QMCPACK is an open source quantum Monte Carlo package for ab initio electronic structure calculations. It supports calculations of metallic and insulating solids, molecules, atoms, and some model Hamiltonians. Implemented real space quantum Monte Carlo algorithms include variational, diffusion, and reptation Monte Carlo. QMCPACK uses Slater-Jastrow type trial wavefunctions in conjunction with a sophisticated optimizer capable of optimizing tens of thousands of parameters. The orbital space auxiliary-field quantum Monte Carlo method is also implemented, enabling cross validation between different highly accurate methods. The code is specifically optimized for calculations with large numbers of electrons on the latest high performance computing architectures, including multicore central processing unit and graphical processing unit systems. We detail the program's capabilities, outline its structure, and give examples of its use in current research calculations. The package is available at http://qmcpack.org.

We present a first-principles scheme that allows the orbital magnetization of a magnetic crystal to be evaluated accurately and efficiently even in the presence of complex Fermi surfaces. Starting from an initial electronic-structure calculation with a coarse ab initio k-point mesh, maximally localized Wannier functions are constructed and used to interpolate the necessary k-space quantities on a fine mesh, in parallel to a previously-developed formalism for the anomalous Hall conductivity [X. Wang, J. Yates, I. Souza, and D. Vanderbilt, Phys. Rev. B 74, 195118 (2006)]. We formulate our new approach in a manifestly gauge-invariant manner, expressing the orbital magnetization in terms of traces over matrices in Wannier space. Since only a few (e.g., of the order of 20) Wannier functions are typically needed to describe the occupied and partially occupied bands, these Wannier matrices are small, which makes the interpolation itself very efficient. The method has been used to calculate the orbital magnetization of bcc Fe, hcp Co, and fcc Ni. Unlike an approximate calculation based on integrating orbital currents inside atomic spheres, our results nicely reproduce the experimentally measured ordering of the orbital magnetization in these three materials.

We extend the recently developed converse NMR approach [T. Thonhauser, D. Ceresoli, A. Mostofi, N. Marzari, R. Resta, and D. Vanderbilt, J. Chem. Phys. 131, 101101 (2009)] such that it can be used in conjunction with norm-conserving, non-local pseudopotentials. This extension permits the efficient ab-initio calculation of NMR chemical shifts for elements other than hydrogen within the convenience of a plane-wave pseudopotential approach. We have tested our approach on several finite and periodic systems, finding very good agreement with established methods and experimental results. 71.15.Mb, 76.60.Cq

We calculate the carbon nuclear magnetic resonance (NMR) shielding for CO2 and the hydrogen shieldings for both H2 and H2O inside the metal organic framework MOF-74-Mg. Our ab initio calculations are at the density functional theory level using the van der Waals including density functional vdW-DF. The shieldings are obtained while placing the small molecules throughout the structure, including the calculated adsorption site for various loading scenarios. We then explore relationships between loading, rotational and positional characteristics, and the NMR shieldings for each adsorbate. Our NMR calculations show a change in the shielding depending on adsorbate, position, and loading in a range that is experimentally observable. We further provide a simple model for the energy and the NMR shieldings throughout the cavity of the MOF. By providing this mapping of shielding to position and loading for these adsorbates, we argue that NMR probes could be used to provide additional information about the position at which these small molecules bind within the MOF, as well as the loading of the adsorbed molecule.

This paper presents Unified Communication X (UCX), a set of network APIs and their implementations for high throughput computing. UCX comes from the combined effort of national laboratories, industry, and academia to design and implement a high-performing and highly-scalable network stack for next generation applications and systems. UCX design provides the ability to tailor its APIs and network functionality to suit a wide variety of application domains and hardware. We envision these APIs to satisfy the networking needs of many programming models such as Message Passing Interface (MPI), OpenSHMEM, Partitioned Global Address Space (PGAS) languages, task-based paradigms and I/O bound applications. To evaluate the design we implement the APIs and protocols, and measure the performance of overhead-critical network primitives fundamental for implementing many parallel programming models and system libraries. Our results show that the latency, bandwidth, and message rate achieved by the portable UCX prototype is very close to that of the underlying driver. With UCX, we achieved a message exchange latency of 0.89 us, a bandwidth of 6138.5 MB/s, and a message rate of 14 million messages per second. As far as we know, this is the highest bandwidth and message rate achieved by any network stack (publicly known) on this hardware.

No abstract

No abstract

No abstract