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 1,621 records · Page 90

Adapting Wave-front Algorithms to Efficiently Utilize Systems with Deep Communication Hierarchies

Large-scale systems increasingly exhibit a differential between intra-chip and inter-chip communication performance especially in hybrid systems using accelerators. Processorcores on the same socket are able to communicate at lower latencies, and with higher bandwidths, than cores on different sockets either within the same node or between nodes. A key challenge is to efficiently use this communication hierarchy and hence optimize performance. We consider here the class of applications that contains wavefront processing. In these applications data can only be processed after their upstream neighbors have been processed. Similar dependencies result between processors in which communication is required to pass boundary data downstream and whose cost is typically impacted by the slowest communication channel in use. In this work we develop a novel hierarchical wave-front approach that reduces the use of slower communications in the hierarchy but at the cost of additional steps in the parallel computation and higher use of on-chip communications. This tradeoff is explored using a performance model. An implementation using the Reverse-acceleration programming model on the petascale Roadrunner system demonstrates a 27% performance improvement at full system-scale on a kernel application. The approach is generally applicable to large-scale multi-core and accelerated systems where a differential in system communication performance exists.

43 PARTICLE ACCELERATORS↗

Comparative Implementation of High Performance Computing for Power System Dynamic Simulations

Dynamic simulation for transient stability assessment is one of the most important, but intensive, computations for power system planning and operation. Present commercial software is mainly designed for sequential computation to run a single simulation, which is very time consuming with a single processer. The application of High Performance Computing (HPC) to dynamic simulations is very promising in accelerating the computing process by parallelizing its kernel algorithms while maintaining the same level of computation accuracy. This paper describes the comparative implementation of four parallel dynamic simulation schemes in two state-of-the-art HPC environments: Message Passing Interface (MPI) and Open Multi-Processing (OpenMP). These implementations serve to match the application with dedicated multi-processor computing hardware and maximize the utilization and benefits of HPC during the development process.

97 MATHEMATICS AND COMPUTING↗

The suitability of the ILLIAC IV architecture for image processing

The major architectural features of the ILLIAC IV large scale, array processor are summarized along with their applicability to image processing. Several image processing algorithms are considered, including multispectral classification, texture feature extraction, two-dimensional Fourier transform, and synthetic aperture radar processing. The basic parallelism of the ILLIAC IV (64 processing elements acting in lock-step) is usually fully utilized by the image processing applications. The major architectural aspect of the system with respect to image processing is the relatively small local scratch-pad memory and the long latency time to access the main storage device. The major precision used for the image processing applications is the 32-bit floating point, given a choice of 8-bit integers and 64-bit floating point.

Stevenson, D. K.↗

Multi-level Hierarchical Poly Tree computer architectures

Based on the concept of hierarchical substructuring, this paper develops an optimal multi-level Hierarchical Poly Tree (HPT) parallel computer architecture scheme which is applicable to the solution of finite element and difference simulations. Emphasis is given to minimizing computational effort, in-core/out-of-core memory requirements, and the data transfer between processors. In addition, a simplified communications network that reduces the number of I/O channels between processors is presented. HPT configurations that yield optimal superlinearities are also demonstrated. Moreover, to generalize the scope of applicability, special attention is given to developing: (1) multi-level reduction trees which provide an orderly/optimal procedure by which model densification/simplification can be achieved, as well as (2) methodologies enabling processor grading that yields architectures with varying types of multi-level granularity.

Padovan, Joe↗

Implementation of a blade element UH-60 helicopter simulation on a parallel computer architecture in real-time

A high-performance platform for development of real-time helicopter flight simulations based on a simulation development and analysis platform combining a parallel simulation development and analysis environment with a scalable multiprocessor computer system is described. Simulation functional decomposition is covered, including the sequencing and data dependency of simulation modules and simulation functional mapping to multiple processors. The multiprocessor-based implementation of a blade-element simulation of the UH-60 helicopter is presented, and a prototype developed for a TC2000 computer is generalized in order to arrive at a portable multiprocessor software architecture. It is pointed out that the proposed approach coupled with a pilot's station creates a setting in which simulation engineers, computer scientists, and pilots can work together in the design and evaluation of advanced real-time helicopter simulations.

