Capabilities Of GPUs Research Articles

A local time stepping algorithm for GPU-accelerated 2D shallow water models

Abstract In the simulation of flooding events, mesh refinement is often required to capture local bathymetric features and/or to detail areas of interest; however, if an explicit finite volume scheme is adopted, the presence of small cells in the domain can restrict the allowable time step due to the stability condition, thus reducing the computational efficiency. With the aim of overcoming this problem, the paper proposes the application of a Local Time Stepping (LTS) strategy to a GPU-accelerated 2D shallow water numerical model able to handle non-uniform structured meshes. The algorithm is specifically designed to exploit the computational capability of GPUs, minimizing the overheads associated with the LTS implementation. The results of theoretical and field-scale test cases show that the LTS model guarantees appreciable reductions in the execution time compared to the traditional Global Time Stepping strategy, without compromising the solution accuracy.

Performance studies of GooFit on GPUs vs RooFit on CPUs while estimating the statistical significance of a new physical signal

In order to test the computing capabilities of GPUs with respect to traditional CPU cores a high-statistics toy Monte Carlo technique has been implemented both in ROOT/RooFit and GooFit frameworks with the purpose to estimate the statistical significance of the structure observed by CMS close to the kinematical boundary of the J/ψϕ invariant mass in the three-body decay B+ → J/ψϕK+. GooFit is a data analysis open tool under development that interfaces ROOT/RooFit to CUDA platform on nVidia GPU. The optimized GooFit application running on GPUs hosted by servers in the Bari Tier2 provides striking speed-up performances with respect to the RooFit application parallelised on multiple CPUs by means of PROOF-Lite tool. The considerable resulting speed-up, evident when comparing concurrent GooFit processes allowed by CUDA Multi Process Service and a RooFit/PROOF-Lite process with multiple CPU workers, is presented and discussed in detail. By means of GooFit it has also been possible to explore the behaviour of a likelihood ratio test statistic in different situations in which the Wilks Theorem may or may not apply because its regularity conditions are not satisfied.

A distributed in-memory key-value store system on heterogeneous CPU–GPU cluster

In-memory key-value stores play a critical role in many data-intensive applications to provide high-throughput and low latency data accesses. In-memory key-value stores have several unique properties that include (1) data-intensive operations demanding high memory bandwidth for fast data accesses, (2) high data parallelism and simple computing operations demanding many slim parallel computing units, and (3) a large working set. However, our experiments show that homogeneous multicore CPU systems are increasingly mismatched to the special properties of key-value stores because they do not provide massive data parallelism and high memory bandwidth; the powerful but the limited number of computing cores does not satisfy the demand of the unique data processing task; and the cache hierarchy may not well benefit to the large working set. In this paper, we present the design and implementation of Mega-KV, a distributed in-memory key-value store system on a heterogeneous CPU–GPU cluster. Effectively utilizing the high memory bandwidth and latency hiding capability of GPUs, Mega-KV provides fast data accesses and significantly boosts overall performance and energy efficiency over the homogeneous CPU architectures. Mega-KV shows excellent scalability and processes up to 623-million key-value operations per second on a cluster installed with eight CPUs and eight GPUs, while delivering an efficiency of up to 299-thousand operations per Watt (KOPS/W).

GPU-based volume reconstruction for freehand 3D ultrasound imaging.

Volume reconstruction plays an important role in improving image quality for freehand three-dimensional (3D) ultrasound imaging. The kernel regression provides an effective method for volume reconstruction in 3D ultrasound imaging, but it requires heavily computational time-cost. In this paper, a programmable graphic-processor-unit-(GPU) based fast kernel regression method is proposed for freehand 3D ultrasound volume reconstruction. The most significant aspect of our method is the adopting of powerful data-parallel computing capability of GPU to improve the overall efficiency. To produce higher image quality, the results of the kernel regression with various parameter settings is deeply investigated under the help of the fast implementation of the algorithm. Experimental results demonstrate that the computational performance of the proposed GPU-based method can be over 200 times faster than that on CPU. Better image quality for speckle reduction and details preservation can be obtained with the parameter setting of kernel window size of 5×5×5 and kernel bandwidth of 1.0.

