Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel processors”

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.

At least 1,711 records · Page 95

OpenACC unified programming environment for GPU and FPGA multi-hybrid acceleration

Attached accelerators have been frequently used in recent High Per- formance Computing (HPC) systems because of their high performance/power ratio. In particular, the Graphics Processing Unit (GPU) is the most popu- lar accelerator owing to its high peak FLOPS performance and high memory bandwidth supported by HBM2, etc. However, the performance of GPU depends highly on a large degree of SIMD parallelism and has difficulty sustaining a high performance on programs with frequent branch operations or a partially low degree of parallelism.By contrast, a Field Programmable Gate Array (FPGA) has received attention as a different type of accelerator than GPU as a fully reconfigurable processor fitting the target applications. The high performance of FPGA is mainly provided by a pipelined operation and optimized circuit suitable for any operation even with frequent conditional branches. We have been focusing on the flexibility of FPGA to compensate for the weakness of GPU. We believe that the coupling of GPU with FPGA can result in one of the most powerful accelerating platforms available.However, the program coding of GPU and FPGA coupling can be quite difficult for application users. Traditionally, CUDA by NVIDIA has been the most popular programming language with the largest share of GPUs used in HPC, whereas a hardware description language such as Verilog HDL has been used in FPGA programming. OpenCL coding has recently become available even on high-end FPGAs. Moreover, several recent studies have also enabled the OpenACC coding for use in FPGA. In this study, we provide a unified programming system based on OpenACC for a platform equipped with both GPU and FPGA aiming at the next-generation accelerated supercomputer framework. Our programming environment is called Multi-Hybrid OpenACC Translator (MHOAT), and in this paper, we describe the basic concept and prototype system of MHOAT based on an evaluation on the amount of coding required and the performance of a hybrid multi-device accelerated system.

Tsunashima, Ryuta↗

Towards Microsecond Biological Molecular Dynamics Simulations on Hybrid Processors

Biomolecular simulations continue to become an increasingly important component of molecular biochemistry and biophysics investigations. Performance improvements in the simulations based on molecular dynamics (MD) codes are widely desired. This is particularly driven by the rapid growth of biological data due to improvements in experimental techniques. Unfortunately, the factors, which allowed past performance improvements of MD simulations, particularly the increase in microprocessor clock frequencies, are no longer improving. Hence, novel software and hardware solutions are being explored for accelerating the performance of popular MD codes. In this paper, we describe our efforts to port and optimize LAMMPS, a popular MD framework, on hybrid processors: graphical processing units (GPUs) accelerated multi-core processors. Our implementation is based on porting the computationally expensive, non-bonded interaction terms on the GPUs, and overlapping the computation on the CPU and GPUs. This functionality is built on top of message passing interface (MPI) that allows multi-level parallelism to be extracted even at the workstation level with the multi-core CPUs as well as extend the implementation on GPU clusters. The results from a number of typically sized biomolecular systems are provided and analysis is performed on 3 generations of GPUs from NVIDIA. Our implementation allows up to 30-40 ns/day throughput on a single workstation as well as significant speedup over Cray XT5, a high-end supercomputing platform. Moreover, detailed analysis of the implementation indicates that further code optimization and improvements in GPUs will allow {approx}100 ns/day throughput on workstations and inexpensive GPU clusters, putting the widely-desired microsecond simulation time-scale within reach to a large user community.

59 BASIC BIOLOGICAL SCIENCES↗

SALT: The Simulator for the Analysis of LWP Timing

With the emergence of new processor architectures that are highly multithreaded, and support features such as full/empty memory semantics and split-phase memory transactions, the need for a processor simulator to handle these features becomes apparent. This paper describes such a simulator, called SALT.

