No abstract

Specialized computational chemistry packages have permanently reshaped the landscape of chemical and materials science by providing tools to support and guide experimental efforts and for the prediction of atomistic and electronic properties. In this regard, electronic structure packages have played a special role by using first-principle-driven methodologies to model complex chemical and materials processes. Over the past few decades, the rapid development of computing technologies and the tremendous increase in computational power have offered a unique chance to study complex transformations using sophisticated and predictive many-body techniques that describe correlated behavior of electrons in molecular and condensed phase systems at different levels of theory. In enabling these simulations, novel parallel algorithms have been able to take advantage of computational resources to address the polynomial scaling of electronic structure methods. In this paper, we briefly review the NWChem computational chemistry suite, including its history, design principles, parallel tools, current capabilities, outreach, and outlook.

A parallel coupled cluster algorithm that combines distributed and shared memory techniques for the CCSD(T) method (singles + doubles with perturbative triples) is described. The implementation of the massively parallel CCSD(T) algorithm uses a hybrid molecular and "direct" atomic integral driven approach. Shared memory is used to minimize redundant replicated storage per compute process. The algorithm is targeted at modern cluster based architectures that are comprised of multiprocessor nodes connected by a dedicated communication network. Parallelism is achieved on two levels: parallelism within a compute node via shared memory parallel techniques and parallelism between nodes using distributed memory techniques. The new parallel implementation is designed to allow for the routine evaluation of mid-(500−750 basis function) to large-scale (750−1000 basis function) CCSD(T) energies. Sample calculations are performed on five low-lying isomers of water hexamer using the aug-cc-pVTZ basis set. Disciplines Chemistry | Computer Sciences CommentsThe following article appeared Abstract: A parallel coupled cluster algorithm that combines distributed and shared memory techniques for the CCSD(T) method (singles + doubles with perturbative triples) is described. The implementation of the massively parallel CCSD(T) algorithm uses a hybrid molecular and "direct" atomic integral driven approach. Shared memory is used to minimize redundant replicated storage per compute process. The algorithm is targeted at modern cluster based architectures that are comprised of multiprocessor nodes connected by a dedicated communication network. Parallelism is achieved on two levels: parallelism within a compute node via shared memory parallel techniques and parallelism between nodes using distributed memory techniques. The new parallel implementation is designed to allow for the routine evaluation of mid-(500-750 basis function) to large-scale (750-1000 basis function) CCSD(T) energies. Sample calculations are performed on five low-lying isomers of water hexamer using the aug-cc-pVTZ basis set.

Both the replicated and distributed data parallel full configuration interaction (FCI) implementations are described. The implementation of the FCI algorithm is organized in a hybrid strings-integral driven approach. Redundant communication is avoided, and the network performance is further optimized by an improved distributed data interface library. Examples show linear scalability of the distributed data code on both PC and workstation clusters. The new parallel implementation greatly extends the hardware on which parallel FCI calculations can be performed. The timing data on the workstation cluster show great potential for using the new parallel FCI algorithm in expanding applications of complete active space self-consistent field applications. Keywords Configuration interaction Disciplines Chemistry CommentsThe following article appeared in Journal of Chemical Physics 119 (2003) Both the replicated and distributed data parallel full configuration interaction ͑FCI͒ implementations are described. The implementation of the FCI algorithm is organized in a hybrid strings-integral driven approach. Redundant communication is avoided, and the network performance is further optimized by an improved distributed data interface library. Examples show linear scalability of the distributed data code on both PC and workstation clusters. The new parallel implementation greatly extends the hardware on which parallel FCI calculations can be performed. The timing data on the workstation cluster show great potential for using the new parallel FCI algorithm in expanding applications of complete active space self-consistent field applications.