Interactive GPU-based generation of solvent-excluded surfaces

The solvent-excluded surface (SES) is a popular molecular representation that gives the boundary of the molecular volume with respect to a specific solvent. SESs depict which areas of a molecule are accessible by a specific solvent, which is represented as a spherical probe. Despite the popularity of SESs, their generation is still a compute-intensive process, which is often performed in a preprocessing stage prior to the actual rendering (except for small models). For dynamic data or varying probe radii, however, such a preprocessing is not feasible as it prevents interactive visual analysis. Thus, we present a novel approach for the on-the-fly generation of SESs, a highly parallelizable, grid-based algorithm where the SES is rendered using ray-marching. By exploiting modern GPUs, we are able to rapidly generate SESs directly within the mapping stage of the visualization pipeline. Our algorithm can be applied to large time-varying molecules and is scalable, as it can progressively refine the SES if GPU capabilities are insufficient. In this paper, we show how our algorithm is realized and how smooth transitions are achieved during progressive refinement. We further show visual results obtained from real-world data and discuss the performance obtained, which improves upon previous techniques in both the size of the molecules that can be handled and the resulting frame rate.

Statistical significance estimation of a signal within the GooFit framework on GPUs

In order to test the computing capabilities of GPUs with respect to traditional CPU cores a high-statistics toy Monte Carlo technique has been implemented both in ROOT/RooFit and GooFit frameworks with the purpose to estimate the statistical significance of the structure observed by CMS close to the kinematical boundary of the J/ψϕ invariant mass in the three-body decay B + → J/ψϕK + . GooFit is a data analysis open tool under development that interfaces ROOT/RooFit to CUDA platform on nVidia GPU. The optimized GooFit application running on GPUs hosted by servers in the Bari Tier2 provides striking speed-up performances with respect to the RooFit application parallelised on multiple CPUs by means of PROOF-Lite tool. The considerable resulting speed-up, evident when comparing concurrent GooFit processes allowed by CUDA Multi Process Service and a RooFit/PROOF-Lite process with multiple CPU workers, is presented and discussed in detail. By means of GooFit it has also been possible to explore the behaviour of a likelihood ratio test statistic in different situations in which the Wilks Theorem may or may not apply because its regularity conditions are not satisfied.

Accelerating the phylogenetic parsimony function on heterogeneous systems

SummaryThe availability of heterogeneous CPU+GPU systems has opened the door to new opportunities for the development of parallel solutions to tackle complex biological problems. The reconstruction of evolutionary histories among species represents a grand computational challenge, which can be addressed by exploiting this kind of hardware designs. In this research, we study the application of heterogeneous computing with OpenCL to accelerate one of the most well‐known objective functions for inferring phylogenies, the phylogenetic parsimony function. For this purpose, we undertake the design of CPU and GPU kernel implementations of this relevant function, proposing a heterogeneous CPU+GPU multidevice approach that distributes multiple parsimony evaluations among processing devices. Experiments on 6 real nucleotide data sets and comparisons with other parallel implementations give account of the benefits of the proposal in this paper, obtaining significant parallel results by combining CPU and GPU capabilities in accordance with the characteristics of the input data.

A non-uniform efficient grid type for GPU-parallel Shallow Water Equations models

A GPU-parallel numerical model for the solution of the 2D Shallow Water Equations, based on a novel type of grid called Block-Uniform Quadtree (BUQ), is presented. BUQ grids are based on a data structure which allows to exploit the computational capability of GPUs with minimum overheads, while discretizing the domain with non-uniform resolution. Different cases have been simulated in order to assess the efficiency of the BUQ grids. Theoretical and laboratory tests demonstrate that speed-ups of up to one order of magnitude can be achieved in comparison with uniform Cartesian grids. In the simulation of a hypothetical flood event induced by a levee breach in a real 83 km long river reach, with maximum resolution of 5 m, a ratio of physical to computational time of about 12 was obtained, opening scenarios of quasi real-time 2D simulations in large domains, still retaining a high resolution where necessary.

