Manager of Cloud & Software As Infrastructure
Justin Riley joined Research Computing in 2015 as a senior team lead for special projects. Before joining Justin worked in the Software Tools for Academics and Researchers Group (STAR: http://star.mit.edu) at MIT for eight years. There he worked with faculty and researchers at MIT to help bring their research into the classroom. This included supporting students running simulations for homework on real HPC clusters via a web interface, building custom virtual machines for teaching/learning parallel programming with OpenMP and OpenMPI, and helping teach an IAP course (6.963) on parallel programming with GPUs and CUDA.
As a part of this work he created the open source StarCluster (http://star.mit.edu/cluster) project which makes it easy for anyone to bootstrap their own HPC cluster on the cloud custom-tailored to their needs. The StarCluster project started as an internal tool for streamlining STAR’s support of HPC in the classroom but quickly became a general cluster computing toolkit for the cloud. StarCluster continues to grow and support a diverse community of HPC users.
In his spare time he likes to play guitar, snowboard, program, play Team Fortress 2, and hang out with his wife and cats.
- Ivica, C., Riley, J.T. and Shubert, C., 2009, June. StarHPC—Teaching parallel programming within elastic compute cloud. In Information Technology Interfaces, 2009. ITI'09. Proceedings of the ITI 2009 31st International Conference on(pp. 353-356). IEEE.
- Ragan-Kelley, B., Walters, W.A., McDonald, D., Riley, J., Granger, B.E., Gonzalez, A., Knight, R., Perez, F. and Caporaso, J.G., 2013. Collaborative cloud-enabled tools allow rapid, reproducible biological insights. The ISME journal, 7(3), p.461.
- Riley, J., Noss, J., Dillingham, W., Cuff, J. and Llorente, I.M., 2017. A High-Availability Cloud for Research Computing. IEEE Computer, 50(6), pp.92-95.