On the ec3 and ec6 clusters at MSG, the reconstruction software in Priism is configured by default to parallelize the computations across the nodes in the cluster. ec6 has a queueing system so if you configure ssh to allow you to log into ec6 without supplying a password, you can run Priism on a MSG Linux machine and submit reconstruction jobs to the ec6 queue (the data files will have to be on a disk, like /home2, /home5 or /home6, that is visible under the same name to both the MSG Linux machine and the ec6 head node). If you want to set up parallel execution on another system or change the settings on ec3 and ec6, read on.
With EMTAR and EMTIAR, the parameters for controlling parallel execution are accessible by pressing the "Parallel execution settings..." near the bottom of the main dialog. When the adjacent toggle button labeled "default" is on, the button to view the parallel execution parameters is disabled and the values for the parameters are the preconfigured defaults for the queue you selected in the options dialog (press the button labeled "Options" at the bottom of the main dialog to see that dialog). Setting up the preconfigured defaults for queue is beyond the scope of this tutorial; the queue topic in http://msg.ucsf.edu/IVE/IVE4_HTML/BatchRegion.html has more information on how to do that. To change the parallel parameters, turn off the button labeled 'default" and press the "Parallel execution settings..." button. http://msg.ucsf.eedu/IVE/IVE4_HTML/EMCAT/Parallel.html describes how to set the parameters.