Moxon, Bruce C.↗

PLUM: Parallel Load Balancing for Unstructured Adaptive Meshes

Dynamic mesh adaption on unstructured grids is a powerful tool for computing large-scale problems that require grid modifications to efficiently resolve solution features. By locally refining and coarsening the mesh to capture physical phenomena of interest, such procedures make standard computational methods more cost effective. Unfortunately, an efficient parallel implementation of these adaptive methods is rather difficult to achieve, primarily due to the load imbalance created by the dynamically-changing nonuniform grid. This requires significant communication at runtime, leading to idle processors and adversely affecting the total execution time. Nonetheless, it is generally thought that unstructured adaptive- grid techniques will constitute a significant fraction of future high-performance supercomputing. Various dynamic load balancing methods have been reported to date; however, most of them either lack a global view of loads across processors or do not apply their techniques to realistic large-scale applications.

Oliker, Leonid↗

Root-Raised Cosine Filter Implementation That Uses Canonical Signed Digits for High-Speed Digital Filter Applications

NASA Lewis Research Center's Space Communications Division has been investigating high-speed digital filters that can operate at a higher speed than those in current use for a digital modulator and demodulator (modem). Using the Canonical Signed Digits (CSD) number representation for filter coefficients is a very effective way to increase the filter's speed while reducing complexity in the digital filter hardware design. This approach is a good alternative to using an expensive parallel-processing design technique or custom, application-specific integrated circuits. Such integrated circuits may not be suitable for applications that require filter speeds faster than what application-specific integrated circuits digital signal processors can offer for a dedicated channel. When a communication channel is a dedicated, multiplication process--a costly, time-consuming process--it can be greatly simplified by a replacement of the filter coefficients with CSD numbers. A computer code written with the MATLAB software package runs the program and generates CSD-represented filter coefficients that are based on minimizing minimum mean square errors. Also, the Alta Group of Cadence's Signal Processing Workstation is used to simulate and analyze the CSD filter responses. The impulse response of the root-raised cosine filter that is used as a base model is defined. From this filter, a set of coefficients is sampled and stored in a file. For the all coefficients, the optimal CSD number for each coefficient is searched on the basis of the minimum-mean-square-errors criterion. Because the distribution of CSD numbers is not uniform, quantization errors tend to be bigger for coefficients greater than 1/2. To offset errors that occur in a region of coefficients between 1/2 to 1 and to better represent fractions with CSD numbers, an extra nonzero digit is allowed for any coefficients exceeding 1/2. This will greatly improve frequency response as well as intersymbol interference at the receiver. The frequency response of a set of collected CSD-represented filter coefficients was compared with the same filter that was conventionally implemented. Analyses show CSD-implemented filters perform as well as conventional filters. Comparison of eye diagrams and bit-error-rate curves between CSD filters and traditionally implemented filters are almost indistinguishable. However, filter complexity was reduced from almost 3.5 to 1 for CSD filters. Complete computer simulation results are available. In the near future, work will focus on building actual working digital filter hardware in a field programmable gate array (FPGA).

Kim, Heechul↗

Partitioning strategies for parallel KIVA-4 engine simulations

Parallel KIVA-4 is described and simulated in four different engine geometries. The Message Passing-Interface (MPl) was used to parallelize KIVA-4. Par itioning strategies ar accesed in light of the fact that cells can become deactivated and activated during the course of an engine simulation which will affect the load balance between processors.

