I am interested in the design of large, complex, network-oriented applications and systems. Most recently, I have become concerned with the architecture of such systems in support of data-intensive large-scale computation associated with machine learning workloads.
My most recent project, as part of the RiseLab, is focused on exploiting "serverless computing", a way to harness lightweight low cost stateless virtual machine images typically found in cloud computing environments, to perform long running data-intensive computations. See https://rise.cs.berkeley.edu/projects/cirrus/.
My research style is engineering-oriented, prototype- and artifact-driven, and highly interdisciplinary, leading to extensive collaboration with other faculty. Our methodology depends critically on the three step process of (1) evaluating existing systems to understand their performance and capability limitations, (2) extensive simulation-based analyses to explore the design space of new solutions and architectures, and (3) implementation and measurement of the most attractive design to uncover its implementation complexities and to validate the simulation models used in the preceeding step. In a typical project, these steps are iterated two to three times.