Search NASA⌕ Search

SEARCH · Search NASA

Results for “processor performance factors”

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.

115 records · Page 7

International Linear Collider Accelerator Physics R&D

ILC work at Illinois has concentrated primarily on technical issues relating to the design of the accelerator. Because many of the problems to be resolved require a working knowledge of classical mechanics and electrodynamics, most of our research projects lend themselves well to the participation of undergraduate research assistants. The undergraduates in the group are scientists, not technicians, and find solutions to problems that, for example, have stumped PhD-level staff elsewhere. The ILC Reference Design Report calls for 6.7 km circumference damping rings (which prepare the beams for focusing) using “conventional” stripline kickers driven by fast HV pulsers. Our primary goal was to determine the suitability of the 16 MeV electron beam in the AØ region at Fermilab for precision kicker studies.We found that the low beam energy and lack of redundancy in the beam position monitor system complicated the analysis of our data. In spite of these issues we concluded that the precision we could obtain was adequate to measure the performance and stability of a production module of an ILC kicker, namely 0.5%. We concluded that the kicker was stable to an accuracy of ~2.0% and that we could measure this precision to an accuracy of ~0.5%. As a result, a low energy beam like that at AØ could be used as a rapid-turnaround facility for testing ILC production kicker modules. The ILC timing precision for arrival of bunches at the collision point is required to be 0.1 picosecond or better. We studied the bunch-to-bunch timing accuracy of a “phase detector” installed in AØ in order to determine its suitability as an ILC bunch timing device. A phase detector is an RF structure excited by the passage of a bunch. Its signal is fed through a 1240 MHz high-Q resonant circuit and then down-mixed with the AØ 1300 MHz accelerator RF. We used a kind of autocorrelation technique to compare the phase detector signal with a reference signal obtained from the phase detector’s response to an event at the beginning of the run. We determined that the device installed in our beam, which was instrumented with an 8-bit 500 MHz ADC, could measure the beam timing to an accuracy of 0.4 picoseconds. Simulations of the device showed that an increase in ADC clock rate to 2 GHz would improve measurement precision by the required factor of four. As a result, we felt that a device of this sort, assuming matters concerning dynamic range and long-term stability can be addressed successfully, would work at the ILC. Cost effective operation of the ILC will demand highly reliable, fault tolerant and adaptive solutions for both hardware and software. The large numbers of subsystems and large multipliers associated with the modules in those subsystems will cause even a strong level of unit reliability to become an unacceptable level of system availability. An evaluation effort is underway to evaluate standards associated with high availability, and to guide ILC development with standard practices and well-supported commercial solutions. One area of evaluation involves the Advanced Telecom Computing Architecture (ATCA) hardware and software. We worked with an ATCA crate, processor monitors, and a small amount of ATCA circuit boards in order to develop a backplane “spy” board that would let us watch the ATCA backplane communications and pursue development of an inexpensive processor monitor that could be used as a physics-driven component of the crate-level controls system. We made good progress, and felt that we had determined a productive direction to extend this work. We felt that we had learned enough to begin designing a workable processor monitor chip if there were to be sufficient interest in ATCA shown by the ILC community. Fault recognition is a challenging issue in the crafting a high reliability controls system. With tens of thousands of independent processors running hundreds of thousands of critical processes, how can the system identify that a problem has arisen and determine the appropriate steps to take to correct, or compensate, for the failure? One possible solution might come through the use of the OpenClovis supervisory system, which runs on Linux processors and allows a select set of processors to monitor the behavior of individual processes and processors in a large, distributed controls network. We found that OpenClovis exhibited an irritating amount of sensitivity to the exact version of the Linux kernel running on the processors, and that it was poorly equipped to help us sort through problems that arose through conflicts so deep in the operating systems of the processors. But once this issue was addressed, we found that it performed as expected, recognizing crashes and process (and processor) failures.

