GPU-based Implementation Research Articles

A Fading Channel Simulator Implementation Based on GPU Computing Techniques

Channel simulators are powerful tools that permit performance tests of the individual parts of a wireless communication system. This is relevant when new communication algorithms are tested, because it allows us to determine if they fulfill the communications standard requirements. One of these tests consists of evaluating the system performance when a communication channel is considered. In this sense, it is possible to model the channel as an FIR filter with time-varying random coefficients. If the number of coefficients is increased, then a better approach to real scenarios can be achieved; however, in that case, the computational complexity is increased. In order to address this issue, a design methodology for computing the time-varying coefficients of the fading channel simulators using consumer-designed graphic processing units (GPUs) is proposed. With the use of GPUs and the proposed methodology, it is possible for nonspecialized users in parallel computing to accelerate their simulation developments when compared to conventional software. Implementation results show that the proposed approach allows the easy generation of communication channels while reducing the processing time. Finally, GPU-based implementation takes precedence when compared with the CPU-based implementation, due to the scattered nature of the channel.

USING PARALLEL BRANCH-AND-BOUND ALGORITHM ON GPUs FOR OPTIMAL DESIGN OF MULTI-PRODUCT BATCH PLANTS

In this paper, we propose a novel implementation of the branch-andbound algorithm on a system with GPUs (Graphics Processing Units) using the Compute Unified Device Architecture (CUDA) approach. The developed implementation is used for solving a real-world problem – the optimal selection of the chemical equipment for multi-product batch plants. We propose two implementations for the searching algorithm – an iterative and a recursive one – and we describe their optimizations, as well as compare them to each other. We report experimental results about the speedup of our GPU-based implementations as compared to the sequential CPU version.

Teraflops and Beyond: GPU-based MD Studies of Emergent Phenomena

Molecular dynamics simulations of emergent phenomena are often computationally demanding because of the broad range of length and time scales that must be covered, ranging from the individual particles, out to where the collective behavior is expressed; the fact that simulations of this type are often subject to unpredictable outcomes is a further complication. The computations re- quired for these studies benefit substantially from massively parallel GPU-based implementation, with even a single GPU typically providing an order of magnitude performance gain over a conventional CPU. A sampling of recently obtained exploratory results involving atomistic hydrodynamics, granular segregation and self-assembly are discussed, along with key aspects of the methodol- ogy; the rich behavior observed provides a hint of the kinds of phenomena that can be explored, and what might be achieved given adequate resources.

Enumerating Joint Weight of a Binary Linear Code Using Parallel Architectures: multi-core CPUs and GPUs

In this paper, we present a parallel algorithm for enumerating joint weight of a binary linear $(n,k)$ code, aiming at accelerating assessment of its decoding error probability for network coding. Our algorithm is implemented on a multi-core CPU system and an NVIDIA graphics processing unit (GPU) system using OpenMP and compute unified device architecture (CUDA), respectively. To reduce the number of pairs of codewords to be investigated, our parallel algorithm reduces dimension $k$ by focusing on the all-one vector included in many practical codes. We also employ a population count instruction to compute joint weight of codewords with a less number of instructions. Furthermore, an efficient atomic vote and reduce scheme is deployed in our GPU-based implementation. We apply our CPU- and GPU-based implementations to a subcode of a (127,22) BCH code to evaluate the impact of acceleration.

A GPU-Based Implementation of the Firefly Algorithm for Variable Selection in Multivariate Calibration Problems.

Several variable selection algorithms in multivariate calibration can be accelerated using Graphics Processing Units (GPU). Among these algorithms, the Firefly Algorithm (FA) is a recent proposed metaheuristic that may be used for variable selection. This paper presents a GPU-based FA (FA-MLR) with multiobjective formulation for variable selection in multivariate calibration problems and compares it with some traditional sequential algorithms in the literature. The advantage of the proposed implementation is demonstrated in an example involving a relatively large number of variables. The results showed that the FA-MLR, in comparison with the traditional algorithms is a more suitable choice and a relevant contribution for the variable selection problem. Additionally, the results also demonstrated that the FA-MLR performed in a GPU can be five times faster than its sequential implementation.

