Search NASA⌕ Search

SEARCH · Search NASA

Results for “QDP++”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

QDP-JIT/PTX: A QDP++ Implementation for CUDA-Enabled GPUs

These proceedings describe briefly the QDP-JIT/PTX framework for lattice field theory calcula- tions on the CUDA architecture. The framework generates compute kernels in the PTX assembler language which can be compiled to efficient GPU machine code by the NVIDIA JIT compiler. A comprehensive memory management was added to the framework so that applications, e.g. Chroma, can run unaltered on GPU clusters and supercomputers.

Winter, Frank T.↗

Automatic Offloading C++ Expression Templates to CUDA Enabled GPUs

In the last few years, many scientific applications have been developed for powerful graphics processing units (GPUs) and have achieved remarkable speedups. This success can be partially attributed to high performance host callable GPU library routines that are offloaded to GPUs at runtime. These library routines are based on C/C++-like programming toolkits such as CUDA from NVIDIA and have the same calling signatures as their CPU counterparts. Recently, with the sufficient support of C++ templates from CUDA, the emergence of template libraries have enabled further advancement in code reusability and rapid software development for GPUs. However, Expression Templates (ET), which have been very popular for implementing data parallel scientific software for host CPUs because of their intuitive and mathematics-like syntax, have been underutilized by GPU development libraries. The lack of ET usage is caused by the difficulty of offloading expression templates from hosts to GPUs due to the inability to pass instantiated expressions to GPU kernels as well as the absence of the exact form of the expressions for the templates at the time of coding. This paper presents a general approach that enables automatic offloading of C++ expression templates to CUDA enabled GPUs by using the C++ metaprogramming technique and Just-In-Time (JIT) compilation methodology to generate and compile CUDA kernels for corresponding expression templates followed by executing the kernels with appropriate arguments. This approach allows developers to port applications to run on GPUs with virtually no code modifications. More specifically, this paper uses a large ET based data parallel physics library called QDP++ as an example to illustrate many aspects of the approach to offload expression templates automatically and to demonstrate very good speedups for typical QDP++ applications running on GPUs against running on CPUs using this method of offloading. In addition, this approach of automatic offlo- ding expression templates could be applied to other many-core accelerators that provide C++ programming toolkits with the support of C++ template.

97 MATHEMATICS AND COMPUTING↗

Computing the Properties of Hadrons, Nuclei, and Nuclear Matter from Quantum Chromodynamics, UNC-CH Final Report.

This is the final technical report for the UNC-CH component of a SciDAC3 project. This project served to coordinate the software development effort which the nuclear physics lattice QCD community is developing to ensure that lattice calculations can make optimal use of forthcoming leadership-class and dedicated hardware, including those of the national laboratories, and prepares for the exploitation of future computational resources in the exascale era. The project developed and optimized simulation software for lattice QCD calculations on leadership class computers and on GPUaccelerated heterogeneous architectures. The participating team at multiple instritutions improved and extended software libraries (Chroma, CPS, MILC, QLUA, QPhiX, QDP++, QDP-jit) by providing interfaces for heterogeneous computing environments and through the optimization of gauge field evolution algorithms and sparse matrix inverters. The former includes the development of new multi-time scale integrators and the latter focused on the development of new multi-grid and domain decomposition inverters for several QCD fermion discretization schemes. The sub-project at the University of North Carolina worked towards the development a domain specific language for lattice QCD computations with the express purpose of generating code with uncompromised performance.

97 MATHEMATICS AND COMPUTING↗

The QDP/PLT user's guide

PLT is a high level plotting package. A Programmer can create a default plot suited for the data being displayed. At run times, users can then interact with the plot overriding any or all of these defaults. The user is also provided the capability to fit functions to the displayed data. This ability to display, interact with, and to fit the data make PLT a useful tool in the analysis of data. The Quick and Dandy Plotter (QDP) program will read ASCII text files that contain PLT commands and data. Thus, QDP provides and easy way to use the PLT software QPD files provide a convenient way to exchange data. The QPD/PLT software is written in standard FORTRAN 77 and has been ported to VAX VMS, SUN UNIX, IBM AIX, NeXT NextStep, and MS-DOS systems.

Tennant, Allyn F.↗

A Framework for Lattice QCD Calculations on GPUs

Efficient calculation of the greatest common divisor (GCD) for big integers each whose number of bits is greater than or equal to 1024 has drawn a considerable amount of attention because it can be used to detect a weakness of the RSA security infrastructure. This paper presents a parallel binary GCD algorithm and its implementation for big integers on the Intel Xeon Phi coprocessor. This algorithm is capable of computing GCDs efficiently on many pairs of big integers in parallel by utilizing all cores on a Xeon Phi coprocessor as well as taking advantage of all vector processing units of the coprocessor to speed up critical integer operations within the algorithm. Using 240 threads on a Xeon Phi coprocessor to carry out GCD calculations for a large amount of 2048-bit integers, the implementation achieves the speedup of 30 times over a sequential binary GCD algorithm implementation on a single CPU core, and it delivers twice amount of performance in comparison to the same sequential binary GCD implementation running on 240 threads of the Xeon Phi.

