Alexander Gaenko

Updated core libraries of the ALPS project

Abstract 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 systems. The libraries provide a convenient set of well-documented and reusable components for developing condensed matter physics simulation code, and the applications strive to make commonly used and proven computational algorithms available to a non-expert community. In this paper we present an updated and refactored version of the core ALPS libraries geared at the computational physics software development community, rewritten with focus on documentation, ease of installation, and software maintainability. Program summary Program Title: ALPS Core libraries Program Files doi: http://dx.doi.org/10.17632/fckj5d7wtr.1 Programming language: C++ Licensing provisions: GNU GPLv3 Nature of problem: Need for modern, lightweight, tested and documented libraries covering the basic requirements of rapid development of efficient physics simulation codes, especially for modeling strongly correlated electron systems. Solution method: We present a C++ open source computational library that provides a convenient set of components for developing parallel physics simulation code. The library features a short development cycle and up-to-date user documentation. External routines/libraries: CMake , MPI , Boost , HDF5 .

Energy saving strategies for parallel applications with point-to-point communication phases

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.

Shared-memory parallelization of a local correlation multi-reference CI program

Abstract 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.

Adaptive Application Composition in Quantum Chemistry

Component interfaces, as advanced by the Common Component Architecture (CCA), enable easy access to complex software packages for high-performance scientific computing. A recent focus has been incorporating support for computational quality of service (CQoS), or the automatic composition, substitution, and dynamic reconfiguration of component applications. Several leading quantum chemistry packages have achieved interoperability by adopting CCA components. Running these computations on diverse computing platforms requires selection among many algorithmic and hardware configuration parameters; typical educated guesses or trial and error can result in unexpectedly low performance. Motivated by the need for faster runtimes and increased productivity for chemists, we present a flexible CQoS approach for quantum chemistry that uses a generic CQoS database component to create a training database with timing results and metadata for a range of calculations. The database then interacts with a chemistry CQoS component and other infrastructure to facilitate adaptive application composition for new calculations.

Runtime Procedure for Energy Savings in Applications with Point-to-Point Communications

Although high-performance computing has always been about efficient application execution, both energy and power consumption have become critical concerns owing to their effect on operating costs and failure rates of large-scale computing platforms. Modern microprocessors are equipped with the capabilities to reduce their power consumption using techniques such as dynamic voltage and frequency scaling (DVFS) and CPU clock modulation (called throttling). Without careful application, however, DVFS and throttling may cause a significant performance loss due to system overhead. This work presents design considerations for a runtime procedure that dynamically analyzes blocking point-to-point communications, groups them according to the proposed criteria, and applies frequency scaling by analyzing both communication and architectural parameters without penalizing the performance much. Experiments, performed on NAS parallel benchmarks verify the proposed design by exhibiting energy savings of as much as 11% with a performance loss as low as 2%.

Energy efficient communications in quantum chemistry applications

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.

Fragment Molecular Orbital Method Adaptations for Heterogeneous Computing Platforms

Abstract Modern electronic structure calculations are characterized by unprecedented complexity and accuracy. They de-mand the full power of high-performance computing and must be in tune with the given architecture for superior efficiency. Thus, it is desirable to enable their static and dynamic adaptations using some external software (middle-ware), which may monitor both system availability and application needs, rather than mix science with system-related calls inside the application.Building on the successful usage of the NICAN middleware with the computational chemistry package GAMESS, the work described in this paper links NICAN with the fragment molecular orbital (FMO) method to augment FMO with adaptive capabilities. Specifically, its fragment scheduling is performed, both statically and dynamically, based on current conditions within a heterogeneous computing environment. Significant execution time and throughput gains have been obtained with static adaptations, while the dynamic ones prevented FMO to abort calculations due to the insuffcient memory available at the runtime.