[97] | 1 | .\" |
---|
| 2 | .\" Copyright (c) 2004-2007 The Trustees of Indiana University and Indiana |
---|
| 3 | .\" University Research and Technology |
---|
| 4 | .\" Corporation. All rights reserved. |
---|
| 5 | .\" Copyright (c) 2008 Sun Microsystems, Inc. All rights reserved. |
---|
| 6 | .\" |
---|
| 7 | .\" Man page for OMPI's ompi-restart command |
---|
| 8 | .\" |
---|
| 9 | .\" .TH name section center-footer left-footer center-header |
---|
| 10 | .TH OMPI-RESTART 1 "Dec 08, 2009" "1.4" "Open MPI" |
---|
| 11 | .\" ************************** |
---|
| 12 | .\" Name Section |
---|
| 13 | .\" ************************** |
---|
| 14 | .SH NAME |
---|
| 15 | . |
---|
| 16 | ompi-restart, orte-restart \- Restart a previously checkpointed parallel job |
---|
| 17 | using the Open PAL Checkpoint/Restart Service (CRS) |
---|
| 18 | . |
---|
| 19 | .PP |
---|
| 20 | . |
---|
| 21 | \fBNOTE:\fP \fIompi-restart\fP, and \fIorte-restart\fP are all exact |
---|
| 22 | synonyms for each other. Using any of the names will result in exactly |
---|
| 23 | identical behavior. |
---|
| 24 | . |
---|
| 25 | .\" ************************** |
---|
| 26 | .\" Synopsis Section |
---|
| 27 | .\" ************************** |
---|
| 28 | .SH SYNOPSIS |
---|
| 29 | . |
---|
| 30 | .B ompi-restart |
---|
| 31 | .R [ options ] |
---|
| 32 | .B <GLOBAL SNAPSHOT HANDLE> |
---|
| 33 | . |
---|
| 34 | .\" ************************** |
---|
| 35 | .\" Options Section |
---|
| 36 | .\" ************************** |
---|
| 37 | .SH Options |
---|
| 38 | . |
---|
| 39 | \fIompi-restart\fR will attempt to restart a previously checkpointed parallel |
---|
| 40 | job from the global snapshot handle reference returned by \fIompi_checkpoint\fP. |
---|
| 41 | . |
---|
| 42 | .TP 10 |
---|
| 43 | .B <GLOBAL SNAPSHOT HANDLE> |
---|
| 44 | The global snapshot handle reference returned by \fIompi_checkpoint\fP, used to |
---|
| 45 | restart the job. This is required to be the last argument to this command. |
---|
| 46 | . |
---|
| 47 | . |
---|
| 48 | .TP |
---|
| 49 | .B -h | --help |
---|
| 50 | Display help for this command |
---|
| 51 | . |
---|
| 52 | . |
---|
| 53 | .TP |
---|
| 54 | .B -p | --preload |
---|
| 55 | Preload the checkpoint files on the remote systems before restarting the |
---|
| 56 | application. Disabled by default. |
---|
| 57 | . |
---|
| 58 | . |
---|
| 59 | .TP |
---|
| 60 | .B --fork |
---|
| 61 | Fork off a new process, which is the restarted process. By default, the |
---|
| 62 | restarted process will replace \fIompi-restart\fR. |
---|
| 63 | . |
---|
| 64 | . |
---|
| 65 | .TP |
---|
| 66 | .B -s | --seq |
---|
| 67 | The sequence number of the checkpoint to restart from. By default, the most |
---|
| 68 | recent sequence number is used (specified by -1). |
---|
| 69 | . |
---|
| 70 | . |
---|
| 71 | .TP |
---|
| 72 | .B -hostfile | --hostfile |
---|
| 73 | The hostfile from which to restart the application. Useful in unscheduled |
---|
| 74 | environments. (Same behavior as --machinefile option) |
---|
| 75 | . |
---|
| 76 | . |
---|
| 77 | .TP |
---|
| 78 | .B -machinefile | --machinefile |
---|
| 79 | The machinefile from which to restart the application. Useful in unscheduled |
---|
| 80 | environments. (Same behavior as --hostfile option) |
---|
| 81 | . |
---|
| 82 | . |
---|
| 83 | .TP |
---|
| 84 | .B -v | --verbose |
---|
| 85 | Enable verbose output for debugging. |
---|
| 86 | . |
---|
| 87 | . |
---|
| 88 | .TP |
---|
| 89 | .B -gmca | --gmca \fR<key> <value>\fP |
---|
| 90 | Pass global MCA parameters that are applicable to all contexts. \fI<key>\fP is |
---|
| 91 | the parameter name; \fI<value>\fP is the parameter value. |
---|
| 92 | . |
---|
| 93 | . |
---|
| 94 | .TP |
---|
| 95 | .B -mca | --mca <key> <value> |
---|
| 96 | Send arguments to various MCA modules. |
---|
| 97 | . |
---|
| 98 | . |
---|
| 99 | .\" ************************** |
---|
| 100 | .\" Description Section |
---|
| 101 | .\" ************************** |
---|
| 102 | .SH DESCRIPTION |
---|
| 103 | . |
---|
| 104 | .PP |
---|
| 105 | \fIompi-restart\fR can be invoked multiple, non-overlapping times. This |
---|
| 106 | allows the user to restart a previously running parallel job. |
---|
| 107 | . |
---|
| 108 | . |
---|
| 109 | .\" ************************** |
---|
| 110 | .\" See Also Section |
---|
| 111 | .\" ************************** |
---|
| 112 | . |
---|
| 113 | .SH SEE ALSO |
---|
| 114 | orte-ps(1), orte-clean(1), ompi-checkpoint(1), opal-checkpoint(1), opal-restart(1), opal_crs(7) |
---|
| 115 | . |
---|