Where are you at this time and what are you most likely to be doing?

Running simulations from my computer — sometimes I do this from my desk in the WID, but often I am at home, sending code remotely to CHTC (the Center for High Throughput Computing). My research involves simulating the progression of chemical reactions to examine how abiotic reaction networks might have displayed evolutionary and ecological characteristics before the origin of life. I need to run many replicates of my simulations to collect data and it would be impossible to do this if I had to run them all sequentially on my own computer, but with CHTC I can run many simulations in parallel, though they still often take many days to run. I can submit my simulations remotely and access the data from my own computer whenever they are finished.

How can you tell if your work is going well?

It takes a long time for me to see my results after sending my simulations to the CHTC. First I need to stitch together the output of my (usually hundreds) of  simulations and create a graph to see whether they have produced some sensible output. By this time, it has usually been multiple days since I first sent my simulations off and it is a great relief to see results that, even if surprising, at least seem to follow the basic chemistry my work is meant to simulate and indicate that nothing has gone wrong!

What tells you it’s not going as expected? What do you do when that happens?

I can tell something has gone wrong when my simulations do not take a very long time to run! If I start getting data back immediately it usually means I have done something wrong like misnaming a file or the simulations have run out of memory or some other such error. Whenever I submit my code to CHTC I spend some time checking their queue to see if they quickly get returned to me. If they don’t finish within a few minutes, it usually means they are working.