Chapter 2: Checkpointing and Restarting

LSF provides utilities to save (i.e., checkpoint), and restart an application. The ANSYS Fluent and LSF integration allows ANSYS Fluent to take advantage of the checkpoint and restart features of LSF. At the end of each iteration, ANSYS Fluent looks for the existence of a checkpoint or checkpoint-exit file. If ANSYS Fluent detects the checkpoint file, it writes a case and data file, removes the checkpoint file, and continues iterating. If ANSYS Fluent detects the checkpoint-exit file, it writes a case file and data file, then exits. LSF’s bchkpnt utility can be used to create the checkpoint and checkpoint-exit files, thereby forcing ANSYS Fluent to checkpoint itself, or checkpoint and terminate itself. In addition to writing a case file and data file, ANSYS Fluent also creates a simple journal file with instructions to read the checkpointed case file and data file, and continues iterating. ANSYS Fluent uses that journal file when restarted with LSF’s brestart utility. For more details on checkpointing features and options within ANSYS Fluent, see the ANSYS Fluent User's Guide.

The greatest benefit of the checkpoint facilities occurs when it is used on an automatic basis. By starting jobs with a periodic checkpoint, LSF automatically restarts any jobs that are lost due to host failure from the last checkpoint. This facility can dramatically reduce lost compute time, and also avoids the task of manually restarting failed jobs.

For more information, see the following sections:


Release 18.2 - © ANSYS, Inc. All rights reserved.