No abstract

We describe a mathematical model for heterojunctions in semiconductors which can be used, e.g., for modeling higher efficiency solar cells. The continuum model involves well-known drift-diffusion equations posed away from the interface. These are coupled with interface conditions with a nonhomogeneous jump for the potential, and Robin-like interface conditions for carrier transport. The interface conditions arise from approximating the interface region by a lower-dimensional manifold. The data for the interface conditions are calculated by a Density Functional Theory (DFT) model over a few atomic layers comprising the interface region. We propose a domain decomposition method (DDM) approach to decouple the continuum model on subdomains which is implemented in every step of the Gummel iteration. We show results for CIGS/CdS, Si/ZnS, and Si/GaAs heterojunctions.

No abstract

This article describes a standard API for a set of Batched Basic Linear Algebra Subprograms (Batched BLAS or BBLAS). The focus is on many independent BLAS operations on small matrices that are grouped together and processed by a single routine, called a Batched BLAS routine. The matrices are grouped together in uniformly sized groups, with just one group if all the matrices are of equal size. The aim is to provide more efficient, but portable, implementations of algorithms on high-performance many-core platforms. These include multicore and many-core CPU processors, GPUs and coprocessors, and other hardware accelerators with floating-point compute facility. As well as the standard types of single and double precision, we also include half and quadruple precision in the standard. In particular, half precision is used in many very large scale applications, such as those associated with machine learning.

No abstract