Running apps on the Commandline

Navigate to the folder where your cloudgene.yaml file is located and start your workflow with the following command:

cloudgene run cloudgene.yaml <workflow parameters>

or if you installed an application you can start it by entering the name. For example:

cloudgene run hello-cloudgene

Running a pipeline with Hadoop steps

Cloudgene needs a Hadoop cluster to execute MapReduce steps. If Cloudgene is installed on the Hadoop Namenode you have to use the --conf flag and provide the $HADOOP_CONF folder. For example:

./cloudgene run cloudgene.yaml <workflow parameters> --conf /etc/hadoop/conf

Running Hadoop steps on a remote Hadoop cluster

Navigate to the folder where your cloudgene.yaml file is located and execute your workflow with the --host flag to set the ip address of your remote Hadoop cluster:

./cloudgene run cloudgene.yaml <workflow parameters> --host <remote-ip>

Cloudgene executes your MapReduce steps on the remote cluster. You can use the --user flag to set the username which should be used to execute your job (e.g. it uses the HDFS directory of this user for all files):

./cloudgene run cloudgene.yaml <workflow parameters> --host <remote-ip> --user <remote-user>