Oprimización y paralelización de un algoritmo de sincronización mediante el uso de GPUs y la tecnología CUDA

En el presente documento se hablará acerca del desarrollo de un proyecto para la mejora de un programa de análisis de señales; con ese fin, se hará uso de técnicas de optimización del software y de tecnologías de aceleración, mediante el aprovechamiento del paralelismo del programa. Además se hará un análisis de acerca del uso de dos tecnologías basadas en diferentes paradigmas de programación paralela; una mediante múltiples hilos con memoria compartida y la otra mediante el uso de GPUs como dispositivos de coprocesamiento. This paper will talk about the development of a Project to improve a program that does signals analysis; to that end, it will make use of software optimization techniques and acceleration technologies by exploiting parallelism in the program. In Addition will be done an analysis on the use of two technologies based on two different paradigms; one using multiple threads with shared memory and the other using GPU as co-processing devices.

Aceleración de Algoritmos en Procesadores Multicore, GPGPU y FPGA

El avance en la potencia de cómputo en nuestros días viene dado por la paralelización del procesamiento, dadas las características que disponen las nuevas arquitecturas de hardware. Utilizar convenientemente este hardware impacta en la aceleración de los algoritmos en ejecución (programas). Sin embargo, convertir de forma adecuada el algoritmo en su forma paralela es complejo, y a su vez, esta forma, es específica para cada tipo de hardware paralelo. En la actualidad los procesadores de uso general más comunes son los multicore, procesadores paralelos, también denominados Symmetric Multi-Processors (SMP). Hoy en día es difícil hallar un procesador para computadoras de escritorio que no tengan algún tipo de paralelismo del caracterizado por los SMP, siendo la tendencia de desarrollo, que cada día nos encontremos con procesadores con mayor numero de cores disponibles. Por otro lado, los dispositivos de procesamiento de video (Graphics Processor Units - GPU), a su vez, han ido desarrollando su potencia de cómputo por medio de disponer de múltiples unidades de procesamiento dentro de su composición electrónica, a tal punto que en la actualidad no es difícil encontrar placas de GPU con capacidad de 200 a 400 hilos de procesamiento paralelo. Estos procesadores son muy veloces y específicos para la tarea que fueron desarrollados, principalmente el procesamiento de video. Sin embargo, como este tipo de procesadores tiene muchos puntos en común con el procesamiento científico, estos dispositivos han ido reorientándose con el nombre de General Processing Graphics Processor Unit (GPGPU). A diferencia de los procesadores SMP señalados anteriormente, las GPGPU no son de propósito general y tienen sus complicaciones para uso general debido al límite en la cantidad de memoria que cada placa puede disponer y al tipo de procesamiento paralelo que debe realizar para poder ser productiva su utilización. Los dispositivos de lógica programable, FPGA, son dispositivos capaces de realizar grandes cantidades de operaciones en paralelo, por lo que pueden ser usados para la implementación de algoritmos específicos, aprovechando el paralelismo que estas ofrecen. Su inconveniente viene derivado de la complejidad para la programación y el testing del algoritmo instanciado en el dispositivo. Ante esta diversidad de procesadores paralelos, el objetivo de nuestro trabajo está enfocado en analizar las características especificas que cada uno de estos tienen, y su impacto en la estructura de los algoritmos para que su utilización pueda obtener rendimientos de procesamiento acordes al número de recursos utilizados y combinarlos de forma tal que su complementación sea benéfica. Específicamente, partiendo desde las características del hardware, determinar las propiedades que el algoritmo paralelo debe tener para poder ser acelerado. Las características de los algoritmos paralelos determinará a su vez cuál de estos nuevos tipos de hardware son los mas adecuados para su instanciación. En particular serán tenidos en cuenta el nivel de dependencia de datos, la necesidad de realizar sincronizaciones durante el procesamiento paralelo, el tamaño de datos a procesar y la complejidad de la programación paralela en cada tipo de hardware.

Optimized CUDA-Based PDE Solver for Reaction Diffusion Systems on Arbitrary Surfaces

Partial differential equation (PDE) solvers are commonly employed to study and characterize the parameter space for reaction-diffusion (RD) systems while investigating biological pattern formation. Increasingly, biologists wish to perform such studies with arbitrary surfaces representing ‘real’ 3D geometries for better insights. In this paper, we present a highly optimized CUDA-based solver for RD equations on triangulated meshes in 3D. We demonstrate our solver using a chemotactic model that can be used to study snakeskin pigmentation, for example. We employ a finite element based approach to perform explicit Euler time integrations. We compare our approach to a naive GPU implementation and provide an in-depth performance analysis, demonstrating the significant speedup afforded by our optimizations. The optimization strategies that we exploit could be generalized to other mesh based processing applications with PDE simulations.

