[97] | 1 | .\" |
---|
| 2 | .\" Man page for OMPI's CRCP Functionality |
---|
| 3 | .\" |
---|
| 4 | .\" .TH name section center-footer left-footer center-header |
---|
| 5 | .TH OMPI_CRCP 7 "Dec 08, 2009" "1.4" "Open MPI" |
---|
| 6 | .\" ************************** |
---|
| 7 | .\" Name Section |
---|
| 8 | .\" ************************** |
---|
| 9 | .SH NAME |
---|
| 10 | . |
---|
| 11 | Open MPI MCA Checkpoint/Restart Coordination Protocol (CRCP) Framework \- Overview of Open MPI's CRCP |
---|
| 12 | framework, and selected modules. Open MPI 1.4 |
---|
| 13 | . |
---|
| 14 | .\" ************************** |
---|
| 15 | .\" Description Section |
---|
| 16 | .\" ************************** |
---|
| 17 | .SH DESCRIPTION |
---|
| 18 | . |
---|
| 19 | .PP |
---|
| 20 | The CRCP Framework is used by Open MPI for the encapsulation of various |
---|
| 21 | Checkpoint/Restart Coordination Protocols (e.g., Coordinated, Uncoordinated, |
---|
| 22 | Message/Communication Induced, ...). |
---|
| 23 | . |
---|
| 24 | .\" ************************** |
---|
| 25 | .\" General Process Requirements Section |
---|
| 26 | .\" ************************** |
---|
| 27 | .SH GENERAL PROCESS REQUIREMENTS |
---|
| 28 | .PP |
---|
| 29 | In order for a process to use the Open MPI CRCP components it must adhear to a |
---|
| 30 | few programmatic requirements. |
---|
| 31 | .PP |
---|
| 32 | First, the program must call \fIMPI_INIT\fR early in its execution. |
---|
| 33 | .PP |
---|
| 34 | The program must call \fIMPI_FINALIZE\fR before termination. |
---|
| 35 | .PP |
---|
| 36 | A user may initiate a checkpoint of a parallel application by using the |
---|
| 37 | ompi-checkpoint(1) and ompi-restart(1) commands. |
---|
| 38 | . |
---|
| 39 | .\" ********************************** |
---|
| 40 | .\" Available Components Section |
---|
| 41 | .\" ********************************** |
---|
| 42 | .SH AVAILABLE COMPONENTS |
---|
| 43 | .PP |
---|
| 44 | Open MPI currently ships with one CRCP component: \fIcoord\fR. |
---|
| 45 | . |
---|
| 46 | .PP |
---|
| 47 | The following MCA parameters apply to all components: |
---|
| 48 | . |
---|
| 49 | .TP 4 |
---|
| 50 | crcp_base_verbose |
---|
| 51 | Set the verbosity level for all components. Default is 0, or silent except on error. |
---|
| 52 | . |
---|
| 53 | .\" Coord Component |
---|
| 54 | .\" ****************** |
---|
| 55 | .SS coord CRCP Component |
---|
| 56 | .PP |
---|
| 57 | The \fIcoord\fR component implements a Coordinated Checkpoint/Restart |
---|
| 58 | Coordination Protocol similar to the one implemented in LAM/MPI. |
---|
| 59 | . |
---|
| 60 | .PP |
---|
| 61 | The \fIcoord\fR component has the following MCA parameters: |
---|
| 62 | . |
---|
| 63 | .TP 4 |
---|
| 64 | crcp_coord_priority |
---|
| 65 | The component's priority to use when selecting the most appropriate component |
---|
| 66 | for a run. |
---|
| 67 | . |
---|
| 68 | .TP 4 |
---|
| 69 | crcp_coord_verbose |
---|
| 70 | Set the verbosity level for this component. Default is 0, or silent except on |
---|
| 71 | error. |
---|
| 72 | . |
---|
| 73 | .\" Special 'none' option |
---|
| 74 | .\" ************************ |
---|
| 75 | .SS none CRCP Component |
---|
| 76 | .PP |
---|
| 77 | The \fInone\fP component simply selects no CRCP component. All of the CRCP |
---|
| 78 | function calls return immediately with ORTE_SUCCESS. |
---|
| 79 | . |
---|
| 80 | .PP |
---|
| 81 | This component is the last component to be selected by default. This means that if |
---|
| 82 | another component is available, and the \fInone\fP component was not explicity |
---|
| 83 | requested then Open MPI will attempt to activate all of the available components |
---|
| 84 | before falling back to this component. |
---|
| 85 | . |
---|
| 86 | .\" ************************** |
---|
| 87 | .\" See Also Section |
---|
| 88 | .\" ************************** |
---|
| 89 | . |
---|
| 90 | .SH SEE ALSO |
---|
| 91 | ompi-checkpoint(1), ompi-restart(1), opal-checkpoint(1), opal-restart(1), |
---|
| 92 | orte_snapc(7), orte_filem(7), opal_crs(7) |
---|
| 93 | . |
---|