Module 11: High Performance Computing

This was the final module of the year – and perhaps one of the most relevant modules (although this was only one-week long!) since the majority of us need to speed up our algorithms at some points in our research projects for some forms of screening.

As a good programming practice, one should establish structures for ready access. To do this, we started the module with relational database and SQL languages for data access. Our practical session was on using different methods for compound registration and subsequent data retrieval. On the next day we connected ourselves to our very own computing cluster, based at the Advanced Research Computing. We experimented with OpenMP (shared memory parallelism) and MPI (message passing on distributed memory) on Arcus-B with some numerical integrations and ODEs.  On Tuesday we did this parallelism in C, and Thursday in Python – the message being that, if we were to truly get the most out of computing clusters, we should optimise codes in C and Fortran instead (Python for HPC is simply a mismatch / pretended HPC for lazy coders who want to plug their human-readable Python code on clusters without much transformation … ). Friday was a little bit different with a trainer from the Amazon Web Service taking us through the basics of how we can use the different functionalities on Amazon Cloud and its derivative modules. In my humble opinion, the only usage as of now for the raw academic research is the data storage portion. It is admittedly a trend to run your computer on a cloud server but at present days with the limited availability of powerful processing units, model building activities would be difficulty on such a platform.

There you go, this marks the end of the 6-month training before formal research work is carried out. Shortly I will be starting two rotation projects: one between April and early July, and one between mid-July and October. The first one will be based in Computer Science, on heart mechanics; while the second project will be in Statistics, working on immunoinformatics. Hopefully I will be able to keep a regular record of the project progress in here!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s