Dual-level parallelization, Taskgroups parallelization and Distributed linear algebra

Atomic wavefunction initialization in ab initio molecular dynamics using distributed Lanczos

Parallel Computing, 34(6-8):441-450.
Bekas, C, Curioni, A, and Andreoni, W (2008)

Abstract

We present a distributed scheme for initialization from atomic wavefunctions in ab initio molecular dynamics simulations. Good initial guesses for approximate wave functions are very important in order to enable practical simulations with thousands of atoms. The new scheme is based on a distributed implementation of the Lanczos algorithm for very large dense eigenproblems. We show that the massively parallel BG/L (Blue Gene/L) supercomputer with its very fast separate network for collective communications is an ideal platform for the parallel Lanczos algorithm. We have implemented the new scheme in the popular plane-wave code CPMD. We showcase the applicability of the distributed initialization by a series of examples on a family of Silicon super cells ranging from 512 to 2048 atoms. Keywords: Ab initio molecular dynamics; Lanczos; Large scale parallelization; BG/L supercomputer.

URL: http://dx.doi.org/10.1016/j.parco.2008.01.005

Dual-level parallelism for ab initio molecular dynamics: Reaching teraflop performance with the CPMD code

Parallel Computing, 31(1):1–17.
Hutter, J and Curioni, A (2005).

Abstract

We show teraflop performance of the fully featured ab initio molecular dynamics code CPMD on an IBM pSeries 690 cluster. A mixed distributed-memory, coarse-grained parallel approach using the MPI library and shared-memory, fine-grained parallelism using OpenMP directives is used to optimally map the algorithms on the available hardware. The top performance achieved is ≈20\% of the peak performance and an estimated parallel efficiency of ≈45\% on 1024 processors for a system of 1000 atoms. The main limiting factor of parallel efficiency was found to be the latency of the interconnect. Keywords: Car–Parrinello molecular dynamics; Parallelization; OpenMP; MPI PACS: 71.15Pd

URL: http://dx.doi.org/10.1016/j.parco.2004.12.004


Car-Parrinello Molecular Dynamics on Massively Parallel Computers

ChemPhysChem, 6(9):1788–1793.
Juerg Hutter and Alessandro Curioni (2005)

Abstract

The Car–Parrinello molecular dynamics (CPMD) method is shown to be highly suited to modern computer architectures, such as clustered shared‐memory parallel servers and ultra‐dense massively parallel computers. Different parallelization strategies are reviewed to elucidate the most efficient application of the CPMD code.

URL: http://dx.doi.org/10.1002/cphc.200500059