|
|
# TITAN -- It's a tricky beast, but it was [1st](https://www.top500.org/lists/2012/11/) (on the [top 500](https://www.top500.org/lists/2017/11/))!
|
|
|
And it is still 5th.
|
|
|
## AMD Interlagos based machine, like OIC phase 5.
|
|
|
## Differences from the standard Intel Linux Cluster
|
|
|
### AMD Interlagos based machine, like OIC phase 5.
|
|
|
The key is there are only one floating point unit per two cores. So from a science perspective there are only 8 cores per node, not the sixteen listed. Additionally your code needs to bind to just one core per pair. See [Cray XK7 CPU info](https://www.olcf.ornl.gov/support/system-user-guides/titan-user-guide/#333)
|
|
|
## Service nodes run mpimom processes and PBS scripts -- (this complicates what PBS scripts can do)
|
|
|
## There is only one GPU per node, i.e. 16 CPU, 8 FPU, 1 GPU. It's generally on you to manage this.
|
|
|
### Service nodes run mpimom processes and PBS scripts
|
|
|
this complicates what PBS scripts can do. You can only run 50 aprun's that is 50 independent mpi applications in one job script.
|
|
|
### There is only one GPU per node, so 16 CPU, 8 FPU, 1 GPU.
|
|
|
It's generally on you to manage this.
|
|
|
|
|
|
# Qsub chaining jobs
|
|
|
qsub -W depend=afterok:$JOBID fourth_1.sh
|
|
|
```shell-session
|
|
|
qsub -W depend=afterok:JOBID fourth_1.sh
|
|
|
```
|
|
|
* _afterok_ if you need previous job to be successful
|
|
|
* _afterany_ if you don't care
|
|
|
|
|
|
|
|
|
If you are having trouble, and there is not an example in the repo, it's probably best to just contact slack/here @epd [email](mailto:doakpw@ornl.gov) |
|
|
\ No newline at end of file |
|
|
If you are having trouble, and there is not an example in the repo, it's probably best to just contact slack/here @epd or [email](mailto:doakpw@ornl.gov) |
|
|
\ No newline at end of file |