FDTD Acceleration for Cylindrical Resonator Design Based on the Hybrid of Single and Double Precision Floating-Point Computation

Acceleration of FDTD (finite-difference time-domain) is very important for the fields such as computational electromagnetic simulation. We consider the FDTD simulation model of cylindrical resonator design that requires double precision floating-point and cannot be done using single precision. Conventional FDTD acceleration methods have a common problem of memory-bandwidth limitation due to the large amount of parallel data access. To overcome this problem, we propose a hybrid of single and double precision floating-point computation method that reduces the data-transfer amount. We analyze the characteristics of the FDTD simulation to find out when we can use single precision instead of double precision. According to the experimental results, we achieved over 15 times of speed-up compared to the CPU single-core implementation and over 1.52 times of speed-up compared to the conventional GPU-based implementation.

Performance Evaluations of Graph Database using CUDA and OpenMP Compatible Libraries

Graph databases use graph structures to store data sets as nodes, edges, and properties. They are used to store and search the relationships between a large number of nodes, such as social networking services and recommendation engines that use customer social graphs. Since computation cost for graph search queries increases as the graph becomes large, in this pa- per we accelerate the graph search functions (Dijkstra and A* algorithms) of a graph database Neo4j using two ways: multi- threaded library and CUDA library for graphics processing units (GPUs). We use 100,000-node graphs generated based on a degree distribution of Facebook social graph for evaluations. Our multi-threaded and GPU-based implementations require an auxiliary adjacency matrix for a target graph. The results show that, when we do not take into account additional overhead to generate the auxiliary adjacency matrix, multi-threaded version improves the Dijkstra and A* search performance by 16.2x and 13.8x compared to the original implementation. The GPU-based implementation improves the Dijkstra and A* search performance by 26.2x and 32.8x. When we take into account the overhead, although the speed-ups by our implementations are reduced, by reusing the auxiliary adjacency matrix for multiple graph search queries we can significantly improve the graph search performance.

Efficient GPU-based implementations of simplex type algorithms

Recent hardware advances have made it possible to solve large scale Linear Programming problems in a short amount of time. Graphical Processing Units (GPUs) have gained a lot of popularity and have been applied to linear programming algorithms. In this paper, we propose two efficient GPU-based implementations of the Revised Simplex Algorithm and a Primal–Dual Exterior Point Simplex Algorithm. Both parallel algorithms have been implemented in MATLAB using MATLAB’s Parallel Computing Toolbox. Computational results on randomly generated optimal sparse and dense linear programming problems and on a set of benchmark problems (netlib, kennington, Mészáros) are also presented. The results show that the proposed GPU implementations outperform MATLAB’s interior point method.

Performance Analysis of Robust Image Features Detection Algorithms

Abstract This paper deals with the challenging task of acquiring stable image features in a sequence of images of the same scene taken under different viewing positions by a digital still camera. Two popular contemporary algorithms for discrete feature detection: SIFT and SURF are regarded. The results of the timing performance analysis of their sequential implementations are presented and discussed. The performance speedup analysis and scalability tests with multi-threading and GPU-based implementations are analyzed

Real-Time Stochastic Optimization of Complex Energy Systems on High-Performance Computers

A scalable approach computes in operationally-compatible time the energy dispatch under uncertainty for electrical power grid systems of realistic size with thousands of scenarios. The authors propose several algorithmic and implementation advances in their parallel solver PIPS-IPM for stochastic optimization problems. New developments include a novel, incomplete, augmented, multicore, sparse factorization implemented within the PARDISO linear solver and new multicore- and GPU-based dense matrix implementations. They show improvement on the interprocess communication on Cray XK7 and XC30 systems. PIPS-IPM is used to solve 24-hour horizon power grid problems with up to 1.95 billion decision variables and 1.94 billion constraints on Cray XK7 and Cray XC30, with observed parallel efficiencies and solution times within an operationally defined time interval. To the authors' knowledge, "real-time"-compatible performance on a broad range of architectures for this class of problems hasn't been possible prior to this work.

Direct Isosurface Ray Casting of NURBS-Based Isogeometric Analysis.

In NURBS-based isogeometric analysis, the basis functions of a 3D model's geometric description also form the basis for the solution space of variational formulations of partial differential equations. In order to visualize the results of a NURBS-based isogeometric analysis, we developed a novel GPU-based multi-pass isosurface visualization technique which performs directly on an equivalent rational Bézier representation without the need for discretization or approximation. Our approach utilizes rasterization to generate a list of intervals along the ray that each potentially contain boundary or isosurface intersections. Depth-sorting this list for each ray allows us to proceed in front-to-back order and enables early ray termination. We detect multiple intersections of a ray with the higher-order surface of the model using a sampling-based root-isolation method. The model's surfaces and the isosurfaces always appear smooth, independent of the zoom level due to our pixel-precise processing scheme. Our adaptive sampling strategy minimizes costs for point evaluations and intersection computations. The implementation shows that the proposed approach interactively visualizes volume meshes containing hundreds of thousands of Bézier elements on current graphics hardware. A comparison to a GPU-based ray casting implementation using spatial data structures indicates that our approach generally performs significantly faster while being more accurate.