43 PARTICLE ACCELERATORS↗

Roofline Analysis in the Intel® Advisor to Deliver Optimized Performance for applications on Intel® Xeon Phi™ Processor

In this session we show, in two case studies, how the roofline feature of Intel Advisor has been utilized to optimize the performance of kernels of the XGC1 and PICSAR codes in preparation for Intel Knights Landing architecture. The impact of the implemented optimizations and the benefits of using the automatic roofline feature of Intel Advisor to study performance of large applications will be presented. This demonstrates an effective optimization strategy that has enabled these science applications to achieve up to 4.6 times speed-up and prepare for future exascale architectures. # Goal/Relevance of Session The roofline model [1,2] is a powerful tool for analyzing the performance of applications with respect to the theoretical peak achievable on a given computer architecture. It allows one to graphically represent the performance of an application in terms of operational intensity, i.e. the ratio of flops performed and bytes moved from memory in order to guide optimization efforts. Given the scale and complexity of modern science applications, it can often be a tedious task for the user to perform the analysis on the level of functions or loops to identify where performance gains can be made. With new Intel tools, it is now possible to automate this task, as well as base the estimates of peak performance on measurements rather than vendor specifications. The goal of this session is to demonstrate how the roofline feature of Intel Advisor can be used to balance memory vs. computation related optimization efforts and effectively identify performance bottlenecks. A series of typical optimization techniques: cache blocking, structure refactoring, data alignment, and vectorization illustrated by the kernel cases will be addressed. # Description of the codes ## XGC1 The XGC1 code [3] is a magnetic fusion Particle-In-Cell code that uses an unstructured mesh for its Poisson solver that allows it to accurately resolve the edge plasma of a magnetic fusion device. After recent optimizations to its collision kernel [4], most of the computing time is spent in the electron push (pushe) kernel, where these optimization efforts have been focused. The kernel code scaled well with MPI+OpenMP but had almost no automatic compiler vectorization, in part due to indirect memory addresses and in part due to low trip counts of low-level loops that would be candidates for vectorization. Particle blocking and sorting have been implemented to increase trip counts of low-level loops and improve memory locality, and OpenMP directives have been added to vectorize compute-intensive loops that were identified by Advisor. The optimizations have improved the performance of the pushe kernel 2x on Haswell processors and 1.7x on KNL. The KNL node-for-node performance has been brought to within 30% of a NERSC Cori phase I Haswell node and we expect to bridge this gap by reducing the memory footprint of compute intensive routines to improve cache reuse. ## PICSAR is a Fortran/Python high-performance Particle-In-Cell library targeting at MIC architectures first designed to be coupled with the PIC code WARP for the simulation of laser-matter interaction and particle accelerators. PICSAR also contains a FORTRAN stand-alone kernel for performance studies and benchmarks. A MPI domain decomposition is used between NUMA domains and a tile decomposition (cache-blocking) handled by OpenMP has been added for shared-memory parallelism and better cache management. The so-called current deposition and field gathering steps that compose the PIC time loop constitute major hotspots that have been rewritten to enable more efficient vectorization. Particle communications between tiles and MPI domain has been merged and parallelized. All considered, these improvements provide speedups of 3.1 for order 1 and 4.6 for order 3 interpolation shape factors on KNL configured in SNC4 quadrant flat mode. Performance is similar between a node of cori phase 1 and KNL at order 1 and better on KNL by a factor 1.6 at order 3 with the considered test case (homogeneous thermal plasma).

97 MATHEMATICS AND COMPUTING↗

AMPS: An Augmented Matrix Formulation for Principal Submatrix Updates with Application to Power Grids

We present an augmented matrix approach to update the solution to a linear system of equations when the coefficient matrix is modified by a few elements within a principal submatrix. This problem arises in the dynamic security analysis of a power grid, where operators need to perform $N-x$ contingency analysis, i.e., determine the state of the system when up to $x$ links from $N$ fail. Our algorithms augment the coefficient matrix to account for the changes in it, and then compute the solution to the augmented system without refactoring the modified matrix. We provide two algorithms, a direct method, and a hybrid direct-iterative method for solving the augmented system. We also exploit the sparsity of the matrices and vectors to accelerate the overall computation. Our algorithms are compared on three power grids with PARDISO, a parallel direct solver, and CHOLMOD, a direct solver with the ability to modify the Cholesky factors of the coefficient matrix. We show that our augmented algorithms outperform PARDISO (by two orders of magnitude), and CHOLMOD (by a factor of up to 5). Further, our algorithms scale better than CHOLMOD as the number of elements updated increases. The solutions are computed with high accuracy. Our algorithms are capable of computing $N-x$ contingency analysis on a $778K$ bus grid, updating a solution with $x=20$ elements in $1.6 \times 10^{-2}$ seconds on an Intel Xeon processor.

Yeung, Yu-Hong↗

The Argonne Leadership Computing Facility 2010 annual report.