Multi-threaded parallel numerical modelling of femtosecond pulse propagation in laser machining

Femtosecond laser microfabrication has emerged over the last decade as a 3D flexible technology in photonics. Numerical simulations provide an important insight into spatial and temporal beam and pulse shaping during the course of extremely intricate nonlinear propagation (see e.g. [1,2]). Electromagnetics of such propagation is typically described in the form of the generalized Non-Linear Schrdinger Equation (NLSE) coupled with Drude model for plasma [3]. In this paper we consider a multi-threaded parallel numerical solution for a specific model which describes femtosecond laser pulse propagation in transparent media [4, 5]. However our approach can be extended to similar models. The numerical code is implemented in NVIDIA Graphics Processing Unit (GPU) which provides an effitient hardware platform for multi-threded computing. We compare the performance of the described below parallel code implementated for GPU using CUDA programming interface [3] with a serial CPU version used in our previous papers [4,5]. © 2011 IEEE.

Analisi della propagazione elettrica nel tessuto cardiaco mediante simulazione numerica implementata su GPU

Il segnale elettrico si propaga nel tessuto cardiaco attraverso gap-junctions che si trovano tra i miociti cardiaci e in ciascuno di essi si avvia un processo chiamato potenziale d'azione (PA). In questa tesi prenderò in considerazione il modello Luo-Rudy 1991 e il difetto oggetto di studio sono le Early Afterdepolarizations (EADs). Si analizzerà la propagazione del potenziale d’azione in un cavo di 300 cellule. Dopo alcune simulazioni preliminari è emersa l’utilità di trovare una soluzione che permettesse di ridurre i tempi di calcolo, il modello è stato quindi implementato in CUDA. Il lavoro è stato sviluppato nei seguenti step: 1) l’impiego dell’ambiente di calcolo MATLAB per implementare il modello, descrivendo ogni cellula attraverso il modello Luo-Rudy 1991 e l’interazione elettrica inter-cellulare, considerando un cavo di 300 cellule; 2) individuazione dei parametri che, adeguatamente modificati, sono in grado di indurre EADs a livello single cell; 3) implementazione del modello in CUDA, creando uno strumento che potrà essere utilizzato per aumentare notevolmente il numero delle simulazioni nell’unità di tempo; 4) messa a punto di un criterio per valutare in modo conciso la bontà (safety factor) della relazione source-sink. L’utilità di un simile criterio è quella di valutare, sia nel caso di propagazione di AP che in quello di eventuale propagazione di EADs, la propensione alla propagazione in un tessuto. Il primo capitolo descriverà il potenziale d’azione, il modello usato e la teoria del cavo. Il secondo capitolo discuterà l’implementazione del modello usato, descriverà CUDA e come il modello sia stato implementato. Il terzo capitolo riguarderà i primi risultati ottenuti dalle simulazioni e come la variazione dei parametri influisce sulla forma delle EADs. L’ultimo capitolo approfondirà i requisiti necessari per far avvenire una propagazione in un cavo.

Computação de alto desempenho aplicada a modelagem numérica de fenômenos atmosféricos

A constante evolução da tecnologia disponibilizou, atualmente, ferramentas computacionais que eram apenas expectativas há 10 anos atrás. O aumento do potencial computacional aplicado a modelos numéricos que simulam a atmosfera permitiu ampliar o estudo de fenômenos atmosféricos, através do uso de ferramentas de computação de alto desempenho. O trabalho propôs o desenvolvimento de algoritmos com base em arquiteturas SIMT e aplicação de técnicas de paralelismo com uso da ferramenta OpenACC para processamento de dados de previsão numérica do modelo Weather Research and Forecast. Esta proposta tem forte conotação interdisciplinar, buscando a interação entre as áreas de modelagem atmosférica e computação científica. Foram testadas a influência da computação do cálculo de microfísica de nuvens na degradação temporal do modelo. Como a entrada de dados para execução na GPU não era suficientemente grande, o tempo necessário para transferir dados da CPU para a GPU foi maior do que a execução da computação na CPU. Outro fator determinante foi a adição de código CUDA dentro de um contexto MPI, causando assim condições de disputa de recursos entre os processadores, mais uma vez degradando o tempo de execução. A proposta do uso de diretivas para aplicar computação de alto desempenho em uma estrutura CUDA parece muito promissora, mas ainda precisa ser utilizada com muita cautela a fim de produzir bons resultados. A construção de um híbrido MPI + CUDA foi testada, mas os resultados não foram conclusivos.

