This report comprises the complete D5.2.1 deliverable as specified for workpackage WP5.2 in Subproject

This paper describes the design of PICO, a C++ framework for implementing general parallel branch-and-bound algorithms. The PICO framework provides a mechanism for the efficient implementation of a wide range of branchand -bound methods on an equally wide range of parallel computing platforms. We first discuss the basic architecture of PICO, including the application class hierarchy and the package's serial and parallel layers. We next describe the design of the serial layer, and its central notion of manipulating subproblem states. Then, we discuss the design of the parallel layer, which includes flexible processor clustering and communication rates, various load balancing mechanisms, and a non-preemptive task scheduler running on each processor. We close by describing the application of the package to a simple branch-and-bound method for mixed integer programming, along with computational results on the ASCI Red massively parallel computer.

A 3d geoscience information system framework THÈSE pour l´obtention du Doctorat de l´Institut National Polytechnique de Lorraine (Spécialité Géosciences) par

In this thesis many-body Quantum Monte Carlo (QMC) calculations are presented for the ground state and excitation energies of Ge atoms, molecules, and clusters as large as Ge29H36 ( ≈ 1.2 nm diameter). The hydrogen terminated GenHm nanoclusters are of particular interest since they have optical properties that are different from small molecules and bulk Ge, and are viewed as an ideal model for quantum confined semiconductor systems. The present QMC results are compared with previous QMC calculations for the corresponding Si molecules and clusters. In addition, the QMC gaps for GenHm are found to be higher than the gaps reported in recent time-dependent density functional studies, by amounts similar to that previously found for Si systems. For materials containing Ge it is necessary to deal with the issue of correlation between the core and valence electrons. The accuracy of QMC for heavy atoms such as Ge (Z=32) is limited by the fact that the core-valence interactions cannot be treated at the same manybody level as the valence-valence interactions. Typically the core-valence interactions are treated at a single-body level via a pseudopotential, but such methods are unsatisfactory for Ge which has a shallow, easily polarizable 3d core. Previous work has proposed using