Security Analysis and Related Usability of Motion-Based CAPTCHAs: Decoding Codewords in Motion

We explore the robustness and usability of moving-image object recognition (video) CAPTCHAS, designing and implementing automated attacks based on computer vision techniques. Our approach is suitable for broad classes of moving-image CAPTCHAS involving rigid objects. We first present an attack that defeats instances of such a CAPTCHA (NuCaptcha) representing the state-of-the-art, involving dynamic text strings called codewords. We then consider design modifications to mitigate the attacks (e.g., overlapping characters more closely, randomly changing the font of individual characters, or even randomly varying the number of characters in the codeword). We implement the modified CAPTCHAS and test if designs modified for greater robustness maintain usability. Our lab-based studies show that the modified captchas fail to offer viable usability, even when the captcha strength is reduced below acceptable targets. Worse yet, our GPU-based implementation shows that our automated approach can decode these captchas faster than humans can, and we can do so at a relatively low cost of roughly 50 cents per 1,000 captchas solved based on Amazon EC2 rates circa 2012. To further demonstrate the challenges in designing usable captchas, we also implement and test another variant of moving text strings using the known emerging images concept. This variant is resilient to our attacks and also offers similar usability to commercially available approaches. We explain why fundamental elements of the emerging images idea resist our current attack where others fail.

Dynamic meshing for deformable image registration

Finite element method (FEM) is commonly used for deformable image registration. However, there is no existing literature studying how the superimposed mesh structure would influence the image registration process. We study this problem in this paper, and propose a dynamic meshing strategy to generate mesh structure for image registration. To construct such a dynamic mesh during image registration, three steps are performed. Firstly, a density field that measures the importance of a pixel/voxel’s displacement to the registration process is computed. Secondly, an efficient contraction–optimization scheme is applied to compute a discrete Centroidal Voronoi Tessellation of the density field. Thirdly, the final mesh structure is constructed by its dual triangulation, with some post-processing to preserve the image boundary. In each iteration of the deformable image registration, the mesh structure is efficiently updated with GPU-based parallel implementation. We conduct experiments of the new dynamic mesh-guided registration framework on both synthetic and real medical images, and compare our results with the other state-of-the-art FEM-based image registration methods.

Heterogeneous computing architecture for fast detection of SNP-SNP interactions.

BackgroundThe extent of data in a typical genome-wide association study (GWAS) poses considerable computational challenges to software tools for gene-gene interaction discovery. Exhaustive evaluation of all interactions among hundreds of thousands to millions of single nucleotide polymorphisms (SNPs) may require weeks or even months of computation. Massively parallel hardware within a modern Graphic Processing Unit (GPU) and Many Integrated Core (MIC) coprocessors can shorten the run time considerably. While the utility of GPU-based implementations in bioinformatics has been well studied, MIC architecture has been introduced only recently and may provide a number of comparative advantages that have yet to be explored and tested.ResultsWe have developed a heterogeneous, GPU and Intel MIC-accelerated software module for SNP-SNP interaction discovery to replace the previously single-threaded computational core in the interactive web-based data exploration program SNPsyn. We report on differences between these two modern massively parallel architectures and their software environments. Their utility resulted in an order of magnitude shorter execution times when compared to the single-threaded CPU implementation. GPU implementation on a single Nvidia Tesla K20 runs twice as fast as that for the MIC architecture-based Xeon Phi P5110 coprocessor, but also requires considerably more programming effort.ConclusionsGeneral purpose GPUs are a mature platform with large amounts of computing power capable of tackling inherently parallel problems, but can prove demanding for the programmer. On the other hand the new MIC architecture, albeit lacking in performance reduces the programming effort and makes it up with a more general architecture suitable for a wider range of problems.