99 GENERAL AND MISCELLANEOUS↗

...And Eat it Too: High Read Performance in Write-Optimized HPC I/O Middleware File Formats

As HPC applications run on increasingly high process counts on larger and larger machines, both the frequency of checkpoints needed for fault tolerance and the resolution and size of Data Analysis Dumps are expected to increase proportionally. In order to maintain an acceptable ratio of time spent performing useful computation work to time spent performing I/O, write bandwidth to the underlying storage system must increase proportionally to this increase in the checkpoint and computation size. Unfortunately, popular scientific self-describing file formats such as netCDF and HDF5 are designed with a focus on portability and flexibility. Extra care and careful crafting of the output structure and API calls is required to optimize for write performance using these APIs. To provide sufficient write bandwidth to continue to support the demands of scientific applications, the HPC community has developed a number of I/O middleware layers, that structure output into write-optimized file formats. However, the obvious concern with any write optimized file format would be a corresponding penalty on reads. In the log-structured filesystem, for example, a file generated by random writes could be written efficiently, but reading the file back sequentially later would result in very poor performance. Simulation results require efficient read-back for visualization and analytics, and though most checkpoint files are never used, the efficiency of a restart is very important in the face of inevitable failures. The utility of write speed improving middleware would be greatly diminished if it sacrificed acceptable read performance. In this paper we examine the read performance of two write-optimized middleware layers on large parallel machines and compare it to reading data natively in popular file formats. The two I/O middleware layer examined in this paper are the Adaptable IO System (ADIOS), a library-based approach developed at Oak Ridge National Laboratory to provide a high-level IO API that can be used in place of netCDF or HDF5 to do much more aggressive write-behind and efficient reordering of data locations within the file; and the Parallel Log-structured Filesystem (PLFS), a stackable FUSE filesystem approach developed at Los Alamos National Laboratory that decouples concurrent writes to improve the speed of checkpoints. Since ADIOS is an I/O componentization that affords selection of different I/O methods at or during runtime, through a single API, users can have access to MPI-IO, Posix-IO, HDF5, netCDF, and staging methods. The ADIOS BP file format is a new log-file format that has a superset of the features of both HDF5 and netCDF, but is designed to be portable and flexible while being optimized for writing. PLFS, on the other hand, is mounted as a stackable filesystem on top of an existing parallel filesystem. Reads or writes to the PLFS filesystem are transparently translated into operations on per-process log files stored in the underlying parallel filesystem. Since PLFS performs this translation without application modification, users can write in HDF5, netCDF, or app-specific file formats and PLFS will store the writes in a set of efficiently written log-formatted files, while presenting the user with a logical 'flat' file on reads. Despite their different approaches, the commonality behind both of these middleware systems is that they both write to a log file format. As shown in previous publications, writes are fully optimized in both systems, sometimes resulting in 100x improvements over writing data in popular file formats. But as mentioned above, writing is only one part of the story. In this paper we examine the read performance of our middleware layers on large parallel machines and compare these to reading data either natively or from other popular file formats. We compare the reading performance in two different scenarios: (1) Reading back restarts from the same number of processors as wrote the data and (2) Reading back restart data from a different number of processors. We observe that not only can write-optimized I/O middleware be built to not penalize read speeds, but for important workloads, techniques that improve write performance can, perhaps counterintuitively, improve read speeds over reading from a contiguously organized file format. In the remainder of this paper, we investigate this further through case studies of PLFS and ADIOS on simulation checkpoint restarts.

99 GENERAL AND MISCELLANEOUS↗

Substructure analysis using NICE/SPAR and applications of force to linear and nonlinear structures

Parallel computing studies are presented for a variety of structural analysis problems. Included are the substructure planar analysis of rectangular panels with and without a hole, the static analysis of space mast, using NICE/SPAR and FORCE, and substructure analysis of plane rigid-jointed frames using FORCE. The computations are carried out on the Flex/32 MultiComputer using one to eighteen processors. The NICE/SPAR runstream samples are documented for the panel problem. For the substructure analysis of plane frames, a computer program is developed to demonstrate the effectiveness of a substructuring technique when FORCE is enforced. Ongoing research activities for an elasto-plastic stability analysis problem using FORCE, and stability analysis of the focus problem using NICE/SPAR are briefly summarized. Speedup curves for the panel, the mast, and the frame problems provide a basic understanding of the effectiveness of parallel computing procedures utilized or developed, within the domain of the parameters considered. Although the speedup curves obtained exhibit various levels of computational efficiency, they clearly demonstrate the excellent promise which parallel computing holds for the structural analysis problem. Source code is given for the elasto-plastic stability problem and the FORCE program.