Researchers found more ways than ever to conduct transformative science at the Argonne Leadership Computing Facility (ALCF) in 2010. Both familiar initiatives and innovative new programs at the ALCF are now serving a growing, global user community with a wide range of computing needs. The Department of Energy's (DOE) INCITE Program remained vital in providing scientists with major allocations of leadership-class computing resources at the ALCF. For calendar year 2011, 35 projects were awarded 732 million supercomputer processor-hours for computationally intensive, large-scale research projects with the potential to significantly advance key areas in science and engineering. Argonne also continued to provide Director's Discretionary allocations - 'start up' awards - for potential future INCITE projects. And DOE's new ASCR Leadership Computing (ALCC) Program allocated resources to 10 ALCF projects, with an emphasis on high-risk, high-payoff simulations directly related to the Department's energy mission, national emergencies, or for broadening the research community capable of using leadership computing resources. While delivering more science today, we've also been laying a solid foundation for high performance computing in the future. After a successful DOE Lehman review, a contract was signed to deliver Mira, the next-generation Blue Gene/Q system, to the ALCF in 2012. The ALCF is working with the 16 projects that were selected for the Early Science Program (ESP) to enable them to be productive as soon as Mira is operational. Preproduction access to Mira will enable ESP projects to adapt their codes to its architecture and collaborate with ALCF staff in shaking down the new system. We expect the 10-petaflops system to stoke economic growth and improve U.S. competitiveness in key areas such as advancing clean energy and addressing global climate change. Ultimately, we envision Mira as a stepping-stone to exascale-class computers that will be faster than petascale-class computers by a factor of a thousand. Pete Beckman, who served as the ALCF's Director for the past few years, has been named director of the newly created Exascale Technology and Computing Institute (ETCi). The institute will focus on developing exascale computing to extend scientific discovery and solve critical science and engineering problems. Just as Pete's leadership propelled the ALCF to great success, we know that that ETCi will benefit immensely from his expertise and experience. Without question, the future of supercomputing is certainly in good hands. I would like to thank Pete for all his effort over the past two years, during which he oversaw the establishing of ALCF2, the deployment of the Magellan project, increases in utilization, availability, and number of projects using ALCF1. He managed the rapid growth of ALCF staff and made the facility what it is today. All the staff and users are better for Pete's efforts.

54 ENVIRONMENTAL SCIENCES↗

Quantum Monte Carlo Endstation for Petascale Computing

The major achievements enabled by QMC Endstation grant include * Performance improvement on clusters of x86 multi-core systems, especially on Cray XT systems * New and improved methods for the wavefunction optimizations * New forms of trial wavefunctions * Implementation of the full application on NVIDIA GPUs using CUDA The scaling studies of QMCPACK on large-scale systems show excellent parallel efficiency up to 216K cores on Jaguarpf (Cray XT5). The GPU implementation shows speedups of 10-15x over the CPU implementation on older generation of x86. We have implemented hybrid OpenMP/MPI scheme in QMC to take advantage of multi-core shared memory processors of petascale systems. Our hybrid scheme has several advantages over the standard MPI-only scheme. * Memory optimized: large read-only data to store one-body orbitals and other shared properties to represent the trial wave function and many-body Hamiltonian can be shared among threads, which reduces the memory footprint of a large-scale problem. * Cache optimized: the data associated with an active Walker are in cache during the compute-intensive drift-diffusion process and the operations on an Walker are optimized for cache reuse. Thread-local objects are used to ensure the data affinity to a thread. * Load balanced: Walkers in an ensemble are evenly distributed among threads and MPI tasks. The two-level parallelism reduces the population imbalance among MPI tasks and reduces the number of point-to-point communications of large messages (serialized objects) for the Walker exchange. * Communication optimized: the communication overhead, especially for the collective operations necessary to determine ET and measure the properties of an ensemble, is significantly lowered by using less MPI tasks. The multiple forms of parallelism afforded by QMC algorithms make them ideal candidates for acceleration in the many-core paradigm. We presented the results of our effort to port the QMCPACK simulation code to the NVIDIA CUDA GPU platform. We restructured the CPU algorithms to express additional parallelism, minimize GPU-CPU communication, and efficiently utilize the GPU memory hierarchy. Using mixed precision on GT200 GPUs and MPI for intercommunication and load balancing, we observe typical full-application speedups of approximately 10x to 15x relative to quad-core Xeon CPUs alone, while reproducing the double-precision CPU results within statistical error. We developed an all-electron quantum Monte Carlo (QMC) method for solids that does not rely on pseudopotentials, and used it to construct a primary ultra-high-pressure calibration based on the equation of state of cubic boron nitride. We computed the static contribution to the free energy with the QMC method and obtained the phonon contribution from density functional theory, yielding a high-accuracy calibration up to 900 GPa usable directly in experiment. We computed the anharmonic Raman frequency shift with QMC simulations as a function of pressure and temperature, allowing optical pressure calibration. In contrast to present experimental approaches, small systematic errors in the theoretical EOS do not increase with pressure, and no extrapolation is needed. This all-electron method is applicable to first-row solids, providing a new reference for ab initio calculations of solids and benchmarks for pseudopotential accuracy. We compared experimental and theoretical results on the momentum distribution and the quasiparticle renormalization factor in sodium. From an x-ray Compton-profile measurement of the valence-electron momentum density, we derived its discontinuity at the Fermi wavevector finding an accurate measure of the renormalization factor that we compared with quantum-Monte-Carlo and G0W0 calculations performed both on crystalline sodium and on the homogeneous electron gas. Our calculated results are in good agreement with the experiment. We have been studying the heat of formation for various Kubas complexes of molecular hydrogen on Ti(1,2)ethylene-nH2 using Diffusion Monte Carlo. This work has been started and is ongoing. We are studying systems involving 1 and 2 Ti bonding sites with up to 10 hydrogen molecules in numerous configurations. This work will establish a benchmark that will test the accuracy of density functional calculations and establish the feasibility of our methods for similar systems.