Balancing Energy and Performance in Dense Linear System Solvers for Hybrid ARM+GPU platforms

The high performance computing community has traditionally focused uniquely on the reduction of execution time, though in the last years, the optimization of energy consumption has become a main issue. A reduction of energy usage without a degradation of performance requires the adoption of energy-efficient hardware platforms accompanied by the development of energy-aware algorithms and computational kernels. The solution of linear systems is a key operation for many scientific and engineering problems. Its relevance has motivated an important amount of work, and consequently, it is possible to find high performance solvers for a wide variety of hardware platforms. In this work, we aim to develop a high performance and energy-efficient linear system solver. In particular, we develop two solvers for a low-power CPU-GPU platform, the NVIDIA Jetson TK1. These solvers implement the Gauss-Huard algorithm yielding an efficient usage of the target hardware as well as an efficient memory access. The experimental evaluation shows that the novel proposal reports important savings in both time and energy-consumption when compared with the state-of-the-art solvers of the platform.

Random path sampling applied to vector models of magnetism

The recently reported Monte Carlo Random Path Sampling method (RPS) is here improved and its application is expanded to the study of the 2D and 3D Ising and discrete Heisenberg models. The methodology was implemented to allow use in both CPU-based high-performance computing infrastructures (C/MPI) and GPU-based (CUDA) parallel computation, with significant computational performance gains. Convergence is discussed, both in terms of free energy and magnetization dependence on field/temperature. From the calculated magnetization-energy joint density of states, fast calculations of field and temperature dependent thermodynamic properties are performed, including the effects of anisotropy on coercivity, and the magnetocaloric effect. The emergence of first-order magneto-volume transitions in the compressible Ising model is interpreted using the Landau theory of phase transitions. Using metallic Gadolinium as a real-world example, the possibility of using RPS as a tool for computational magnetic materials design is discussed. Experimental magnetic and structural properties of a Gadolinium single crystal are compared to RPS-based calculations using microscopic parameters obtained from Density Functional Theory.

Procesamiento de imágenes cerebrales en GPU

Abstract: As time has passed, the general purpose programming paradigm has evolved, producing different hardware architectures whose characteristics differ widely. In this work, we are going to demonstrate, through different applications belonging to the field of Image Processing, the existing difference between three Nvidia hardware platforms: two of them belong to the GeForce graphics cards series, the GTX 480 and the GTX 980 and one of the low consumption platforms which purpose is to allow the execution of embedded applications as well as providing an extreme efficiency: the Jetson TK1. With respect to the test applications we will use five examples from Nvidia CUDA Samples. These applications are directly related to Image Processing, as the algorithms they use are similar to those from the field of medical image registration. After the tests, it will be proven that GTX 980 is both the device with the highest computational power and the one that has greater consumption, it will be seen that Jetson TK1 is the most efficient platform, it will be shown that GTX 480 produces more heat than the others and we will learn other effects produced by the existing difference between the architecture of the devices.

A three-dimensional representation method for noisy point clouds based on growing self-organizing maps accelerated on GPUs

The research described in this thesis was motivated by the need of a robust model capable of representing 3D data obtained with 3D sensors, which are inherently noisy. In addition, time constraints have to be considered as these sensors are capable of providing a 3D data stream in real time. This thesis proposed the use of Self-Organizing Maps (SOMs) as a 3D representation model. In particular, we proposed the use of the Growing Neural Gas (GNG) network, which has been successfully used for clustering, pattern recognition and topology representation of multi-dimensional data. Until now, Self-Organizing Maps have been primarily computed offline and their application in 3D data has mainly focused on free noise models, without considering time constraints. It is proposed a hardware implementation leveraging the computing power of modern GPUs, which takes advantage of a new paradigm coined as General-Purpose Computing on Graphics Processing Units (GPGPU). The proposed methods were applied to different problem and applications in the area of computer vision such as the recognition and localization of objects, visual surveillance or 3D reconstruction.

