论文信息 - Improving the parallel scaling of ABINIT

Improving the parallel scaling of ABINIT

This work presents the results of a three week stay of T.H. at the High Performance Systems Department of the CINECA Institute in Bologna, Italy. The main goals of the short project to understand and parallelize the core functions of the application ABINIT have been achieved. A huge share of the time consuming part of the code has been parallelized efficiently and scales well on modern supercomputers. We demonstrate parallel scaling on the IBM SP5 in CINECA and a small Opteron PC cluster.

Torsten Hoefler | Wolfgang Rehm | Rebecca Janisch

[1] Allan,et al. Solution of Schrödinger's equation for large systems. , 1989, Physical review. B, Condensed matter.

[2] M. B. Vigil. High performance systems , 1995 .

[3] Car,et al. Unified approach for molecular dynamics and density-functional theory. , 1985, Physical review letters.

[4] Torsten Hoefler,et al. A Performance Analysis of ABINIT on a Cluster System , 2006 .

[5] Carlo Cavazzoni,et al. A parallel and modular deformable cell Car–Parrinello code , 1999 .

[6] George Bosilca,et al. Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation , 2004, PVM/MPI.

[7] T. Arias,et al. Iterative minimization techniques for ab initio total energy calculations: molecular dynamics and co , 1992 .

[8] Xavier Gonze,et al. A brief introduction to the ABINIT software package , 2005 .

[9] W. Kohn,et al. Self-Consistent Equations Including Exchange and Correlation Effects , 1965 .

[10] P. Hohenberg,et al. Inhomogeneous Electron Gas , 1964 .

[11] V. Eyert. A Comparative Study on Methods for Convergence Acceleration of Iterative Vector Sequences , 1996 .

[12] G. Amdhal,et al. Validity of the single processor approach to achieving large scale computing capabilities , 1967, AFIPS '67 (Spring).

[13] Matthieu Verstraete,et al. First-principles computation of material properties: the ABINIT software project , 2002 .