Prostate Segmentation: An Efficient Convex Optimization Approach With Axial Symmetry Using 3-D TRUS and MR Images

We propose a novel global optimization-based approach to segmentation of 3-D prostate transrectal ultrasound (TRUS) and T2 weighted magnetic resonance (MR) images, enforcing inherent axial symmetry of prostate shapes to simultaneously adjust a series of 2-D slice-wise segmentations in a "global" 3-D sense. We show that the introduced challenging combinatorial optimization problem can be solved globally and exactly by means of convex relaxation. In this regard, we propose a novel coherent continuous max-flow model (CCMFM), which derives a new and efficient duality-based algorithm, leading to a GPU-based implementation to achieve high computational speeds. Experiments with 25 3-D TRUS images and 30 3-D T2w MR images from our dataset, and 50 3-D T2w MR images from a public dataset, demonstrate that the proposed approach can segment a 3-D prostate TRUS/MR image within 5-6 s including 4-5 s for initialization, yielding a mean Dice similarity coefficient of 93.2%±2.0% for 3-D TRUS images and 88.5%±3.5% for 3-D MR images. The proposed method also yields relatively low intra- and inter-observer variability introduced by user manual initialization, suggesting a high reproducibility, independent of observers.

Optimizing Hopfield Neural Network for Spectral Mixture Unmixing on GPU Platform

The Hopfield neural network (HNN) has been demonstrated to be an effective tool for the spectral mixture unmixing of hyperspectral images. However, it is extremely time consuming for such per-pixel algorithm to be utilized in real-world applications. In this letter, the implementation of a multichannel structure of HNN (named as MHNN) on a graphics processing unit (GPU) platform is proposed. According to the unmixing procedure of MHNN, three levels of parallelism, including thread, block, and stream, are designed to explore the peak computing capacity of a GPU device. In addition, constant and texture memories are utilized to further improve its computational performance. Experiments on both synthetic and real hyperspectral images demonstrated that the proposed GPU-based implementation works on the peak computing ability of a GPU device and obtains several hundred times of acceleration versus the CPU-based implementation while its unmixing performance remains unchanged.

Towards a GPU-based implementation of interaction nets

We present ingpu, a GPU-based evaluator for interaction nets that heavily utilizes their potential for parallel evaluation. We discuss advantages and challenges of the ongoing implementation of ingpu and compare its performance to existing interaction nets evaluators.

A computational comparison of scaling techniques for linear optimization problems on a graphical processing unit

Preconditioning techniques are important in solving linear problems, as they improve their computational properties. Scaling is the most widely used preconditioning technique in linear optimization algorithms and is used to reduce the condition number of the constraint matrix, to improve the numerical behavior of the algorithms and to reduce the number of iterations required to solve linear problems. Graphical processing units (GPUs) have gained a lot of popularity in the recent years and have been applied for the solution of linear optimization problems. In this paper, we review and implement ten scaling techniques with a focus on the parallel implementation of them on GPUs. All these techniques have been implemented under the MATLAB and CUDA environment. Finally, a computational study on the Netlib set is presented to establish the practical value of GPU-based implementations. On average the speedup gained from the GPU implementations of all scaling methods is about 7×.

Scalable CAIM discretization on multiple GPUs using concurrent kernels

Class-attribute interdependence maximization (CAIM) is one of the state-of-the-art algorithms for discretizing data for which classes are known. However, it may take a long time when run on high-dimensional large-scale data, with large number of attributes and/or instances. This paper presents a solution to this problem by introducing a graphic processing unit (GPU)-based implementation of the CAIM algorithm that significantly speeds up the discretization process on big complex data sets. The GPU-based implementation is scalable to multiple GPU devices and enables the use of concurrent kernels execution capabilities of modern GPUs. The CAIM GPU-based model is evaluated and compared with the original CAIM using single and multi-threaded parallel configurations on 40 data sets with different characteristics. The results show great speedup, up to 139 times faster using four GPUs, which makes discretization of big data efficient and manageable. For example, discretization time of one big data set is reduced from 2 h to $$<$$ < 2 min.

Towards Affordable Computing: SiftCU a Simple but Elegant GPU-based Implementation of SIFT

This article presents a fully functional GPU-based implementation of Scale Invariant Feature Transform (SIFT) algorithm. SIFT is a popular image feature extraction algorithm. Although it is a powerful algorithm for image matching but it is also computationally very expensive. This makes it difficult to use especially in real time applications. We purpose to expedite SIFT through GPU-based implementation. There has been some related works on this issue since SIFT was introduced. Our focus is solely on describing GPU-based implementation. We will discuss our implementation in detail. Our implementation is simpler and more efficient than previous works. Part of this paper‟s purpose is to discuss challenges and strategies related to implementing SIFT like image processing algorithms on GPU. In addition, we are going to present a full comparison between serial implementations of SIFT and our GPU-based implementation, namely siftCU, both in accuracy and time consumption.