[ https://issues.apache.org/jira/browse/MAHOUT-2025?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16965300#comment-16965300 ]
Andrew Palumbo commented on MAHOUT-2025: ---------------------------------------- As well we may want a {{cuda}} based \{{Dockerfile}}. > Add a script to launch an EC2 cluster, install Mahout and JCuda or ViennaCL > to Examples > --------------------------------------------------------------------------------------- > > Key: MAHOUT-2025 > URL: https://issues.apache.org/jira/browse/MAHOUT-2025 > Project: Mahout > Issue Type: Task > Affects Versions: 0.13.0 > Reporter: Andrew Palumbo > Assignee: Andrew Palumbo > Priority: Major > Fix For: 0.13.2, 14.1, 14.2, 15.0 > > > Create a script for {{$MAHOUT_HOME/examples/bin}} to launch a spark cluster > on EC2 using e.g.: [spark-ec2|https://github.com/amplab/spark-ec2], to: > # launch an EC2 Cluster > # Install Mahout > # Install either JCuda or ViennaCL > # RSYNC CUDA drivers and libraries in the case of JCuda > Much of this is already done in the {{.travis.yml}} configuration. > This should lessen the learning curve for those wishing to make use of the > native solvers. -- This message was sent by Atlassian Jira (v8.3.4#803005)