====== CUDA ======

Compiling CUDA applications typically requires you to pass the level of Compute Capability to the compiler. For example, Genga is compiled by typing

<code>
$ make SM=XX
</code>

where XX is the Compute Capability of the cards the code will run on. For cards available at the ICS, use the following.

^Slurm Partition^Card^Compute Capability^
|tasna|GeForce GTX 590|20|
|ceres|GeForce GTX 680|30|
|vesta|Tesla K80X|37|

:!: **You should compile on nodes of the partition you are running on. To do so, launch an interactive job as described on the [[slurm|SLURM Scheduler]] page.** :!:

Note that the GTX 680 cards have gimped double precision performance compared to the GTX 590. Do not be surprised if your code runs slower than you would expect.