A Comparison of Programming Models for Multiprocessors with Explicitly Managed Memory Hierarchies

On multiprocessors with explicitly managed memory hierarchies (EMM), software has the responsibility of moving data in and out of fast local memories. This task can be complex and error-prone even for expert programmers. Before we can allow compilers to handle the complexity for us, we must identify the abstractions that are general enough to allow us to write applications with reasonable effort, yet speci?c enough to exploit the vast on-chip memory bandwidth of EMM multi-processors. To this end, we compare two programming models against hand-tuned codes on the STI Cell, paying attention to programmability and performance. The ?rst programming model, Sequoia, abstracts the memory hierarchy as private address spaces, each corresponding to a parallel task. The second, Cellgen, is a new framework which provides OpenMP-like semantics and the abstraction of a shared address spaces divided into private and shared data. We compare three applications programmed using these models against their hand-optimized counterparts in terms of abstractions, programming complexity, and performance.

BDDT: Block-level Dynamic Dependence Analysis for Task-Based Parallelism

We present BDDT, a task-parallel runtime system that dynamically discovers and resolves dependencies among parallel tasks. BDDT allows the programmer to specify detailed task footprints on any memory address range, multidimensional array tile or dynamic region. BDDT uses a block-based dependence analysis with arbitrary granularity. The analysis is applicable to existing C programs without having to restructure object or array allocation, and provides flexibility in array layouts and tile dimensions.
We evaluate BDDT using a representative set of benchmarks, and we compare it to SMPSs (the equivalent runtime system in StarSs) and OpenMP. BDDT performs comparable to or better than SMPSs and is able to cope with task granularity as much as one order of magnitude finer than SMPSs. Compared to OpenMP, BDDT performs up to 3.9× better for benchmarks that benefit from dynamic dependence analysis. BDDT provides additional data annotations to bypass dependence analysis. Using these annotations, BDDT outperforms OpenMP also in benchmarks where dependence analysis does not discover additional parallelism, thanks to a more efficient implementation of the runtime system.

TProf: An energy profiler for task-parallel programs

We present TProf, an energy profiling tool for OpenMP-like task-parallel programs. To compute the energy consumed by each task in a parallel application, TProf dynamically traces the parallel execution and uses a novel technique to estimate the per-task energy consumption. To achieve this estimation, TProf apportions the total processor energy among cores and overcomes the limitation of current works which would otherwise make parallel accounting impossible to achieve. We demonstrate the value of TProf by characterizing a set of task parallel programs, where we find that data locality, memory access patterns and task working sets are responsible for significant variance in energy consumption between seemingly homogeneous tasks. In addition, we identify opportunities for fine-grain energy optimization by applying per-task Dynamic Voltage and Frequency Scaling (DVFS).

Runtime Support for Adaptive Power Capping on Heterogeneous SoCs

Power capping is a fundamental method for reducing the energy consumption of a wide range of modern computing environments, ranging from mobile embedded systems to datacentres. Unfortunately, maximising performance and system efficiency under static power caps remains challenging, while maximising performance under dynamic power caps has been largely unexplored. We present an adaptive power capping method that reduces the power consumption and maximizes the performance of heterogeneous SoCs for mobile and server platforms. Our technique combines power capping with coordinated DVFS, data partitioning and core allocations on a heterogeneous SoC with ARM processors and FPGA resources. We design our framework as a run-time system based on OpenMP and OpenCL to utilise the heterogeneous resources. We evaluate it through five data-parallel benchmarks on the Xilinx SoC which allows fully voltage and frequency control. Our experiments show a significant performance boost of 30% under dynamic power caps with concurrent execution on ARM and FPGA, compared to a naive separate approach.

