We present BHM, a tool for restoring a smooth function from a sampled histogram using the bin hierarchy method. It is particularly useful for the analysis of data from large-scale numerical simulations of physical systems, such as diagrammatic Monte Carlo simulations of quantum many-body problems. The theoretical background of the method is presented in Goulko et al., (2018). The code automatically generates a smooth polynomial spline with the minimal acceptable number of knots from the input data. It works universally for any sufficiently regular shaped distribution and any level of data quality (provided that the data are uncorrelated or correlations have been accounted for), requiring almost no external parameter specification. This paper explains the details of the implementation and the use of the program, including a physical example of the restoration of the Frohlich polaron Green's function from data sampled with diagrammatic Monte Carlo. Program summary Program Title: BHM Program Files doi: http://dx.doi.org/10.17632/dvj8gxsxpk.1 Licensing provisions: GPLv3 Programming language: C++ External routines/libraries: CMake, GSL Nature of problem: Restoring a smooth function from a sampled histogram. Solution method: To make use of all information contained in the sampled data, the BHM algorithm generates a hierarchy of overlapping bins of different sizes from the initially supplied fine histogram. The bin hierarchy is fitted to a polynomial spline with the minimal acceptable number of knots, the positions of which are determined automatically. The output is a smooth function with error band. (C) 2018 Elsevier B.V. All rights reserved.

The open source ALPS (Algorithms and Libraries for Physics Simulations) project provides a collection of physics libraries and applications, with a focus on simulations of lattice models and strongly correlated electron systems. The libraries provide a convenient set of well-documented and reusable components for developing condensed matter physics simulation codes, and the applications strive to make commonly used and proven computational algorithms available to a non-expert community. In this paper we present an update of the core ALPS libraries. We present in particular new Monte Carlo libraries and new Green's function libraries.

This chapter presents an overview of the components of the dynamical nucleation theory effective fragment potential (DNTEFP) model, the use of the DNTEFP for the representative molecular systems of atmospheric importance, and implementation and performance of these methods using two-level parallelism on high performance computers. In DNTEFP, several important concepts are combined to facilitate the computation of the evaporation rates—dynamical nucleation theory Monte Carlo (DNTMC), the effective fragment potential, and two-level parallelism. The chapter presents a brief introduction to each of the concepts to acclimate the reader to the ideas. The chapter also presents the results of the use of the DNTEFP method to study several chemical systems. Next, it discusses an assessment of the performance of the software. The investigated chemical systems are representative of different types of calculations that can be performed using the DNTEFP method—homogeneous water clusters, heterogeneous water clusters with ions, and homogeneous dehydrated systems.

Submitted for the MAR17 Meeting of The American Physical Society Hypothesis testing of quantum Monte Carlo simulations1 MARKUS WALLERBERGER, ALEXANDER GAENKO, EMANUEL GULL, University of Michigan, Ann Arbor — The large implementation complexity of modern quantum Monte Carlo solvers makes careful testing of the algorithm as well as verification of the results an imperative. Due to their deterministic nature, traditional unit tests are unsuited for verifying probabilistic results: they are prone to false positives in the case of outliers or changes to the implementation. Therefore, Monte Carlo data are often checked by visual inspection only, which is susceptible to incomplete and non-continuous test coverage. Statistical hypothesis testing provides a non-deterministic alternative: we choose an exact result (which exists for certain limits) as the null hypothesis and compute the statistical significance score for the Monte Carlo data. Rejection or too strong acceptance of the null hypothesis then amounts to a failed test, thus providing a test criterion for both the Monte Carlo estimate and its error bars. While this does not provide a binary answer, ambiguous cases can be systematically refined by lengthening the Monte Carlo run, and the procedure lends itself to automation. We develop a testing framework and illustrate the procedure for the two-dimensional Ising model as well as for Continuous-time quantum Monte Carlo data for the single impurity Anderson model. 1The authors are funded by the Simons Foundation as part of the Simons Collaboration on the many-electron problem. Markus Wallerberger University of Michigan, Ann Arbor Date submitted: 16 Nov 2016 Electronic form version 1.4

