Strained Silicon on Silicon by Wafer Bonding and Layer Transfer from Relaxed SiGe Buffer

We report the creation of strained silicon on silicon (SSOS) substrate technology. The method uses a relaxed SiGe buffer as a template for inducing tensile strain in a Si layer, which is then bonded to another Si handle wafer. The original Si wafer and the relaxed SiGe buffer are subsequently removed, thereby transferring a strained-Si layer directly to Si substrate without intermediate SiGe or oxide layers. Complete removal of Ge from the structure was confirmed by cross-sectional transmission electron microscopy as well as secondary ion mass spectrometry. A plan-view transmission electron microscopy study of the strained-Si/Si interface reveals that the lattice-mismatch between the layers is accommodated by an orthogonal array of edge dislocations. This misfit dislocation array, which forms upon bonding, is geometrically necessary and has an average spacing of approximately 40nm, in excellent agreement with established dislocation theory. To our knowledge, this is the first study of a chemically homogeneous, yet lattice-mismatched, interface.

Fast Sorting on a Distributed-Memory Architecture

We consider the often-studied problem of sorting, for a parallel computer. Given an input array distributed evenly over p processors, the task is to compute the sorted output array, also distributed over the p processors. Many existing algorithms take the approach of approximately load-balancing the output, leaving each processor with Θ(n/p) elements. However, in many cases, approximate load-balancing leads to inefficiencies in both the sorting itself and in further uses of the data after sorting. We provide a deterministic parallel sorting algorithm that uses parallel selection to produce any output distribution exactly, particularly one that is perfectly load-balanced. Furthermore, when using a comparison sort, this algorithm is 1-optimal in both computation and communication. We provide an empirical study that illustrates the efficiency of exact data splitting, and shows an improvement over two sample sort algorithms.