Razzaq, Zia↗

Development of parallel DEM for the open source code MFIX

The paper presents the development of a parallel Discrete Element Method (DEM) solver for the open source code, Multiphase Flow with Interphase eXchange (MFIX) based on the domain decomposition method. The performance of the code was evaluated by simulating a bubbling fluidized bed with 2.5 million particles. The DEM solver shows strong scalability up to 256 processors with an efficiency of 81%. Further, to analyze weak scaling, the static height of the fluidized bed was increased to hold 5 and 10 million particles. The results show that global communication cost increases with problem size while the computational cost remains constant. Further, the effects of static bed height on the bubble hydrodynamics and mixing characteristics are analyzed.

42 ENGINEERING↗

A Parallel Domain Decomposition BEM Algorithm for Three Dimensional Exponentially Graded Elasticity

A parallel domain decomposition boundary integral algorithm for three-dimensional exponentially graded elasticity has been developed. As this subdomain algorithm allows the grading direction to vary in the structure, geometries arising from practical FGM applications can be handled. Moreover, the boundary integral algorithm scales well with the number of processors, also helping to alleviate the high computational cost of evaluating the Green's function. Numerical results for cylindrical geometries show excellent agreement with the new analytical solution deduced for axisymmetric plane strain states in a radially graded material.

97 MATHEMATICS AND COMPUTING↗

Efficacy of Code Optimization on Cache-based Processors

The current common wisdom in the U.S. is that the powerful, cost-effective supercomputers of tomorrow will be based on commodity (RISC) micro-processors with cache memories. Already, most distributed systems in the world use such hardware as building blocks. This shift away from vector supercomputers and towards cache-based systems has brought about a change in programming paradigm, even when ignoring issues of parallelism. Vector machines require inner-loop independence and regular, non-pathological memory strides (usually this means: non-power-of-two strides) to allow efficient vectorization of array operations. Cache-based systems require spatial and temporal locality of data, so that data once read from main memory and stored in high-speed cache memory is used optimally before being written back to main memory. This means that the most cache-friendly array operations are those that feature zero or unit stride, so that each unit of data read from main memory (a cache line) contains information for the next iteration in the loop. Moreover, loops ought to be 'fat', meaning that as many operations as possible are performed on cache data-provided instruction caches do not overflow and enough registers are available. If unit stride is not possible, for example because of some data dependency, then care must be taken to avoid pathological strides, just ads on vector computers. For cache-based systems the issues are more complex, due to the effects of associativity and of non-unit block (cache line) size. But there is more to the story. Most modern micro-processors are superscalar, which means that they can issue several (arithmetic) instructions per clock cycle, provided that there are enough independent instructions in the loop body. This is another argument for providing fat loop bodies. With these restrictions, it appears fairly straightforward to produce code that will run efficiently on any cache-based system. It can be argued that although some of the important computational algorithms employed at NASA Ames require different programming styles on vector machines and cache-based machines, respectively, neither architecture class appeared to be favored by particular algorithms in principle. Practice tells us that the situation is more complicated. This report presents observations and some analysis of performance tuning for cache-based systems. We point out several counterintuitive results that serve as a cautionary reminder that memory accesses are not the only factors that determine performance, and that within the class of cache-based systems, significant differences exist.

