CFD on Graphic Cards

We present a novel heterogeneous parallel matrix multiplication algorithm that utilizes both central processing units (CPUs) and graphics processing units (GPUs) for large-scale matrices. Based on Strassen’s method, we represent matrix multiplication work as a set of matrix addition and multiplication tasks among their sub-matrices. Then, we distribute the tasks to CPUs and GPUs while considering the characteristics of the tasks and computing resources to minimize the data communication overhead and fully utilize the available computing power. To handle a large matrix efficiently with limited GPU memory, we also propose a block-based work decomposition method. We then further improve the performance of our method by exploiting the concurrent execution abilities of a heterogeneous parallel computing system. We implemented our method on five different heterogeneous systems and applied it to matrices of various sizes. Our method generally shows higher performance than the prior GPU-based matrix multiplication methods. Moreover, compared with the state-of-the-art GPU matrix multiplication library (i.e., CUBLAS), our method achieved up to 1.97 times higher performance using the same GPUs and CPU cores. In some cases, our method using a low-performance GPU (e.g., GTX 1060, 3 GB) achieved performance comparable to that of CUBLAS using a high-performance GPU (e.g., RTX 2080, 8 GB). Also, our method continually improves performance as we use more computing resources like additional CPU cores and GPUs. We could achieve such high performance because our approach fully utilized the capacities of the given heterogeneous parallel computing systems while employing the Strassen’s method, which has a lower asymptotic complexity. These results demonstrate the efficiency and robustness of our algorithm.

Bioinspired methods usually need a high amount of computational resources. For this reason, parallelization is an interesting alternative in order to decrease the execution time and to provide accurate results. In this sense, recently there has been a growing interest in developing parallel algorithms using graphic processing units (GPU) also refered as GPU computation. Advances in the video gaming industry have led to the production of low-cost, high-performance graphics processing units (GPUs) that possess more memory bandwidth and computational capability than central processing units (CPUs). As GPUs are available in personal computers, and they are easy to use and manage through several GPU programming languages (CUDA, OpenCL, etc.), graphics engines are being adopted widely in scientific computing applications, particularly in the fields of computational biology and bioinformatics. This paper reviews the use of GPUs to solve scientific problems, giving an overview of current software systems.

To understand the mechanism of information processing by a biological neural network, computer simulation of a large-scale spiking neural network is an important method. However, because of a high computation cost of the simulation of a large-scale spiking neural network, the simulation requires high performance computing implemented by a supercomputer or a computer cluster. Recently, hardware for parallel computing such as a multi-core CPU and a graphics card with a graphics processing unit (GPU) is built in a gaming computer and a workstation. Thus, parallel computing using this hardware is becoming widespread, allowing us to obtain powerful computing power for simulation of a large-scale spiking neural network. However, it is not clear how much increased performance the parallel computing method using a new GPU yields in the simulation of a large-scale spiking neural network. In this study, we compared computation time between the computing methods with CPUs and GPUs in a simulation of neuronal models. We developed computer programs of neuronal simulations for the computing systems that consist of a gaming graphics card with new architecture (the NVIDIA GTX 1080) and an accelerator board using a GPU (the NVIDIA Tesla K20C). Our results show that the computing systems can perform a simulation of a large number of neurons faster than CPU-based systems. Furthermore, we investigated the accuracy of a simulation using single precision floating point. We show that the simulation results of single precision were accurate enough compared with those of double precision, but chaotic neuronal response calculated by a GPU using single precision is prominently different from that calculated by a CPU using double precision. Furthermore, the difference in chaotic dynamics appeared even if we used double precision of a GPU. In conclusion, the GPU-based computing system exhibits a higher computing performance than the CPU-based system, even if the GPU system includes data transfer from a graphics card to host memory.

