want all the metadata, you may not want all of the grid variables—most simulations have extra variables that you likely will not use in analysis (e.g., do you really need both "eint" and "ener" to do analysis, even though the code does for evolution? Probably not).  

Plotfiles give you the ability to dump only the variables you will use in analysis more frequently than checkpoints, which are (as you said) usually thought to be only for restarts. You can also dump particle files. I typically dump both plotfiles and particle files at the same cadence. In the end, you will likely save a lot of disk space if you use plotfiles and particle files. 

Best,

John

> On Jan 18, 2017, at 2:58 PM, Marissa Adams <madams at pas.rochester.edu> wrote:
> 
> Hi all,
> 
> I am wonder if there is an advantage to using plt files in comparison to chk files as I am considering just using chk for primary output. I want to make sure I am not losing any physics by doing so.
> 
> The documentation clearly alludes to chk files being necessary for restarts. One can make the plotfileIntervalTime = checkpointFileIntervalTime, as well as have one be < and > the other, and the code still runs.
> 
> I also considered if perhaps plot/refinement variables specified in a flash.par would not necessarily be there -- all the information is there, as to be expected (re: restarts).
> 
> I suppose my questions can be well defined as, aren't plt files a little redundant if checkpoint files are sufficient, and contain more information (for restarts, particles, etc)? Are plt files more easily handled in some visualization software other than VisIt and yt (since checkpoint files are handled just fine there)?
> 
> Best,
> Marissa
> 
> 
> -- 
> Marissa Adams                                 E-mail:  <>madams at pas.rochester.edu <mailto:madams at pas.rochester.edu>
> PhD Student                                   Ph: (585) 402-5779
> University of Rochester                       Website: http://www.pas.rochester.edu/~madams/ <http://www.pas.rochester.edu/~madams/>
> Department of Physics & Astronomy
> 108 Bausch & Lomb Hall
> P.O. Box 270171

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20170118/a740bd87/attachment.htm>