97 MATHEMATICS AND COMPUTING↗

On the Degradation of the Effectiveness of Nonlinear Diffusion Acceleration with Parallel Block Jacobi Splitting

MOOSE's radiation transport module Rattlesnake comprises a first order S N solver that can be accelerated with a flexible Nonlinear Diffusion Acceleration. At each NDA iteration, the streaming and collision operator of the S{sub N} equations are (approximately) inverted using a transport sweep. When executed on multiple processors parallel block Jacobi (PBJ) splitting is utilized lagging the processor sub-domain's incoming angular fluxes making the implemented method asynchronous, i.e. the convergence properties of NDA become dependent on the number of concurrent tasks and details of the partitioning. Within this work, we study the NDA convergence properties for the two-dimensional C5G7 benchmark problem PBJ with p = 1, 2, 4, 8, 12, 16, 24, 32, 48, 60, 72, 96 and 144 concurrent tasks. Rosa studied the convergence properties of PBJ splitting methods with and without transport synthetic acceleration (TSA) in slab geometry and two-dimensional Cartesian geometry, respectively. Rosa's iteration scheme lags both the incoming angular fluxes and the scattering source making it very similar to the method described here. Rosa found by means of Fourier analysis that convergence of unaccelerated PBJ can degrade for optically thin subdomains, i.e. for a large number of processors given a fixed domain size. Depending on the choice of parameters application of TSA can lead to an effective algorithm. In contrast to Rosa's work, this work focuses on PBJ splitting for NDA accelerated solution of the multigroup eigenvalue problem. Later, Zerr and Anistratov used their integral transport matrix method in conjunction with PBJ. This iteration scheme is different as it does not lag the scattering source. Nevertheless, Fourier analysis with Diamond Difference spatial discretization showed that the method degrades with increasing the overall size of the spatial domain while keeping the sub-domain size constant. In the case of an infinite medium, the observed spectral radius becomes unity. These references establish the iterative performance of PBJ for one-group fixed source problems discretized on rectangular cells, but for the Rattlesnake first order NDA method, an emphasis lies on multigroup eigenvalue problems. In this setting, the additional complication arises that the low order equations are not exactly inverted, but only to some iterative criterion. Further, as the low order problem is solved using Jacobian-Free Newton Krylov methods, a suitable preconditioner for the GMRES iterations is required. As the effectiveness of the preconditioners may decrease when more concurrent processes are used, we expect the low order solve to impact the overall effectiveness of the NDA method in parallel execution as well. Therefore, this work focuses on the question of how detrimental the asynchronicity of the high and low order iterative solutions is for the effectiveness of NDA solutions of realistic, unstructured grid, multigroup eigenvalue problems. The iterative convergence properties of Rattlesnake's NDA method for multigroup eigenvalue problems are studied for varying number of concurrent processes. Rattlesnake's NDA method is asynchronous because the high order transport sweep uses PBJ splitting and because the low order solve's preconditioners, algebraic multigrid and additive Schwarz, are asynchronous. We found that up to 144 processes the number of Picard iterations increases by a factor of 2-3 affecting but not destroying NDA's parallel performance for the C5G7 benchmark problem, a representative LWR (light water reactor) configuration. The solution of the NDA low order eigenvalue problem posed greater problems than expected for the solvers currently available in Rattlesnake mainly because of problems to effectively precondition the modified low-order problem. As a matter of fact, the low order problem becomes an advection-diffusion problem. Non-robustness was observed for the AMG preconditioned coarse mesh acceleration for more than 60 concurrent processes, which was fixed by increasing the number of permissible V-cycles per AMG call. For fine mesh acceleration AMG proved to be ineffective and we resorted to ASM. For both AMG and ASM, increasing number of linear iterations per nonlinear are observed when increasing the number of processes, but for ASM this problem appears to be worse. The results from this work also indicate that initial development efforts should be diverted from implementation of a high-order parallel sweeping algorithm to the effective and robust iterative solution of the low order NDA multigroup eigenvalue problem. (authors)