simulation↗

Accelerating Subsurface Transport Simulation on Heterogeneous Clusters

Reactive transport numerical models simulate chemical and microbiological reactions that occur along a flowpath. These models have to compute reactions for a large number of locations. They solve the set of ordinary differential equations (ODEs) that describes the reaction for each location through the Newton-Raphson technique. This technique involves computing a Jacobian matrix and a residual vector for each set of equation, and then solving iteratively the linearized system by performing Gaussian Elimination and LU decomposition until convergence. STOMP, a well known subsurface flow simulation tool, employs matrices with sizes in the order of 100x100 elements and, for numerical accuracy, LU factorization with full pivoting instead of the faster partial pivoting. Modern high performance computing systems are heterogeneous machines whose nodes integrate both CPUs and GPUs, exposing unprecedented amounts of parallelism. To exploit all their computational power, applications must use both the types of processing elements. For the case of subsurface flow simulation, this mainly requires implementing efficient batched LU-based solvers and identifying efficient solutions for enabling load balancing among the different processors of the system. In this paper we discuss two approaches that allows scaling STOMP's performance on heterogeneous clusters. We initially identify the challenges in implementing batched LU-based solvers for small matrices on GPUs, and propose an implementation that fulfills STOMP's requirements. We compare this implementation to other existing solutions. Then, we combine the batched GPU solver with an OpenMP-based CPU solver, and present an adaptive load balancer that dynamically distributes the linear systems to solve between the two components inside a node. We show how these approaches, integrated into the full application, provide speed ups from 6 to 7 times on large problems, executed on up to 16 nodes of a cluster with two AMD Opteron 6272 and a Tesla M2090 per node.

GPU↗

Probing Quantum Telecloning on Superconducting Quantum Processors

Quantum information cannot be perfectly cloned, but approximate copies of quantum information can be generated. Quantum telecloning combines approximate quantum cloning, more typically referred to as quantum cloning, and quantum teleportation. Quantum telecloning allows approximate copies of quantum information to be constructed by separate parties, using the classical results of a Bell measurement made on a prepared quantum telecloning state. Quantum telecloning can be implemented as a circuit on quantum computers using a classical coprocessor to compute classical feedforward instructions using if statements based on the results of a midcircuit Bell measurement in real time. We present universal symmetric optimal 1 → M telecloning circuits and experimentally demonstrate these quantum telecloning circuits for M = 2 up to M = 10, natively executed with real-time classical control systems on IBM Quantum superconducting processors, known as dynamic circuits. We perform the cloning procedure on many different message states across the Bloch sphere, on seven IBM Quantum processors, optionally using the error suppression technique X–X sequence digital dynamical decoupling. Two circuit optimizations are utilized: one that removes ancilla qubits for M = 2,3 , and one that reduces the total number of gates in the circuit but still uses ancilla qubits. Parallel single-qubit tomography with maximum likelihood estimation density matrix reconstruction is used in order to compute the mixed-state density matrices of the clone qubits, and clone quality is measured using quantum fidelity. These results present one of the largest and most comprehensive noisy intermediate-scale quantum computer experimental analyses on (single qubit) quantum telecloning to date. The clone fidelity sharply decreases to 0.5 for M > 5 , but for M = 2 , we are able to achieve a mean clone fidelity of up to 0.79 using dynamical decoupling.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Three-dimensional direct particle simulation on the Connection Machine

This paper presents the algorithms necessary for an efficient data parallel implementation of a 3D particle simulation. In particular, a general master/slave algorithm and a fast sorting algorithm are described and the use of these algorithms in a particle simulation is outlined. A particle simulation using these algorithms has been implemented on a 32768 processor Connection Machine that is capable of simulating over 30 million particles at an average rate of 2.4-microsec/particle/step. Results are presented from the simulation of flow over an Aeroassisted Flight Experiment geometry at 100 km altitude.

Dagum, Leonardo↗

Parallel computation of aerodynamic influence coefficients for aeroelastic analysis on a transputer network

Aeroelastic analysis is mult-disciplinary and computationally expensive. Hence, it can greatly benefit from parallel processing. As part of an effort to develop an aeroelastic analysis capability on a distributed-memory transputer network, a parallel algorithm for the computation of aerodynamic influence coefficients is implemented on a network of 32 transputers. The aerodynamic influence coefficients are calculated using a three-dimensional unsteady aerodynamic model and a panel discretization. Efficiencies up to 85 percent are demonstrated using 32 processors. The effects of subtask ordering, problem size and network topology are presented. A comparison to results on a shared-memory computer indicates that higher speedup is achieved on the distributed-memory system.