GPUs for statistical data analysis in HEP: a performance study of GooFit on GPUs vs. RooFit on CPUs

In order to test the computing capabilities of GPUs with respect to traditional CPU cores a high-statistics toy Monte Carlo technique has been implemented both in ROOT/RooFit and GooFit frameworks with the purpose to estimate the statistical significance of the structure observed by CMS close to the kinematical boundary of the Jψϕ invariant mass in the three-body decay B+→JψϕK+. GooFit is a data analysis open tool under development that interfaces ROOT/RooFit to CUDA platform on nVidia GPU. The optimized GooFit application running on GPUs hosted by servers in the Bari Tier2 provides striking speed-up performances with respect to the RooFit application parallelised on multiple CPUs by means of PROOF-Lite tool. The considerably resulting speed-up, while comparing concurrent GooFit processes allowed by CUDA Multi Process Service and a RooFit/PROOF-Lite process with multiple CPU workers, is presented and discussed in detail. By means of GooFit it has also been possible to explore the behaviour of a likelihood ratio test statistic in different situations in which the Wilks Theorem may apply or does not apply because its regularity conditions are not satisfied.

Performance of Сellular Automata-based Stream Ciphers in GPU Implementation

Earlier the author had developed methods to build high-performance generalized cellular automata-based symmetric ciphers, which allow obtaining the encryption algorithms that show extremely high performance in hardware implementation. However, their implementation based on the conventional microprocessors lacks high performance. The mere fact is quite common - it shows a scope of applications for these ciphers. Nevertheless, the use of graphic processors enables achieving an appropriate performance for a software implementation. The article is extension of a series of the articles, which study various aspects to construct and implement cryptographic algorithms based on the generalized cellular automata. The article is aimed at studying the capabilities to implement the GPU-based cryptographic algorithms under consideration. Representing a key generator, the implemented encryption algorithm comprises 2k generalized cellular automata. The cellular automata graphs are Ramanujan’s ones. The cells of produced k gamma streams alternate, thereby allowing the GPU capabilities to be better used. To implement was used OpenCL, as the most universal and platform-independent API. The software written in C ++ was designed so that the user could set various parameters, including the encryption key, the graph structure, the local communication function, various constants, etc. To test were used a variety of graphics processors (NVIDIA GTX 650; NVIDIA GTX 770; AMD R9 280X). Depending on operating conditions, and GPU used, a performance range is from 0.47 to 6.61 Gb / s, which is comparable to the performance of the countertypes. Thus, the article has demonstrated that using the GPU makes it is possible to provide efficient software implementation of stream ciphers based on the generalized cellular automata. This work was supported by the RFBR, the project №16-07-00542.

SU-G-IeP2-13: Toward Heavy Ion Computed Tomography with Carbon Ions: A Monte Carlo Study

Purpose: In the present Monte Carlo study, we investigated the use of Carbon ions for computed tomography (CT) with a relatively low imaging dose. This will enable us to avoid any conversion of X-ray CT numbers to the relative stopping power (or relative electron density) values and the associated uncertainties in Carbon dose calculation. Methods: In the first stage, we studied the propagation of Carbon nuclei through a water phantom using the Geant4 specially to understand their lateral displacement inside the phantom. In the second stage, we used our GPU-based Monte Carlo code, which has been cross validated against Geant4, to create the 2D map of the water equivalent path length (WEPL) inside a human head size phantom by acquiring 240 projections each 1.5° apart. Subsequently the 3D relative electron density map of the phantom was reconstructed from the 2D WEPL map using the Algebraic Reconstruction Technique (ART) coupled with total variation (TV) minimization Results: A high quality image of the relative electron density inside the phantom was reconstructed by ARTTV. The mean relative error between the reconstructed image for low contrast object (PMMA) was about 1.74%. The delivered dose per scan at the center of the phantom was about 0.1 Gy. Conclusion: We have been able to obtain a 3D map of the electron density using a human head size phantom while keeping the delivered dose to relatively low value. Using the GPU capabilities of our simulation engine, we believe that a real time CT with Carbon ions could be a reality in future.

A Hybrid Sorting Algorithm on Heterogeneous Architectures

Nowadays high performance computing devices are more common than ever before. The capacity of main memories becomes very huge, CPUs get more cores and computing units that have greater performance. There are more and more machines get accelerators such as GPUs, too. Take full advantages of modern machines that use heterogeneous architectures to get higher performance solutions is a real challenge. There are so much literatures on only use CPUs or GPUs, however, research on algorithms that utilize heterogeneous architectures is comparatively few. In this paper, we propose a novel hybrid sorting algorithm that let CPU cooperate with GPU. To fully utilize computing capability of both CPU and GPU, we used SIMD intrinsic instructions to implement sorting kernels that run on CPU, and adopted radix sort kernels that implemented by CUDA(Compute Unified Device Architecture) that run on GPU. Performance evaluation is promising that our algorithm can sort one billion 32-bit float data in no more than 5 seconds.