Arithmetic bit recycling data compression

La compression des données est la technique informatique qui vise à réduire la taille de l’information pour minimiser l’espace de stockage nécessaire et accélérer la transmission des données dans les réseaux à bande passante limitée. Plusieurs techniques de compression telles que LZ77 et ses variantes souffrent d’un problème que nous appelons la redondance causée par la multiplicité d’encodages. La multiplicité d’encodages (ME) signifie que les données sources peuvent être encodées de différentes manières. Dans son cas le plus simple, ME se produit lorsqu’une technique de compression a la possibilité, au cours du processus d’encodage, de coder un symbole de différentes manières. La technique de compression par recyclage de bits a été introduite par D. Dubé et V. Beaudoin pour minimiser la redondance causée par ME. Des variantes de recyclage de bits ont été appliquées à LZ77 et les résultats expérimentaux obtenus conduisent à une meilleure compression (une réduction d’environ 9% de la taille des fichiers qui ont été compressés par Gzip en exploitant ME). Dubé et Beaudoin ont souligné que leur technique pourrait ne pas minimiser parfaitement la redondance causée par ME, car elle est construite sur la base du codage de Huffman qui n’a pas la capacité de traiter des mots de code (codewords) de longueurs fractionnaires, c’est-à-dire qu’elle permet de générer des mots de code de longueurs intégrales. En outre, le recyclage de bits s’appuie sur le codage de Huffman (HuBR) qui impose des contraintes supplémentaires pour éviter certaines situations qui diminuent sa performance. Contrairement aux codes de Huffman, le codage arithmétique (AC) peut manipuler des mots de code de longueurs fractionnaires. De plus, durant ces dernières décennies, les codes arithmétiques ont attiré plusieurs chercheurs vu qu’ils sont plus puissants et plus souples que les codes de Huffman. Par conséquent, ce travail vise à adapter le recyclage des bits pour les codes arithmétiques afin d’améliorer l’efficacité du codage et sa flexibilité. Nous avons abordé ce problème à travers nos quatre contributions (publiées). Ces contributions sont présentées dans cette thèse et peuvent être résumées comme suit. Premièrement, nous proposons une nouvelle technique utilisée pour adapter le recyclage de bits qui s’appuie sur les codes de Huffman (HuBR) au codage arithmétique. Cette technique est nommée recyclage de bits basé sur les codes arithmétiques (ACBR). Elle décrit le cadriciel et les principes de l’adaptation du HuBR à l’ACBR. Nous présentons aussi l’analyse théorique nécessaire pour estimer la redondance qui peut être réduite à l’aide de HuBR et ACBR pour les applications qui souffrent de ME. Cette analyse démontre que ACBR réalise un recyclage parfait dans tous les cas, tandis que HuBR ne réalise de telles performances que dans des cas très spécifiques. Deuxièmement, le problème de la technique ACBR précitée, c’est qu’elle requiert des calculs à précision arbitraire. Cela nécessite des ressources illimitées (ou infinies). Afin de bénéficier de cette dernière, nous proposons une nouvelle version à précision finie. Ladite technique devienne ainsi efficace et applicable sur les ordinateurs avec les registres classiques de taille fixe et peut être facilement interfacée avec les applications qui souffrent de ME. Troisièmement, nous proposons l’utilisation de HuBR et ACBR comme un moyen pour réduire la redondance afin d’obtenir un code binaire variable à fixe. Nous avons prouvé théoriquement et expérimentalement que les deux techniques permettent d’obtenir une amélioration significative (moins de redondance). À cet égard, ACBR surpasse HuBR et fournit une classe plus étendue des sources binaires qui pouvant bénéficier d’un dictionnaire pluriellement analysable. En outre, nous montrons qu’ACBR est plus souple que HuBR dans la pratique. Quatrièmement, nous utilisons HuBR pour réduire la redondance des codes équilibrés générés par l’algorithme de Knuth. Afin de comparer les performances de HuBR et ACBR, les résultats théoriques correspondants de HuBR et d’ACBR sont présentés. Les résultats montrent que les deux techniques réalisent presque la même réduction de redondance sur les codes équilibrés générés par l’algorithme de Knuth.

Design and Qualitative/Quantitative Analysis of Multi-Agent Spatial Simulation Library

Thesis (Master's)--University of Washington, 2012

A framework for the development of parallel and distributed real-time embedded systems

Embedded real-time applications increasingly present high computation requirements, which need to be completed within specific deadlines, but that present highly variable patterns, depending on the set of data available in a determined instant. The current trend to provide parallel processing in the embedded domain allows providing higher processing power; however, it does not address the variability in the processing pattern. Dimensioning each device for its worst-case scenario implies lower average utilization, and increased available, but unusable, processing in the overall system. A solution for this problem is to extend the parallel execution of the applications, allowing networked nodes to distribute the workload, on peak situations, to neighbour nodes. In this context, this report proposes a framework to develop parallel and distributed real-time embedded applications, transparently using OpenMP and Message Passing Interface (MPI), within a programming model based on OpenMP. The technical report also devises an integrated timing model, which enables the structured reasoning on the timing behaviour of these hybrid architectures.

Evaluating the state of the art of parallel programming systems

This paper describes our plans to evaluate the present state of affairs concerning parallel programming and its systems. Three subprojects are proposed: a survey among programmers and scientists, a comparison of parallel programming systems using a standard set of test programs, and a wiki resource for the parallel programming community - the Parawiki. We would like to invite you to participate and turn these subprojects into true community efforts.

Observations on the Publicity and Usage of Parallel Programming Systems and Languages: A Survey Approach

In this publication, we report on an online survey that was carried out among parallel programmers. More than 250 people worldwide have submitted answers to our questions, and their responses are analyzed here. Although not statistically sound, the data we provide give useful insights about which parallel programming systems and languages are known and in actual use. For instance, the collected data indicate that for our survey group MPI and (to a lesser extent) C are the most widely used parallel programming system and language, respectively.

INFO2009 2012-13 Resource Group 10 Facebook Privacy

Security and Privacy Online - this is the resource and flyer created for INFO2009 which provides an interactive web presentation to make the general public aware of the dangers of using the internet unsafely, and how they can protect themselves.