INVESTIGATIONS INTO MECHANISMS UNDERLYING EXTREME WAVE FORMATIONS AND COMPUTATIONALLY INTENSIVE SIMULATIONS

Various mechanisms have been proposed to explain extreme waves or rogue waves in an oceanic environment including directional focusing, dispersive focusing, wave-current interaction, and nonlinear modulational instability. The Benjamin-Feir instability (nonlinear modulational instability), however, is considered to be one of the primary mechanisms for rogue-wave occurrence. The nonlinear Schrodinger equation is a well-established approximate model based on the same assumptions as required for the derivation of the Benjamin-Feir theory. Solutions of the nonlinear Schrodinger equation, including new rogue-wave type solutions are presented in the author's dissertation work. The solutions are obtained by using a predictive eigenvalue map based predictor-corrector procedure developed by the author. Features of the predictive map are explored and the influences of certain parameter variations are investigated. The solutions are rescaled to match the length scales of waves generated in a wave tank. Based on the information provided by the map and the details of physical scaling, a framework is developed that can serve as a basis for experimental investigations into a variety of extreme waves as well localizations in wave fields. To derive further fundamental insights into the complexity of extreme wave conditions, Smoothed Particle Hydrodynamics (SPH) simulations are carried out on an advanced Graphic Processing Unit (GPU) based parallel computational platform. Free surface gravity wave simulations have successfully characterized water-wave dispersion in the SPH model while demonstrating extreme energy focusing and wave growth in both linear and nonlinear regimes. A virtual wave tank is simulated wherein wave motions can be excited from either side. Focusing of several wave trains and isolated waves has been simulated. With properly chosen parameters, dispersion effects are observed causing a chirped wave train to focus and exhibit growth. By using the insights derived from the study of the nonlinear Schrodinger equation, modulational instability or self-focusing has been induced in a numerical wave tank and studied through several numerical simulations. Due to the inherent dissipative nature of SPH models, simulating persistent progressive waves can be problematic. This issue has been addressed and an observation-based solution has been provided. The efficacy of SPH in modeling wave focusing can be critical to further our understanding and predicting extreme wave phenomena through simulations. A deeper understanding of the mechanisms underlying extreme energy localization phenomena can help facilitate energy harnessing and serve as a basis to predict and mitigate the impact of energy focusing.

Visualization of flow fields in the web platform

Visualization of vector fields plays an important role in research activities nowadays -- Web applications allow a fast, multi-platform and multi-device access to data, which results in the need of optimized applications to be implemented in both high-performance and low-performance devices -- Point trajectory calculation procedures usually perform repeated calculations due to the fact that several points might lie over the same trajectory -- This paper presents a new methodology to calculate point trajectories over highly-dense and uniformly-distributed grid of points in which the trajectories are forced to lie over the points in the grid -- Its advantages rely on a highly parallel computing architecture implementation and in the reduction of the computational effort to calculate the stream paths since unnecessary calculations are avoided, reusing data through iterations -- As case study, the visualization of oceanic currents through in the web platform is presented and analyzed, using WebGL as the parallel computing architecture and the rendering Application Programming Interface

Improvements in Hardware Transactional Memory for GPU Architectures

In the multi-core CPU world, transactional memory (TM)has emerged as an alternative to lock-based programming for thread synchronization. Recent research proposes the use of TM in GPU architectures, where a high number of computing threads, organized in SIMT fashion, requires an effective synchronization method. In contrast to CPUs, GPUs offer two memory spaces: global memory and local memory. The local memory space serves as a shared scratch-pad for a subset of the computing threads, and it is used by programmers to speed-up their applications thanks to its low latency. Prior work from the authors proposed a lightweight hardware TM (HTM) support based in the local memory, modifying the SIMT execution model and adding a conflict detection mechanism. An efficient implementation of these features is key in order to provide an effective synchronization mechanism at the local memory level. After a quick description of the main features of our HTM design for GPU local memory, in this work we gather together a number of proposals designed with the aim of improving those mechanisms with high impact on performance. Firstly, the SIMT execution model is modified to increase the parallelism of the application when transactions must be serialized in order to make forward progress. Secondly, the conflict detection mechanism is optimized depending on application characteristics, such us the read/write sets, the probability of conflict between transactions and the existence of read-only transactions. As these features can be present in hardware simultaneously, it is a task of the compiler and runtime to determine which ones are more important for a given application. This work includes a discussion on the analysis to be done in order to choose the best configuration solution.

HPC Accelerators with 3D Memory

After a decade evolving in the High Performance Computing arena, GPU-equipped supercomputers have con- quered the top500 and green500 lists, providing us unprecedented levels of computational power and memory bandwidth. This year, major vendors have introduced new accelerators based on 3D memory, like Xeon Phi Knights Landing by Intel and Pascal architecture by Nvidia. This paper reviews hardware features of those new HPC accelerators and unveils potential performance for scientific applications, with an emphasis on Hybrid Memory Cube (HMC) and High Bandwidth Memory (HBM) used by commercial products according to roadmaps already announced.