Parallel Implementation of Scheduling Algorithms on GPU using CUDA

future of computation is the GPU, i.e. the Graphical Processing Unit. The graphics cards have shown the tremendous power in the field of image processing and accelerated generating of 3D scenes, and the computational capability of GPUs have promised its developing into great parallel computing units. It is quite simple to program a graphical processor to perform many parallel tasks. But after understanding the various aspects of the graphical processor, it can be used to perform other useful tasks as well. This paper shows how CUDA can fully utilize the tremendous power of these GPUs. CUDA is NVIDIA's parallel computing architecture which enables terrible increase in computing performance, by gearing the power of the GPU. In the first phase, several operating system algorithms in single threaded CPU environment are implemented using C language, then the same algorithms are implemented on CUDA and CUDA enabled GPU in a parallel environment and finally comparison of their performance and results to their implementation in GPU and CPU are shown.

Дослідження можливостей графічних процесорів при реалізації алгоритмів симетричного шифрування

The analysis of GPUs information technology has been conducted. The possibility of using them as additional computational resources to solve some problems that require significant computing resources (including cryptographic tasks) has been determined. Modern graphic processors can contain up to thousands of cores. One can achieve greater performance by paralleling. Block ciphers are an excellent target for parallelization - while encrypting large amounts of data you can encrypt one block in the flow. AES block encryption algorithm using OpenCL has been implemented. OpenCL is a completely open standard; its use is not subjected to royalties. The purpose of OpenCL is to complement OpenGL and OpenAL, which are open branch standards for three-dimensional computer graphics and sound, using GPU capabilities. OpenCL standard provides parallelism at the level of instructions and data level as well and is the development of GPGPU technology, which uses stream processors for non-graphical data. OpenCL appendix consists of a host programme and a set of nuclei (kernels). The dependence of speed encryption upon data volume has been studied

A Parallel Algorithm for Game Tree Search Using GPGPU

Game tree search is a classical problem in the field of game theory and artificial intelligence. Fast game tree search algorithm is critical for computer games asking for real-time responses. In this paper, we focus on how to leverage massive parallelism capabilities of GPU to accelerate the speed of game tree search algorithms and propose a concise and general parallel game tree search algorithm on GPU. The performance model of our algorithm is presented and analyzed theoretically. We implement the algorithm for two real computer games called Connect6 and Chess. We also use these two games to verify the effectiveness and efficiency of our algorithm. Experiments support our theoretical results and show good performance of our approach. Compared to classical CPU-based game tree search algorithms, our algorithm can achieve speedups of 89.95x for Connect6 and 11.43x for Chess, in case of no pruning. When pruning is considered, which means the practical performance of our algorithm, the speedup can reach about 10.58x for Connect6 and 7.26x for Chess. The insight of our work is that using GPU is a feasible way to improve the performance of game tree search algorithms.

Achieving Exascale Capabilities through Heterogeneous Computing

This article provides an overview of AMD's vision for exascale computing, and in particular, how heterogeneity will play a central role in realizing this vision. Exascale computing requires high levels of performance capabilities while staying within stringent power budgets. Using hardware optimized for specific functions is much more energy efficient than implementing those functions with general-purpose cores. However, there is a strong desire for supercomputer customers not to have to pay for custom components designed only for high-end high-performance computing systems. Therefore, high-volume GPU technology becomes a natural choice for energy-efficient data-parallel computing. To fully realize the GPU's capabilities, the authors envision exascale computing nodes that compose integrated CPUs and GPUs (that is, accelerated processing units), along with the hardware and software support to enable scientists to effectively run their scientific experiments on an exascale system. The authors discuss the hardware and software challenges in building a heterogeneous exascale system and describe ongoing research efforts at AMD to realize their exascale vision.

