CONVERGE Horizon
Environment Variables#
You can use the environment variables listed below when running a job with custom commands.
Variable |
Value |
Example |
|---|---|---|
|
The job’s ID. |
|
|
The ID of the team that owns the job. |
|
|
The Case Directory, which is the job’s storage path on the head node. Your commands start here. |
|
|
The folder that storage paths are relative to. |
|
|
Where the job’s storage is mounted. |
|
|
Where applications are installed. |
|
|
Where the job’s application version is installed. |
|
|
The application version the job runs. |
|
|
The MPI implementation the application version uses ( |
|
|
A file in the Case Directory that lists the private IP address of every node (one per line, head node first). |
|
|
The number of nodes. |
|
|
The number of cores per node the job is set to use. See Advanced Options > Cores per Node. |
|
|
The number of cores per node times the number of nodes. |
|
|
For CONVERGE 6 and later versions, the RDMA device that connects the nodes. Empty when the nodes have none. |
|
|
For CONVERGE 6 and later versions, the network interface of the RDMA device. Empty when the nodes have none. |
|
|
Where GT-SUITE is installed for jobs running with GT-SUITE coupling. The default snippet runs CONVERGE on 2 fewer cores than |
|