Running exareme in distributed mode

There are two ways to run exareme in distributed mode (more than 1 workers): a) Either you setup a vm for each worker and one for the master b) or you can use a test main, thawhich allows exareme to run in a pseudo-distributed mode (more than one workers in a single machine). The pseudo-distributed mode is only meant to be used for experimentation and quick testing when developing.

If you opt for the first setup you need to follow these instructions: https://github.com/madgik/exareme/wiki/Getting-Started the files master and workers contain the ips of the vms, that way you differentiate the master instances from the workers.

If you opt for the second setup: you will need to download the dev branch and build it as per the instructions at https://github.com/madgik/exareme/wiki/Getting-Started#development and use the main in madgik.exareme.master.gateway.ExaremeGatewayStart. The ExaremeClusterFactory.createMiniCluster(...) creates the cluster. The last argument of this method is the number of workers that will be created. then you can interface with exareme via the console. You can find the console at this path " exareme/exareme-tools/madis/src/mterm.py" (instruction on how to use the console can be found at the "Getting started" page of this wiki)
The master node of exareme also runs a worker instance.