Mega-KV

In-memory key-value stores play a critical role in data processing to provide high throughput and low latency data accesses. In-memory key-value stores have several unique properties that include (1) data intensive operations demanding high memory bandwidth for fast data accesses, (2) high data parallelism and simple computing operations demanding many slim parallel computing units, and (3) a large working set. As data volume continues to increase, our experiments show that conventional and general-purpose multicore systems are increasingly mismatched to the special properties of key-value stores because they do not provide massive data parallelism and high memory bandwidth; the powerful but the limited number of computing cores do not satisfy the demand of the unique data processing task; and the cache hierarchy may not well benefit to the large working set. In this paper, we make a strong case for GPUs to serve as special-purpose devices to greatly accelerate the operations of in-memory key-value stores. Specifically, we present the design and implementation of Mega-KV, a GPU-based in-memory key-value store system that achieves high performance and high throughput. Effectively utilizing the high memory bandwidth and latency hiding capability of GPUs, Mega-KV provides fast data accesses and significantly boosts overall performance. Running on a commodity PC installed with two CPUs and two GPUs, Mega-KV can process up to 160+ million key-value operations per second, which is 1.4-2.8 times as fast as the state-of-the-art key-value store system on a conventional CPU-based platform.

SP-ChainMail: a GPU-based sparse parallel ChainMail algorithm for deforming medical volumes

ChainMail algorithm is a physically based deformation algorithm that has been successfully used in virtual surgery simulators, where time is a critical factor. In this paper, we present a parallel algorithm, based on ChainMail, and its efficient implementation that reduces the time required to compute deformations over large medical 3D datasets by means of modern GPU capabilities. We also present a 3D blocking scheme that reduces the amount of unnecessary processing threads. For this purpose, this paper describes a new parallel boolean reduction scheme, used to efficiently decide which blocks are computed. Finally, through an extensive analysis, we show the performance improvement achieved by our implementation of the proposed algorithm and the use of the proposed blocking scheme, due to the high spatial and temporal locality of our approach.

Optimization of minimum volume constrained hyperspectral image unmixing on CPU–GPU heterogeneous platform

Hyperspectral unmixing is essential for efficient hyperspectral image processing. Nonnegative matrix factorization based on minimum volume constraint (MVC-NMF) is one of the most widely used methods for unsupervised unmixing for hyperspectral image without the pure-pixel assumption. But the model of MVC-NMF is unstable, and the traditional solution based on projected gradient algorithm (PG-MVC-NMF) converges slowly with low accuracy. In this paper, a novel parallel method is proposed for minimum volume constrained hyperspectral image unmixing on CPU–GPU Heterogeneous Platform. First, a optimized unmixing model of minimum logarithmic volume regularized NMF is introduced and solved based on the second-order approximation of function and alternating direction method of multipliers (SO-MVC-NMF). Then, the parallel algorithm for optimized MVC-NMF (PO-MVC-NMF) is proposed based on the CPU–GPU heterogeneous platform, taking advantage of the parallel processing capabilities of GPUs and logic control abilities of CPUs. Experimental results based on both simulated and real hyperspectral images indicate that the proposed algorithm is more accurate and robust than the traditional PG-MVC-NMF, and the total speedup of PO-MVC-NMF compared to PG-MVC-NMF is over 50 times.

C‐LOD: Context‐aware Material Level‐of‐Detail applied to Mobile Graphics

AbstractAttention‐based Level‐Of–Detail (LOD) managers downgrade the quality of areas that are expected to go unnoticed by an observer to economize on computational resources. The perceptibility of lowered visual fidelity is determined by the accuracy of the attention model that assigns quality levels. Most previous attention based LOD managers do not take into account saliency provoked by context, failing to provide consistently accurate attention predictions. In this work, we extend a recent high level saliency model with four additional components yielding more accurate predictions: an object‐intrinsic factor accounting for canonical form of objects, an object‐context factor for contextual isolation of objects, a feature uniqueness term that accounts for the number of salient features in an image, and a temporal context that generates recurring fixations for objects inconsistent with the context. We conduct a perceptual experiment to acquire the weighting factors to initialize our model. We design C‐LOD, a LOD manager that maintains a constant frame rate on mobile devices by dynamically re‐adjusting material quality on secondary visual features of non‐attended objects. In a proof of concept study we establish that by incorporating C‐LOD, complex effects such as parallax occlusion mapping usually omitted in mobile devices can now be employed, without overloading GPU capability and, at the same time, conserving battery power.