Janetzke, D. C.↗

Satisfiability Test with Synchronous Simulated Annealing on the Fujitsu AP1000 Massively-Parallel Multiprocessor

Solving the hard Satisfiability Problem is time consuming even for modest-sized problem instances. Solving the Random L-SAT Problem is especially difficult due to the ratio of clauses to variables. This report presents a parallel synchronous simulated annealing method for solving the Random L-SAT Problem on a large-scale distributed-memory multiprocessor. In particular, we use a parallel synchronous simulated annealing procedure, called Generalized Speculative Computation, which guarantees the same decision sequence as sequential simulated annealing. To demonstrate the performance of the parallel method, we have selected problem instances varying in size from 100-variables/425-clauses to 5000-variables/21,250-clauses. Experimental results on the AP1000 multiprocessor indicate that our approach can satisfy 99.9 percent of the clauses while giving almost a 70-fold speedup on 500 processors.

Sohn, Andrew↗

Spectral Element Method for the Simulation of Unsteady Compressible Flows

This work uses a discontinuous-Galerkin spectral-element method (DGSEM) to solve the compressible Navier-Stokes equations [1{3]. The inviscid ux is computed using the approximate Riemann solver of Roe [4]. The viscous fluxes are computed using the second form of Bassi and Rebay (BR2) [5] in a manner consistent with the spectral-element approximation. The method of lines with the classical 4th-order explicit Runge-Kutta scheme is used for time integration. Results for polynomial orders up to p = 15 (16th order) are presented. The code is parallelized using the Message Passing Interface (MPI). The computations presented in this work are performed using the Sandy Bridge nodes of the NASA Pleiades supercomputer at NASA Ames Research Center. Each Sandy Bridge node consists of 2 eight-core Intel Xeon E5-2670 processors with a clock speed of 2.6Ghz and 2GB per core memory. On a Sandy Bridge node the Tau Benchmark [6] runs in a time of 7.6s.

Diosady, Laslo Tibor↗

Scalable I/O Tracing and Analysis

As supercomputer performance approached and then surpassed the petaflop level, I/O performance has become a major performance bottleneck for many scientific applications. Several tools exist to collect I/O traces to assist in the analysis of I/O performance problems. However, these tools either produce extremely large trace files that complicate performance analysis, or sacrifice accuracy to collect high-level statistical information. We propose a multi-level trace generator tool, ScalaIOTrace, that collects traces at several levels in the HPC I/O stack. ScalaIOTrace features aggressive trace compression that generates trace files of near constant size for regular I/O patterns and orders of magnitudes smaller for less regular ones. This enables the collection of I/O and communication traces of applications running on thousands of processors. Our contributions also include automated trace analysis to collect selected statistical information of I/O calls by parsing the compressed trace on-the-fly and time-accurate replay of communication events with MPI-IO calls. We evaluated our approach with the Parallel Ocean Program (POP) climate simulation and the FLASH parallel I/O benchmark. POP uses NetCDF as an I/O library while FLASH I/O uses the parallel HDF5 I/O library, which internally maps onto MPI-IO. We collected MPI-IO and low-level POSIX I/O traces to study application I/O behavior. Our results show constant size trace files of only 145KB irrespective of the number of nodes for FLASH I/O benchmark, which exhibits regular I/O and communication pattern. For POP, we observe up to two orders of magnitude reduction in trace file sizes compared to flat traces. Statistical information gathered reveals insight on the number of I/O and communication calls issued in the POP and FLASH I/O. Such concise traces are unprecedented for isolated I/O and combined I/O plus communication tracing.

97 MATHEMATICS AND COMPUTING↗

High performance graphics processor based computed tomography reconstruction algorithms for nuclear and other large scale applications.

The goal of this work is to develop a fast computed tomography (CT) reconstruction algorithm based on graphics processing units (GPU) that achieves significant improvement over traditional central processing unit (CPU) based implementations. The main challenge in developing a CT algorithm that is capable of handling very large datasets is parallelizing the algorithm in such a way that data transfer does not hinder performance of the reconstruction algorithm. General Purpose Graphics Processing (GPGPU) is a new technology that the Science and Technology (S&T) community is starting to adopt in many fields where CPU-based computing is the norm. GPGPU programming requires a new approach to algorithm development that utilizes massively multi-threaded environments. Multi-threaded algorithms in general are difficult to optimize since performance bottlenecks occur that are non-existent in single-threaded algorithms such as memory latencies. If an efficient GPU-based CT reconstruction algorithm can be developed; computational times could be improved by a factor of 20. Additionally, cost benefits will be realized as commodity graphics hardware could potentially replace expensive supercomputers and high-end workstations. This project will take advantage of the CUDA programming environment and attempt to parallelize the task in such a way that multiple slices of the reconstruction volume are computed simultaneously. This work will also take advantage of the GPU memory by utilizing asynchronous memory transfers, GPU texture memory, and (when possible) pinned host memory so that the memory transfer bottleneck inherent to GPGPU is amortized. Additionally, this work will take advantage of GPU-specific hardware (i.e. fast texture memory, pixel-pipelines, hardware interpolators, and varying memory hierarchy) that will allow for additional performance improvements.