VanderWijngaart, Rob F.↗

A general CFD framework for fault-resilient simulations based on multi-resolution information fusion

We develop a general CFD framework for multi-resolution simulations to target multiscale problems but also resilience in exascale simulations, where faulty processors may lead to gappy, in space-time, simulated fields. We combine approximation theory and domain decomposition together with statistical learning techniques, e.g. coKriging, to estimate boundary conditions and minimize communications by performing independent parallel runs. To demonstrate this new simulation approach, we consider two benchmark problems. First, we solve the heat equation (a) on a small number of spatial “patches” distributed across the domain, simulated by finite differences at fine resolution and (b) on the entire domain simulated at very low resolution, thus fusing multi-resolution models to obtain the final answer. Second, we simulate the flow in a lid-driven cavity in an analogous fashion, by fusing finite difference solutions obtained with fine and low resolution assuming gappy data sets. We investigate the influence of various parameters for this framework, including the correlation kernel, the size of a buffer employed in estimating boundary conditions, the coarseness of the resolution of auxiliary data, and the communication frequency across different patches in fusing the information at different resolution levels. In addition to its robustness and resilience, the new framework can be employed to generalize previous multiscale approaches involving heterogeneous discretizations or even fundamentally different flow descriptions, e.g. in continuum-atomistic simulations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Exploring the use of novel programming models in land surface models

A wide range of programming models are currently under rapid development to meet the needs of application developers looking to work on more complex machines. These models fill a variety of roles. Some look to abstract supercomputer architecture, including both processors and memory, to present a strategy for portable performance across a wide range of machines. Others look to expose concurrency by explicitly constructing task-driven dependency graphs that allow a scheduler to find parallelism. Here we explore the implications for application codes of adopting two such programming models, Kokkos and Legion, one from each class of models. We specifically focus on the software design implications on refactoring existing applications, rather than the performance and performance tuning of these models. We identify a strategy for refactoring the Energy Exascale Earth System Model's Land Surface Model, an extremely complex code for climate applications, and prototype a series of mini-apps that explore the adoption of Kokkos and Legion. In doing this, we identify commonalities across the models, leading to a series of conclusions about application software design and refactoring for the adoption of novel programming models.

Coon, Ethan↗

Fast Parallel Computation Of Manipulator Inverse Dynamics

Method for fast parallel computation of inverse dynamics problem, essential for real-time dynamic control and simulation of robot manipulators, undergoing development. Enables exploitation of high degree of parallelism and, achievement of significant computational efficiency, while minimizing various communication and synchronization overheads as well as complexity of required computer architecture. Universal real-time robotic controller and simulator (URRCS) consists of internal host processor and several SIMD processors with ring topology. Architecture modular and expandable: more SIMD processors added to match size of problem. Operate asynchronously and in MIMD fashion.

Fijany, Amir↗

Reusable and Extensible High Level Data Distributions

This paper presents a reusable design of a data distribution framework for data parallel high performance applications. We are implementing the design in the context of the Chapel high productivity programming language. Distributions in Chapel are a means to express locality in systems composed of large numbers of processor and memory components connected by a network. Since distributions have a great effect on,the performance of applications, it is important that the distribution strategy can be chosen by a user. At the same time, high productivity concerns require that the user is shielded from error-prone, tedious details such as communication and synchronization. We propose an approach to distributions that enables the user to refine a language-provided distribution type and adjust it to optimize the performance of the application. Additionally, we conceal from the user low-level communication and synchronization details to increase productivity. To emphasize the generality of our distribution machinery, we present its abstract design in the form of a design pattern, which is independent of a concrete implementation. To illustrate the applicability of our distribution framework design, we outline the implementation of data distributions in terms of the Chapel language.

patterns↗

Systems and methods for rapid processing and storage of data

Systems and methods of building massively parallel computing systems using low power computing complexes in accordance with embodiments of the invention are disclosed. A massively parallel computing system in accordance with one embodiment of the invention includes at least one Solid State Blade configured to communicate via a high performance network fabric. In addition, each Solid State Blade includes a processor configured to communicate with a plurality of low power computing complexes interconnected by a router, and each low power computing complex includes at least one general processing core, an accelerator, an I/O interface, and cache memory and is configured to communicate with non-volatile solid state memory.

97 MATHEMATICS AND COMPUTING↗