Nested parallelism for multi-core HPC systems using Java

Since its introduction in 1993, the Message Passing Interface (MPI) has become a de facto standard for writing High Performance Computing (HPC) applications on clusters and Massively Parallel Processors (MPPs). The recent emergence of multi-core processor systems presents a new challenge for established parallel programming paradigms, including those based on MPI. This paper presents a new Java messaging system called MPJ Express. Using this system, we exploit multiple levels of parallelism - messaging and threading - to improve application performance on multi-core processors. We refer to our approach as nested parallelism. This MPI-like Java library can support nested parallelism by using Java or Java OpenMP (JOMP) threads within an MPJ Express process. Practicality of this approach is assessed by porting to Java a massively parallel structure formation code from Cosmology called Gadget-2. We introduce nested parallelism in the Java version of the simulation code and report good speed-ups. To the best of our knowledge it is the first time this kind of hybrid parallelism is demonstrated in a high performance Java application. (C) 2009 Elsevier Inc. All rights reserved.

Análise de desempenho da rede neural artificial do tipo multilayer perceptron na era multicore

Artificial neural networks are usually applied to solve complex problems. In problems with more complexity, by increasing the number of layers and neurons, it is possible to achieve greater functional efficiency. Nevertheless, this leads to a greater computational effort. The response time is an important factor in the decision to use neural networks in some systems. Many argue that the computational cost is higher in the training period. However, this phase is held only once. Once the network trained, it is necessary to use the existing computational resources efficiently. In the multicore era, the problem boils down to efficient use of all available processing cores. However, it is necessary to consider the overhead of parallel computing. In this sense, this paper proposes a modular structure that proved to be more suitable for parallel implementations. It is proposed to parallelize the feedforward process of an RNA-type MLP, implemented with OpenMP on a shared memory computer architecture. The research consistes on testing and analizing execution times. Speedup, efficiency and parallel scalability are analyzed. In the proposed approach, by reducing the number of connections between remote neurons, the response time of the network decreases and, consequently, so does the total execution time. The time required for communication and synchronization is directly linked to the number of remote neurons in the network, and so it is necessary to investigate which one is the best distribution of remote connections

Escalabilidade Paralela de um Algoritmo de Migração Reversa no Tempo (RTM) Pré-empilhamento

The seismic method is of extreme importance in geophysics. Mainly associated with oil exploration, this line of research focuses most of all investment in this area. The acquisition, processing and interpretation of seismic data are the parts that instantiate a seismic study. Seismic processing in particular is focused on the imaging that represents the geological structures in subsurface. Seismic processing has evolved significantly in recent decades due to the demands of the oil industry, and also due to the technological advances of hardware that achieved higher storage and digital information processing capabilities, which enabled the development of more sophisticated processing algorithms such as the ones that use of parallel architectures. One of the most important steps in seismic processing is imaging. Migration of seismic data is one of the techniques used for imaging, with the goal of obtaining a seismic section image that represents the geological structures the most accurately and faithfully as possible. The result of migration is a 2D or 3D image which it is possible to identify faults and salt domes among other structures of interest, such as potential hydrocarbon reservoirs. However, a migration fulfilled with quality and accuracy may be a long time consuming process, due to the mathematical algorithm heuristics and the extensive amount of data inputs and outputs involved in this process, which may take days, weeks and even months of uninterrupted execution on the supercomputers, representing large computational and financial costs, that could derail the implementation of these methods. Aiming at performance improvement, this work conducted the core parallelization of a Reverse Time Migration (RTM) algorithm, using the parallel programming model Open Multi-Processing (OpenMP), due to the large computational effort required by this migration technique. Furthermore, analyzes such as speedup, efficiency were performed, and ultimately, the identification of the algorithmic scalability degree with respect to the technological advancement expected by future processors

C programs for solving the time-dependent Gross-Pitaevskii equation in a fully anisotropic trap

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Low-energy electron collisions with glycine

We report cross sections for elastic electron scattering by gas phase glycine (neutral form), obtained with the Schwinger multichannel method. The present results are the first obtained with a new implementation that combines parallelization with OpenMP directives and pseudopotentials. The position of the well known pi* shape resonance ranged from 2.3 eV to 2.8 eV depending on the polarization model and conformer. For the most stable isomer, the present result (2.4 eV) is in fair agreement with electron transmission spectroscopy assignments (1.93 +/- 0.05 eV) and available calculations. Our results also point out a shape resonance around 9.5 eV in the A' symmetry that would be weakly coupled to vibrations of the hydroxyl group. Since electron attachment to a broad and lower lying sigma* orbital located on the OH bond has been suggested the underlying mechanism leading to dissociative electron attachment at low energies, we sought for a shape resonance around similar to 4 eV. Though we obtained cross sections with the target molecule at the equilibrium geometry and with stretched OH bond lengths, least-squares fits to the calculated eigenphase sums did not point out signatures of this anion state (though, in principle, it could be hidden in the large background). The low energy (similar to 1 eV) integral cross section strongly scales as the bond length is stretched, and this could indicate a virtual state pole, since dipole supported bound states are not expected at the geometries addressed here. (C) 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.3687345]