97 MATHEMATICS AND COMPUTING↗

A parallel vectorized implementation of triple excitations in CCSD(T) - Application to the binding energies of the AlH3, AlH2F, AlHF2 and AlF3 dimers

An efficient method for various noniterative estimates of connected triple excitations in coupled-cluster theory is outlined and related to a similar expression occurring in Moller-Plesset perturbation theory. The method is highly vectorized and capable of utilizing multiple processors on a shared-memory machine, leading to computational rates in excess of one billion floating-point operations per second on four processors of a CRAY Y-MP. Using the new procedure, the binding energies of the D(2h) diborane-type dimers of AlH3, AlH2F, AlHF2, and AlF3 have been determined to be 32, 40, 20, and 47 kcal/mol, respectively. For Al2F6, the correlation procedure includes 232 molecular orbitals and over 1.5 x 10 to the 6th single and double coupled-cluster amplitudes, effectively accounting for over 2 x 10 to the 9th connected triple excitations.

Rendell, Alistair P.↗

High Fidelity Simulations of Large-Scale Wireless Networks

The worldwide proliferation of wireless connected devices continues to accelerate. There are 10s of billions of wireless links across the planet with an additional explosion of new wireless usage anticipated as the Internet of Things develops. Wireless technologies do not only provide convenience for mobile applications, but are also extremely cost-effective to deploy. Thus, this trend towards wireless connectivity will only continue and Sandia must develop the necessary simulation technology to proactively analyze the associated emerging vulnerabilities. Wireless networks are marked by mobility and proximity-based connectivity. The de facto standard for exploratory studies of wireless networks is discrete event simulations (DES). However, the simulation of large-scale wireless networks is extremely difficult due to prohibitively large turnaround time. A path forward is to expedite simulations with parallel discrete event simulation (PDES) techniques. The mobility and distance-based connectivity associated with wireless simulations, however, typically doom PDES and fail to scale (e.g., OPNET and ns-3 simulators). We propose a PDES-based tool aimed at reducing the communication overhead between processors. The proposed solution will use light-weight processes to dynamically distribute computation workload while mitigating communication overhead associated with synchronizations. This work is vital to the analytics and validation capabilities of simulation and emulation at Sandia. We have years of experience in Sandia’s simulation and emulation projects (e.g., MINIMEGA and FIREWHEEL). Sandia’s current highly-regarded capabilities in large-scale emulations have focused on wired networks, where two assumptions prevent scalable wireless studies: (a) the connections between objects are mostly static and (b) the nodes have fixed locations.

Onunkwo, Uzoma↗

Efficient parallel algorithms for string editing and related problems

The string editing problem for input strings x and y consists of transforming x into y by performing a series of weighted edit operations on x of overall minimum cost. An edit operation on x can be the deletion of a symbol from x, the insertion of a symbol in x or the substitution of a symbol x with another symbol. This problem has a well known O((absolute value of x)(absolute value of y)) time sequential solution (25). The efficient Program Requirements Analysis Methods (PRAM) parallel algorithms for the string editing problem are given. If m = ((absolute value of x),(absolute value of y)) and n = max((absolute value of x),(absolute value of y)), then the CREW bound is O (log m log n) time with O (mn/log m) processors. In all algorithms, space is O (mn).

Apostolico, Alberto↗

Parallel asynchronous hardware implementation of image processing algorithms

Research is being carried out on hardware for a new approach to focal plane processing. The hardware involves silicon injection mode devices. These devices provide a natural basis for parallel asynchronous focal plane image preprocessing. The simplicity and novel properties of the devices would permit an independent analog processing channel to be dedicated to every pixel. A laminar architecture built from arrays of the devices would form a two-dimensional (2-D) array processor with a 2-D array of inputs located directly behind a focal plane detector array. A 2-D image data stream would propagate in neuron-like asynchronous pulse-coded form through the laminar processor. No multiplexing, digitization, or serial processing would occur in the preprocessing state. High performance is expected, based on pulse coding of input currents down to one picoampere with noise referred to input of about 10 femtoamperes. Linear pulse coding has been observed for input currents ranging up to seven orders of magnitude. Low power requirements suggest utility in space and in conjunction with very large arrays. Very low dark current and multispectral capability are possible because of hardware compatibility with the cryogenic environment of high performance detector arrays. The aforementioned hardware development effort is aimed at systems which would integrate image acquisition and image processing.

