Research Experience

@ The UTEP Department of Computer Science

Home

Publications

Bio

GAMESS ECP project

GAMESS stand for General Atomic and Molecular Electronic Structure System. The main focus will be on combined CPU+GPU performance and scalability analysis for the ECP exascale machines. A preliminary assessment of the performance analysis requirements will be fleshed out in more detail through surveying GAMESS ECP developers for their requirements. The framework will be constructed from selected features from the available vendor and ECP Software Technology performance analysis tools, with the addition of project-specific tools, scripts, and benchmark data.

E3SM ECP project

MPAS stands for Model for Prediction Across Scales. It is a collaborative project for developing atmosphere, ocean, and other earth-system simulation components for climate, regional climate, and weather studies. It uses a hexagonal mesh resembling a honeycomb that can be stretched wide in some regions and compressed for higher resolution in others. The MPAS framework code is written in Fortran and usesthe Message Passing Interface (MPI) standardized library (not a language) for the collection of processes communicating via message passing. Shared memory parallelization through OpenMP (an API) is also supported, but the implementation is left up to each core component.

Doctoral Research Topic

The purpose of this research is to design a faster implementation of the spatially variant that improves its performance when it is running on a parallel computer system. The spatially variant is used to synthesize a spatially variant lattice for a periodic electromagnetic structure. The spatially variant has the ability to spatially vary the unit cell orientation and exploit its directional dependencies. The spatially variant produces a lattice that is smooth, continuous and free of defects. The lattice spacing remains strikingly uniform when the unit cell orientation, lattice spacing, fill fraction and more are spatially varied. This is important for maintaining consistent properties throughout the lattice.

Periodic structures like a photonic crystal or metamaterial devices can be enhanced using the spatially variant to unlock new physics applications. Our current effort is to write a portable spatially variant code for parallel architectures. To develop and write the code, we pick a general-purpose programming language that supports structured programming. For the parallel code, we use FFTW for handling the Fourier Transform of the unit cell device and PETSc (Portable, Extensible Toolkit for Scientific Computation) for handling the numerical linear algebra operations. Using Message Passing Interface (MPI) for distributed memory helps us to improve the performance of the spatially variant code when it is executed on a parallel system.