97 MATHEMATICS AND COMPUTING↗

Investigation of Portable Event-Based Monte Carlo Transport Using the NVIDIA Thrust Library

Power consumption considerations are driving future high performance computing platforms toward many-core computing architectures. Los Alamos National Laboratory's Trinity machine, available in 2016, will use both Intel Xeon Haswell processors and Intel Xeon Phi Knights Landing many integrated core (MIC) architecture coprocessors. Lawrence Livermore National Laboratory's Sierra machine, available in 2018, will use an IBM PowerPC architecture along with Nvidia graphics processing unit (GPU) architecture accelerators. These different advanced architectures make the computing landscape in upcoming years complex. Traditional approaches to Monte Carlo transport do not work efficiently on these new computing platforms. MIC architectures require vectorization to operate efficiently, and vectorization is difficult to achieve in Monte Carlo transport. GPU architectures require additional code to explicitly use the hardware, requiring significant code changes or hardware specific branches in the source code. A significant challenge for Monte Carlo transport projects is to simultaneously support within a single source code base efficient simulations for both the current generation of architectures and the different advanced computing architectures. In order to address these challenges, two important changes are typically required: a new algorithmic approach for solving Monte Carlo transport, and explicit use of hardware specific software. In this paper, we describe initial research investigations of an event-based Monte Carlo transport algorithm implemented using the Nvidia Thrust library on a GPU for a Monte Carlo test code. The event-based algorithm targets many-core architectures by increasing SIMD (single instruction multiple data) parallelism, while Thrust potentially provides portable performance by allowing one source code base to compile code targeted for both CPUs and GPUs. We described preliminary investigations of portable event-based Monte Carlo algorithms implemented using the Nvidia Thrust library in a research Monte Carlo test code. We found that an explicit CUDA implementation of an event-based Monte Carlo algorithm performed significantly more efficiently than a Thrust implementation on GPU platforms, most likely as a result of additional flexibility in access to different memory spaces on the GPU. Additionally, we showed that on GPU platforms and at large enough problem sizes the event-based implementations perform more efficiently than the serial history-based implementation running on the host CPU. While investigating this problem, we also discovered that the performance of the event-based algorithm is affected by what tallies are being used. A zonal scalar flux tally requires atomic operations that significantly impacted the performance of the code, in some cases producing slowdowns instead of speedups. We decided to remove the tally in order to focus on the effectiveness of the event-based algorithm. Future work will be required to research more effective ways of handling such tallies. Additionally, we would like to consider new ways for optimizing both the Thrust and CUDA versions, in order to see how much performance we have yet to achieve. The potential trade-off between portability and performance was demonstrated in this investigation. Thrust provides both CPU and GPU versions of the code in one code base, but it does so at a cost. We discovered approximately a factor of 2-5 performance difference between Thrust and CUDA on each GPU platform we tested. Thrust provides a tool to access the GPUs with less effort and specialization, but it does so by giving up fine grained control where extra performance can be found for this application. Future work will be required to determine whether this performance differential between Thrust and CUDA can be reduced for event-based Monte Carlo transport. (authors)

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