-
Notifications
You must be signed in to change notification settings - Fork 8
Supported Adaptors & Resources
The service_url
is part of the pilot_compute_description
. The service URL communicates what type of queueing system or middleware you want to use and where it is. localhost
can be changed to a machine-specific URL, for example: sge://lonestar.tacc.utexas.edu
. The following table explains the supported middleware on XSEDE and FutureGrid. Note: You WILL have to edit the examples for your personal middleware or queueing system.
Supported Adaptors | Description | XSEDE Resources (Examples) |
---|---|---|
fork | Submit jobs only on localhost head node. Example usage: fork://localhost | Any |
SSH | Allows to submit jobs to a remote host via SSH. Password-less login to target machine is required. Example usage: ssh://eric1.loni.org | Any |
SGE(+SSH/GSISSH) | Interfaces with a Sun Grid Engine (SGE) scheduler locally or remotely via SSH. Password-less login to target machine is required. Example usage:
Local: sge://localhost Remote (over SSH): sge+ssh://lonestar.tacc.utexas.edu Remote (over GSISSH): sge+gsissh://lonestar.tacc.utexas.edu |
Lonestar Ranger |
Torque/PBS(+SSH/GSISSH) | Submit jobs using gsissh. Globus certificates are required. Initiate grid proxy (myproxy-logon) before executing the BigJob application. Example usage of URL: Local: pbs://localhost SSH:pbs+ssh://trestles.sdsc.edu GSISSH: xt5torque+gsissh://gsissh.kraken.nics.xsede.org |
Trestles Kraken |
SLURM+SSH | Submit jobs using SLURM and SSH: slurm+ssh://login1.stampede.tacc.utexas.edu | Stampede |
If jobs are submitted via GSISSH, a valid Globus certificates is required. Initiate grid proxy (myproxy-logon) before executing the BigJob application. Please find the GSISSH resource URLs of XSEDE machines at https://www.xsede.org/wwwteragrid/archive/web/user-support/gram-gatekeepers-gateway.html
The attributes required in the PilotComputeDescription depend on the respective backends.
Mandatory Attributes:
service_url
number_of_processes
Most production infrastructures additionally require:
-
processes_per_node
(required for PBS, Torque, SGE) queue
project
walltime
Stampede Example PilotComputeDescription
:
pilot_compute_description = {
"service_url": 'slurm+ssh://login1.stampede.tacc.utexas.edu',
"number_of_processes": 32,
"queue":"normal",
"project":"TG-MCBXXXXXX", # if None default allocation is used
"walltime":10,
"working_directory": os.getcwd()
}