... | ... | @@ -4,16 +4,16 @@ As explained in this [document](http://www.cnms.ornl.gov/capabilities/CNMS_Compu |
|
|
## OIC Phase 5
|
|
|
28 Compute Nodes: Each with two Interlagos Opterons 6274 w/ 16 cores 2.2GHz 16MB L3 Cache, 128GB RAM (DDR3 Registered, Dual Rank, 1333MHz, 1.5v), 1TB HDD
|
|
|
|
|
|
###Getting there
|
|
|
### Getting there
|
|
|
```shell-session
|
|
|
ssh oicphase5.ornl.gov
|
|
|
```
|
|
|
###Accessing CNMS modules
|
|
|
### Accessing CNMS modules
|
|
|
add to your .bashrc
|
|
|
```shell
|
|
|
export MODULEPATH=$MODULEPATH:/projects/cnms/modules
|
|
|
```
|
|
|
###Seeing which modules are available
|
|
|
### Seeing which modules are available
|
|
|
```shell-session
|
|
|
[you@b24l01]$ module avail
|
|
|
...
|
... | ... | @@ -22,17 +22,21 @@ espresso/5.3.0 intelenv lammps/16Feb16 |
|
|
```
|
|
|
currently the modules for vasp, lammps have been tested
|
|
|
|
|
|
###load modules
|
|
|
### load modules
|
|
|
```shell-session
|
|
|
module load module_name
|
|
|
```
|
|
|
|
|
|
to see what they provide
|
|
|
### to see what they provide
|
|
|
```shell-session
|
|
|
module show module_name
|
|
|
```
|
|
|
## Performance on OICPHASE 5
|
|
|
The AMD Interlagos chip has only one FPU per two integer unit cores. As a result for almost all scientific software it can only hurt you to run every core on a node. However only running one core of each pair is a bit of a challenge in the OIC environment. Please see the method in the PBS examples. This results in a 50-100% speed up in VASP and should have a similar effect whenever the code is primarily doing floating point operations.
|
|
|
|
|
|
|
|
|
lammps has undergone only minimal testing via the examples.
|
|
|
|
|
|
And example PBS script is provided in this repo, please clone the repo if you have examples to share. But please work on a branch not master..
|
|
|
|
|
|
[vasp pbs example](https://code.ornl.gov/CNMS/CNMS_Computing_Resources/blob/master/OIC/VASP/vasp_job.pbs) |