Hi YLin, proceed as if you have a single file, just provide the name of the folder for the restart and nektar will take care of the rest. same for FieldConvert. so for example, if you have problemName_100.chk folder which has several files inside, you can use "problemName_100.chk" as the name for the restart file in the XML file or alternatively you can rename the folder to any name you wish and use that name. Just do not change anything inside the folder. Nektar++ has two output format, XML output and Hdf5. You are currently using the XML one that produces a folder. However, if you use the hdf5 format, all those partitions will be written in a single file as a separate group I presume. Nevertheless, you will see a single file, e.g. problemName_100.chk but this time it is a file rather than a folder. I do recommend using the Hdf5 file format since handling the files (especially for transfer) is much easier. Additionally, most supercomputing systems have a limitation on the number of files which can be avoided using hdf5 format. Note that no matter which format you use, the output file/folder extension is .chk or .fld. To use the Hdf5 format, you must build the Nektar++ with MPI support and in the options select "HDF5" and turn on the Thirdparty_build_hdf5 Any further operations on the file are similar to a single XML file. One more thing, if you use FieldConver in serial, the output will be a single file irrespective of XML/hdf5 format or serial/parallel run. However, if you run FieldCovnert in parallel, I know that it will produce several files for the vtk format where you load the main one in paraview and it will load the solution. I am not sure about the .dat format. you can have a look at the userguide. Cheers, Mohsen ________________________________ From: nektar-users-bounces@imperial.ac.uk <nektar-users-bounces@imperial.ac.uk> on behalf of Paul Liu <liuyilin.liu@gmail.com> Sent: Wednesday, February 23, 2022 7:59 PM To: nektar-users <nektar-users@imperial.ac.uk> Subject: [Nektar-users] Restarting and post-processing multi-threaded jobs This email from liuyilin.liu@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list<https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address. Hi Everyone, Upon running a nektar++ AcousticSolver job across multiple CPUS, i noticed that the outputs for each check point are partitioned into separate .fld files (each corresponds to a single node). Stored under say "problemName_100.chk" folder. My question is how might I restart a simulation given the partitioned mesh data at each checkpoint and how to post-process the results into a single .dat file using FieldConvert. I have only noticed the instructions for single outputs. Is there some kind of concatenation required? Thank you. YiLin Liu 403-397-3457 liuyilin.liu@gmail.com<mailto:liuyilin.liu@gmail.com> MASc Student at CML<https://cml.mech.ubc.ca/> UBC