I just noticed I’m using 11GB of storage space on QUT’s HPC server. Almost all of that will be results from remotely running R-INLA over the last few weeks while writing my second paper. The INLA objects are usually tens of megabytes but sometimes hundreds depending on the model I fit (with ~50,000 rows of data) and I think I’d be pushing my relationship with the HPC team if I didn’t clean this up every so often.
Thanks heaps to Håvard Rue and Dan Simpson for their help with getting me sorted out with remote R-INLA.
Edit: if you’re at QUT and want to run R-INLA remotely, here are the instructions that have been worked out by me, Dan, Håvard and QUT HPC’s Ashley Wright.