Diagram of stampede2 directories and what connects to what, and how fast
Stampede2 is a collection of 6,400 computers connected to three file servers, each with unique characteristics.
You need to understand the file servers to know how to use them.
$HOME | $WORK | $SCRATCH | |
|---|---|---|---|
Purged? | No | No | Files can be purged if not accessed for 10 days. |
Backed Up? | Yes | No | No |
Capacity | 10GB | 1TB | Basically infinite. 8.5 PB |
Command to Access | cdh | cdw | cds |
Purpose | Store Executables | Store Files | Run Jobs |
Executables that aren't available on TACC through the "module" command should be stored in $HOME.
If you plan to be using a set of files frequently or would like to save the results of a job, they should be stored in $WORK.
If you're going to run a job, it's a good idea to keep your input files in a directory in $WORK and copy them to a directory in $SCRATCH where you plan to run your job.
This example command might help a bit:
cp $WORK/my_fastq_data/*fastq $SCRATCH/my_project/
Now let's go on to look at how jobs are run on stampede2.