HPC

  • ansible-pi-cluster: A set of Ansible playbooks I used to build a SLURM cluster using Raspberry Pi’s. May be useful to someone interested in learning about HPC clusters.

  • torque_qhistory: A tool for viewing and doing simple queries on PBS/Torque accounting files.

  • modules_logger: A system for logging the usage of environment-modules to help system administrators better manage cluster software. Uses MongoDB as the persistence layer.

  • hpc-helper-scripts: This is just an assortment of scripts, notes, and miscellaneous tools I’ve put together for helping with managing HPC clusters. Some of these are trivial, others are decent tools in their own right; they’re included here in case they are useful to others.

  • Warewulf: a system for provisioning and managing Linux systems, well-optimized for working with HPC clusters, and developed out of Berkeley National Lab. I’ve done a little bit of work for this on IPMI and user interface, and mostly done a lot of bugfixing as I run into problems. One of my primary tools as an HPC admin.

From grad school

  • imagej_morphology: A library based on the popular ImageJ image processing program for carrying out morphological operations, and some associated plugins for ImageJ itself.

  • “Genotyping by Alkaline Dehybridization Using Graphically Encoded Particles”: This project, led by Huaibin Zhang, demonstrated a cool way to do DNA discrimination using pH as a driving force rather than heat, and encapsulated it in a neat polymer-particle and microfluidic device. I was involved in the fabrication of the encoded microparticles, and in the image processing of the microscopy images and some of the subsequent analysis. The supplementary information on this paper is worth reading if you enjoy the analysis portion.

  • M.S. thesis: “Fabrication, dynamics and self-assembly of anisotropic colloidal particles.” Details my work on the fabrication of “patchy” microparticles with non-spherical geometry and non-uniform composition, and some work on measuring and analysing their interaction at low concentration. Really only of interest if you enjoy soft matter materials science, but analyzing the microscopy data was also a good “practical” application of morphological image processing.

  • grad school matlab: A bunch of the matlab scripts I used for processing data for my thesis. Like much academic code it’s uniformly horrible, but it might be worth reading and re-writing if you’re doing related work. Licensed under the CRAPL

Also, see a list of my academic publications.