Jennifer Brana
I recently completed my M.S. in Computer Science at Carnegie Mellon University, where I worked with Professor Nathan Beckmann as a member of the CORGi research group and the Parallel Data Lab. I was supported by an NSF Graduate Research Fellowship (NSF GRFP).
My research interests lie at the intersection of systems, performance modeling, and compilers. I am broadly interested in understanding, predicting, and improving system performance, particularly as increasing hardware heterogeneity makes performance more difficult to reason about and optimize.
My recent research has approached these questions from the architecture side. My master's thesis studied how execution rules and resource constraints determine the parallelism a system can expose, with the goal of understanding what future architectures and runtimes must provide to efficiently reach more parallelism. I am also exploring these questions from the software side through an event-driven SAT solver that reorganizes computation around dynamically available work and dependencies to expose new opportunities for scheduling and optimization.
I did my undergrad in computer science and computer engineering at the University of Portland. During my undergrad, I worked with Nathan Beckmann at Carnegie Mellon University as part of REUSE. In the past, I have also worked on near-data processing at AMD Research and cache coherence for heterogeneous systems at CMU.