Modern supercomputing platform designers are becoming increasingly aware of the operational costs and reliability issues, which are rising due to high power consumption of such systems. At the same time, high-performance application developers are taking pro-active steps towards less energy consumption without a significant performance loss. One way to accomplish energy savings during application execution is to change the processor frequency dynamically when processor is not busy, such as during certain communication stages. Previously, the authors have proposed a runtime procedure that identifies communication phases in parallel applications to apply frequency scaling efficiently and without much overhead. The present work applies the phase detection procedure to parallel electronic structure calculations, performed by a widely used package GAMESS. High computational intensity of these calculations and the GAMESS communication model, which distinguishes computation and communication processes, motivated the investigations in this paper. They have led to several insights as to the role of process-core mapping in the application of dynamic frequency scaling during communications.

We present a shared-memory parallelization of our open-source, local correlation multi-reference framework, TigerCI. Benchmarks of the total parallel speedup show a reasonable scaling for typical modern computing system setups. The efficient use of available computing resources will extend simulations on this high level of theory into a new size regime. We demonstrate our framework using local-correlation multireference computations of alkyl-substituted dioxirane and solvated methyl nitrene as examples.

This work presents a nonadiabatic molecular dynamics study of the nonradiative decay of photoexcited trans-azomethane, using the ab initio multiple spawning (AIMS) program that has been interfaced with the General Atomic and Molecular Electronic Structure System (GAMESS) quantum chemistry package for on-the-fly electronic structure evaluation. The interface strategy is discussed, and the capabilities of the combined programs are demonstrated with a nonadiabatic molecular dynamics study of the nonradiative decay of photoexcited trans-azomethane. Energies, gradients, and nonadiabatic coupling matrix elements were obtained with the state-averaged complete active space self-consistent field method, as implemented in GAMESS. The influence of initial vibrational excitation on the outcome of the photoinduced isomerization is explored. Increased vibrational excitation in the CNNC torsional mode shortens the excited state lifetime. Depending on the degree of vibrational excitation, the excited state lifetime varies from similar to 60200 fs. These short lifetimes are in agreement with time-resolved photoionization mass spectroscopy experiments.

The design and development of scientific software components to provide an interface to the effective fragment potential (EFP) methods are reported. Multiscale modeling of physical and chemical phenomena demands the merging of software packages developed by research groups in significantly different fields. Componentization offers an efficient way to realize new high performance scientific methods by combining the best models available in different software packages without a need for package readaptation after the initial componentization is complete. The EFP method is an efficient electronic structure theory based model potential that is suitable for predictive modeling of intermolecular interactions in large molecular systems, such as liquids, proteins, atmospheric aerosols, and nanoparticles, with an accuracy that is comparable to that of correlated ab initio methods. The developed components make the EFP functionality accessible for any scientific component-aware software package. The performance of the component is demonstrated on a protein interaction model, and its accuracy is compared with results obtained with coupled cluster methods.

Although high-performance computing traditionally focuses on the efficient execution of large-scale applications, both energy and power have become critical concerns when approaching exascale. Drastic increases in the power consumption of supercomputers affect significantly their operating costs and failure rates. In modern microprocessor architectures, equipped with dynamic voltage and frequency scaling (DVFS) and CPU clock modulation (throttling), the power consumption may be controlled in software. Additionally, network interconnect, such as Infiniband, may be exploited to maximize energy savings while the application performance loss and frequency switching overheads must be carefully balanced. This paper advocates for a runtime assessment of such overheads by means of characterizing point-to-point communications into phases followed by analyzing the time gaps between the communication calls. Certain communication and architectural parameters are taken into consideration in the three proposed frequency scaling strategies, which differ with respect to their treatment of the time gaps. The experimental results are presented for NAS parallel benchmark problems as well as for the realistic parallel electronic structure calculations performed by the widely used quantum chemistry package GAMESS. For the latter, three different process-to-core mappings were studied as to their energy savings under the proposed frequency scaling strategies and under the existing state-of-the-art techniques. Close to the maximum energy savings were obtained with a low performance loss of 2% on the given platform. (C) 2013 Elsevier Inc. All rights reserved.