Usage

The main module of the code has been designed as a command line interface (CLI) that supports two arguments:

• [-c,-config]: A user specified yaml file that is required in order to setup the tool according to one's preferences. You can learn more about its structure here. Having provided the path to the file, simply type python3 main.py -c path/to/config to start the resource manager.

• [-i,-info]: An optional argument used to obtain information about either the job queue (i.e. queue parameter) or the current state of the cluster (i.e. state parameter). Some basic examples of how to use this option can be seen below.

Usage within a cluster

In order to use our resource manager alongside a real cluster we must submit our program as a batch job. This is done by writing a script like the following.

A script as such consists, in essence, of two parts. In the first one, some general directives must be defined. These include, most notably, the number of required nodes and processors. Note that the number of requested nodes and cores must be in accordance with the values provided in the config file. The second part of the script is used for both setting the proper environmental variables and loading the modules necessary for running our program. For instance, in this case, we are using the 4.5.2 and 1.8.3 series for GNU and OpenMPI respectively.