Coon, Darryl D.↗

Optimal parallel evaluation of AND trees

A quantitative analysis based on both preemptive and nonpreemptive critical-path scheduling algorithms is presently conducted for the optimal degree of parallelism required in evaluating a given AND tree. The optimal degree of parallelism is found to depend on problem complexity, precedence-graph shape, and task-time distribution along each path. In addition to demonstrating the optimality of the preemptive critical-path scheduling algorithm for evaluating an arbitrary AND tree on a fixed number of processors, the possibility of efficiently ascertaining tight bounds on the number of processors for optimal processor-time efficiency is illustrated.

Wah, Benjamin W.↗

Performance of an Optimized Eta Model Code on the Cray T3E and a Network of PCs

In the year 2001, NASA will launch the satellite TRIANA that will be the first Earth observing mission to provide a continuous, full disk view of the sunlit Earth. As a part of the HPCC Program at NASA GSFC, we have started a project whose objectives are to develop and implement a 3D cloud data assimilation system, by combining TRIANA measurements with model simulation, and to produce accurate statistics of global cloud coverage as an important element of the Earth's climate. For simulation of the atmosphere within this project we are using the NCEP/NOAA operational Eta model. In order to compare TRIANA and the Eta model data on approximately the same grid without significant downscaling, the Eta model will be integrated at a resolution of about 15 km. The integration domain (from -70 to +70 deg in latitude and 150 deg in longitude) will cover most of the sunlit Earth disc and will continuously rotate around the globe following TRIANA. The cloud data assimilation is supposed to run and produce 3D clouds on a near real-time basis. Such a numerical setup and integration design is very ambitious and computationally demanding. Thus, though the Eta model code has been very carefully developed and its computational efficiency has been systematically polished during the years of operational implementation at NCEP, the current MPI version may still have problems with memory and efficiency for the TRIANA simulations. Within this work, we optimize a parallel version of the Eta model code on a Cray T3E and a network of PCs (theHIVE) in order to improve its overall efficiency. Our optimization procedure consists of introducing dynamically allocated arrays to reduce the size of static memory, and optimizing on a single processor by splitting loops to limit the number of streams. All the presented results are derived using an integration domain centered at the equator, with a size of 60 x 60 deg, and with horizontal resolutions of 1/2 and 1/3 deg, respectively. In accompanying charts we report the elapsed time, the speedup and the Mflops as a function of the number of processors for the non-optimized version of the code on the T3E and theHIVE. The large amount of communication required for model integration explains its poor performance on theHIVE. Our initial implementation of the dynamic memory allocation has contributed to about 12% reduction of memory but has introduced a 3% overhead in computing time. This overhead was removed by performing loop splitting in some of the high demanding subroutines. When the Eta code is fully optimized in order to meet the memory requirement for TRIANA simulations, a non-negligeable overhead may appear that may seriously affect the efficiency of the code. To alleviate this problem, we are considering implementation of a new algorithm for the horizontal advection that is computationally less expensive, and also a new approach for marching in time.

Kouatchou, Jules↗

Multi-petascale highly efficient parallel supercomputer

A Multi-Petascale Highly Efficient Parallel Supercomputer of 100 petaOPS-scale computing, at decreased cost, power and footprint, and that allows for a maximum packaging density of processing nodes from an interconnect point of view. The Supercomputer exploits technological advances in VLSI that enables a computing model where many processors can be integrated into a single Application Specific Integrated Circuit (ASIC). Each ASIC computing node comprises a system-on-chip ASIC utilizing four or more processors integrated into one die, with each having full access to all system resources and enabling adaptive partitioning of the processors to functions such as compute or messaging I/O on an application by application basis, and preferably, enable adaptive partitioning of functions in accordance with various algorithmic phases within an application, or if I/O or other processors are underutilized, then can participate in computation or communication nodes are interconnected by a five dimensional torus network with DMA that optimally maximize the throughput of packet communications between nodes and minimize latency.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