Quick Start

Jump to bottom Edit New page

Kazuaki Ishizaki edited this page Feb 2, 2016 · 6 revisions

Assumption: x86_64 or ppc64le linux installed CUDA 7.0 or newer

Install Spark-gpu

$ wget https://s3.amazonaws.com/spark-gpu-public/spark-gpu-latest-bin-hadoop2.4.tgz
$ tar xf spark-gpu-latest-bin-hadoop2.4.tgz && cd spark-gpu

Run a sample program for GPU (Logistic regression)

$ LD_LIBRARY_PATH=/usr/local/cuda/lib64 MASTER='local[2]' ./spark-gpu/bin/run-example SparkGPULR 8 3200 32 5