Competence centre LifeWatch/Citizen Science/Task 4.2
Caffe is a deep learning framework created at UC Berkeley.
For the installation of Caffe in Altamira Supercomputer at IFCA (Spain) we have followed both the official installation guide and a specific guide for installing Caffe on a Supercomputer cluster.
Caffe has been installed locally at /gpfs/res_scratch/lifewatch/iheredia/.usr/local/src/caffe. For more information on the local installation of software in a supercomputer please check [1]. The software/libraries already available at Altamira were PYTHON 2.7.10, CUDA 7.0.28, OPENMPI 1.8.3, BOOST 1.54.0, PROTOBUF 2.5.0, GCC 4.6.3 and HDF5 1.8.10. The remainingn libraries had been installed locally at /gpfs/res_scratch/lifewatch/iheredia/.usr/local/lib. Modules can be loaded all at once by loading the share/modulefiles/common.
Comments:
At this moment Altamira runs with Tesla M2090 GPUs with CUDA capability 2.0. Therefore Caffe has been compiled without CuDNN (the GPU-accelerated library of primitives for deep neural networks) which requires GPUs with CUDA capability of 3.0 or higher. Hopefully they will soon be replaced with Tesla Kepler GPUs.