Winter, F. T.↗

Parallelizing the QUDA Library for Multi-GPU Calculations in Lattice Quantum Chromodynamics

Graphics Processing Units (GPUs) are having a transformational effect on numerical lattice quantum chromodynamics (LQCD) calculations of importance in nuclear and particle physics. The QUDA library provides a package of mixed precision sparse matrix linear solvers for LQCD applications, supporting single GPUs based on NVIDIA's Compute Unified Device Architecture (CUDA). This library, interfaced to the QDP++/Chroma framework for LQCD calculations, is currently in production use on the "9g" cluster at the Jefferson Laboratory, enabling unprecedented price/performance for a range of problems in LQCD. Nevertheless, memory constraints on current GPU devices limit the problem sizes that can be tackled. In this contribution we describe the parallelization of the QUDA library onto multiple GPUs using MPI, including strategies for the overlapping of communication and computation. We report on both weak and strong scaling for up to 32 GPUs interconnected by InfiniBand, on which we sustain in excess of 4 Tflops.

43 PARTICLE ACCELERATORS↗

Modelling default and likelihood reasoning as probabilistic reasoning

A probabilistic analysis of plausible reasoning about defaults and about likelihood is presented. Likely and by default are in fact treated as duals in the same sense as possibility and necessity. To model these four forms probabilistically, a qualitative default probabilistic (QDP) logic and its quantitative counterpart DP are derived that allow qualitative and corresponding quantitative reasoning. Consistency and consequent results for subsets of the logics are given that require at most a quadratic number of satisfiability tests in the underlying propositional logic. The quantitative logic shows how to track the propagation error inherent in these reasoning forms. The methodology and sound framework of the system highlights their approximate nature, the dualities, and the need for complementary reasoning about relevance.

Buntine, Wray↗

Modelling default and likelihood reasoning as probabilistic

A probabilistic analysis of plausible reasoning about defaults and about likelihood is presented. 'Likely' and 'by default' are in fact treated as duals in the same sense as 'possibility' and 'necessity'. To model these four forms probabilistically, a logic QDP and its quantitative counterpart DP are derived that allow qualitative and corresponding quantitative reasoning. Consistency and consequence results for subsets of the logics are given that require at most a quadratic number of satisfiability tests in the underlying propositional logic. The quantitative logic shows how to track the propagation error inherent in these reasoning forms. The methodology and sound framework of the system highlights their approximate nature, the dualities, and the need for complementary reasoning about relevance.

Buntine, Wray↗

Long-Wavelength Infrared (LWIR) Quantum Dot Infrared Photodetector (QDIP) Focal Plane Array

We have exploited the artificial atomlike properties of epitaxially self-assembled quantum dots for the development of high operating temperature long wavelength infrared (LWIR) focal plane arrays. Quantum dots are nanometer-scale islands that form spontaneously on a semiconductor substrate due to lattice mismatch. QDIPs are expected to outperform quantum well infrared detectors (QWIPs) and are expected to offer significant advantages over II-VI material based focal plane arrays. QDIPs are fabricated using robust wide bandgap III-V materials which are well suited to the production of highly uniform LWIR arrays. We have used molecular beam epitaxy (MBE) technology to grow multi-layer LWIR quantum dot structures based on the InAs/InGaAs/GaAs material system. JPL is building on its significant QWIP experience and is basically building a Dot-in-the-Well (DWELL) device design by embedding InAs quantum dots in a QWIP structure. This hybrid quantum dot/quantum well device offers additional control in wavelength tuning via control of dot-size and/or quantum well sizes. In addition the quantum wells can trap electrons and aide in ground state refilling. Recent measurements have shown a 10 times higher photoconductive gain than the typical QWIP device, which indirectly confirms the lower relaxation rate of excited electrons (photon bottleneck) in QDPs. Subsequent material and device improvements have demonstrated an absorption quantum efficiency (QE) of approx. 3%. Dot-in-the-well (DWELL) QDIPs were also experimentally shown to absorb both 45 deg. and normally incident light. Thus we have employed a reflection grating structure to further enhance the quantum efficiency. JPL has demonstrated wavelength control by progressively growing material and fabricating devices structures that have continuously increased in LWIR response. The most recent devices exhibit peak responsivity out to 8.1 microns. Peak detectivity of the 8.1 micrometer devices has reached approx. 1 x 10(exp 10) Jones at 77 K. Furthermore, we have fabricated the first long-wavelength 640x512 pixels QDP focal plane array. This QDIP focal plane may has produced excellent infrared imagery with noise equivalent temperature difference of 40 mK at 60K operating temperature. In addition, we have managed to increase the quantum efficiency of these devices from 0.1% (according to the data published in literature) to 20% in discrete devices. This is a factor of 200 increase in quantum efficiency. With these excellent results, for the first time QDIP performance has surpassed the QWIP performance. Our goal is to operate these long-wavelength detectors at much higher operating temperature than 77K which can be passively achieved in space. This will be a huge leap in high performance infrared detectors specifically applicable to space science instruments.

infrared imaging↗