WESTGRID (http://www.westgrid.ca/) provides high performance computing, networking, and collaboration tools hosted by Western universities (Simon Fraser University, The Banff Centre, TRIUMF, University of Alberta, University of British Columbia, University of Calgary, University of Lethbridge, University of Victoria, University of Northern British Columbia, Athabasca University, University of Saskatchewan, University of Regina, University of Manitoba, University of Winnipeg, Brandon University). Since WESTGRID is CFI funded, WESTGRID also welcomes external users from any other institutions in Canada (ie: any student enrolled with any Canadian university).

Here's quick list of general facilities available:

1) storage: storage facility

2) shared memory: nexus group, cortex group

3) cluster: glacier, robson

4) cluster with fast interconnect: lattice, matrix, tantalus

5) visualization: hydra

Note #1: Here is their hardware in detail: http://www.westgrid.ca/support/Facilities

Note #2: Here is their software in detail: http://www.westgrid.ca/support/software

Get an account here: https://rsg.nic.ualberta.ca/new_account.php

Note: To get an account, one must be supervised by a project leader. Any professor with any Canadian university may become a project supervisor and approve accounts under his project. Use the same website above to get a project leader account and a project ID number.

Here are some things found out about WESTGRID that might help someone who is new to WESTGRID (after reading the WESTGRID FAQ).

1) Opening an account is more of a formality than anything. Everyone gets accepted.

2) Matlab is available but only in single processor format on WESTGRID GLACIER.

3) I have made a Matlab script that will take an M-file (whose primary function is to run many identical test cases with different parameters after some setup code at the top of the M-file) and create a queue to submit to the WESTGRID queue. See a graphics explanation here. See the code here westgrid.m.

(Note: The picture was intended for CLUMEQ, but it serves to explain the WESTGRID as well).

4) Resource allocation is a complete mystery despite the assurance of these documents.

5) The admin for GLACIER, Roman Baranowski, if a very useful and responsive. When troubles arise, do contact him ( roman@chemDELETEthisTEXT.ubc.ca ).

6) CLUMEQ is the same thing as GLACIER WESTGRID. It is woefully under used most of the time. It might be worth it to race your queues on both systems.

7) WESTGRID GLACIER makes 2 files, stdout and errout, after running your job queue file (ie: PBS file). However, the error file sometimes cuts off if Matlab is writing to that error file (unknown why). This makes debugging Matlab sometimes difficult. Beware....

8) I have made a matlab script that will kill all the user queues. Check it out here qdelall.m. One uses it like this under matlab: qdelall('your_user_name');

9) Sometimes there aren't enough licenses free to execute your program. This is a pain. However, someone has solved this Stupid Cluster Tricks. Good luck with this.

10) To solve the point above for matlab licenses, one may include the following line of execution into your matlab code:

    % halt execution for a minute if we can't get the image toolbox
    while (~license('checkout','image_toolbox'))
        pause(60); 
    end;

Note: This can be done for every single toolbox required. However, to prevent possible license-required deadlocks, I highly suggest one manually (within your matlab script, and most probably during your initilisation procedure) control the checking out of required matlab toolbox licenses as the matlab license() function does not allow one to release a toolbox license. But, of course, this only works if one can initiate a maltab session.

11) To solve the above 2 problems, one can take the ultimate step and compile the M-file into a system executable. This is done via matlab's compiler. However, one must be aware of the licensing limitations associated with a stand-alone system executable from the matlab compiler. Of course, if one uses a call in the "Can not be compiled" column, then one must revert back to the solution above. Gee, this is getting complicated.

12) The ultimate matlab licensing solution would involved a system executable which attempts to launch a matlab engine. Whilst monitoring the standard error output stream (ie: stderr), if it detects an error "-4" (the error number associated with insufficient matlab licenses), then it can relaunch the matlab engine until it doesn't fail. The problem with this solution is that it is racing against all the other processes in a polling manner-- quite inelegant. So really this isn't the ultimate solution.

Note: Please feel free to edit this page. If something isn't clear, then feel free to email Albert Law.