I have recently completed my PhD in Computer Science at Stanford, advised by Alex Aiken. I am currently a Principal Research Scientist at NVIDIA. My interests include programming models, parallel computing architectures, and trying to keep up with my two daughters as they explore and develop their own interests.

I am interested in the design of high-performance computing systems, with a focus on addressing the challenges of communication and coordination between processing elements in massively-parallel architectures. I believe the best solutions will come from a co-design of the programming models, compilers, runtimes, and actual hardware architecture that provides a path to high performance while preserving portability (both functional and performance) across systems and improving the productivity of the programmers who are often the most critical resource.

Curriculum Vita
Research Statement
Teaching Statement

Legion, a data-centric parallel programming system for writing portable high performance programs targeted at distributed heterogeneous architectures.

Dependent Partitioning Regent: A high-productivity programming language for HPC with logical regions Structure slicing: Extending logical regions with fields Realm: An event-based low-level runtime for distributed memory architectures Singe: leveraging warp specialization for high performance on GPUs. Language support for dynamic, hierarchical data partitioning   (extended version with proofs here) Legion: expressing locality and independence with logical regions EcoG: A power-efficient GPU cluster architecture for scientific computing