Hi All,
I am also in favor of CUDA suffixes. CUDA containing toolchain is only
needed when you need CUDA aware MPI (It does not work with Intel as far
as I know). In other words, if you want to use GPUs in different boxes.
Sincerely,
Balazs
On 19/03/2018 15:43, Joachim Hein wrote:
Hi,
I am currently installing tensorflow via easybuild (I assume many of
us do these days) and am trying to understand EasyBuild’s ideas on
toolchains supporting cuda.
I looked at TensorFlow-1.5.0-goolfc-2017b-Python-3.6.3.eb, which
builds ontop of a toolchain containing GCC, Cuda (installed as a
compiler module), and OpenMPI, Blas, FFTW etc.
I now noticed that there is a new
TensorFlow-1.6.0-foss-2018a-Python-3.6.4-CUDA-9.1.85.eb, which is
accepted into the development branch (PR 6016). This builds ontop a
“vanilla” foss-2018a toolchain, using a Cuda and cuDNN modules
installed as a core module (system compiler).
I am wondering how do we want to organise us in future? Do we want to
continue with the goolfc idea or do we go for a “core” cuda and cuDNN?
I feel this needs standardising soonish. It is also something I feel
I need to document for my users, who want to build their own cuda
based software. What models should be loaded to build software.
Any comments, how we take this further?
Best wishes
Joachim
--
HPC consultant
HPC/VSC Support and System Administration
Computing Center
ULB/VUB
Avenue Adolphe Buyllaan 91 - CP 197
1050 Brussels
Belgium