We present a high performance graphics processing unit (GPU) algorithm ISSD (Improved Sorting considering Special Distributions) implemented with the Compute Unified Device Architecture (CUDA). The ISSD focuses on two aspects to improve performance. One is how to decompose the tasks into independent and balanced subtasks which can then be easily distributed to thousands of threads to realize the concept of parallel sorting as well as to efficiently explore the power of GPU. The other one is how to take advantage of special data distributions to further optimize the algorithms and improve their performance. The algorithm is redesigned based on our previous general data distribution version and optimized both on general implementation methods and special input distributions. Experimental results show that for the general data distribution inputs, the ISSD outperforms the existing algorithms by about 10% in performance due to its practical optimization in implementation, and for three special data distribution inputs, the ISSD outperforms the existing algorithms by more than 40% due to its special optimization based on the data distributions. Therefore, the algorithm is viable and efficient when dealing with specific data distributions.

BackgroundSearching for similarities in protein and DNA databases has become a routine procedure in Molecular Biology. The Smith-Waterman algorithm has been available for more than 25 years. It is based on a dynamic programming approach that explores all the possible alignments between two sequences; as a result it returns the optimal local alignment. Unfortunately, the computational cost is very high, requiring a number of operations proportional to the product of the length of two sequences. Furthermore, the exponential growth of protein and DNA databases makes the Smith-Waterman algorithm unrealistic for searching similarities in large sets of sequences. For these reasons heuristic approaches such as those implemented in FASTA and BLAST tend to be preferred, allowing faster execution times at the cost of reduced sensitivity. The main motivation of our work is to exploit the huge computational power of commonly available graphic cards, to develop high performance solutions for sequence alignment.ResultsIn this paper we present what we believe is the fastest solution of the exact Smith-Waterman algorithm running on commodity hardware. It is implemented in the recently released CUDA programming environment by NVidia. CUDA allows direct access to the hardware primitives of the last-generation Graphics Processing Units (GPU) G80. Speeds of more than 3.5 GCUPS (Giga Cell Updates Per Second) are achieved on a workstation running two GeForce 8800 GTX. Exhaustive tests have been done to compare our implementation to SSEARCH and BLAST, running on a 3 GHz Intel Pentium IV processor. Our solution was also compared to a recently published GPU implementation and to a Single Instruction Multiple Data (SIMD) solution. These tests show that our implementation performs from 2 to 30 times faster than any other previous attempt available on commodity hardware.ConclusionsThe results show that graphic cards are now sufficiently advanced to be used as efficient hardware accelerators for sequence alignment. Their performance is better than any alternative available on commodity hardware platforms. The solution presented in this paper allows large scale alignments to be performed at low cost, using the exact Smith-Waterman algorithm instead of the largely adopted heuristic approaches.

Online courses have potential to reach broader audience when compared to traditional learning methods, as they can be made instantly available to groups of students dispersed over wide range of locations and time zones. Some learning topics are easily adopted to this online environment, while others may be challenging to implement as online courses due to their specific requirements and associated cost. If mastering certain topic requires hands-on experience then online course must be made interactive to match the experience of a traditional instructor-led class. Otherwise, online course will be perceived as unsatisfactory and lacking positive impact. Computer Aided Design (CAD) training requires both high performance Graphics Processing Unit (GPU) and hands-on experience with specific CAD software. Both these requirements are difficult to meet on a remote student computer without additional cost. In this paper, grouping CAD workstations into cells is proposed in order to perform required graphics processing using existing hardware and software resources available in an average CAD laboratory. Proposed cell framework uses video capture and GPU hardware encoding to stream the content to the remote students while capturing their interactive feedback for the near real-time hands-on experience. The framework requires single capture card per cell and utilizes 20% of the resources for the cell overhead processing. Remaining 80% of workstations are fully available to the online students and instructors.

SummaryMany e‐business or social network servers have been constructed on cloud. On such open environments, private data of massive users have to be protected by encrypting, such as using Advanced Encryption Standard (AES), and furthermore, this process must be finished in a short time for users' better experience. This gives huge pressure on cloud servers, especially common servers, such as web servers. We urgently need an inexpensive and highly efficient method to relieve cloud servers' pressure. Fortunately, many cores of a graphics processing unit (GPU) can undertake this hard mission because of stronger computing power and lower price. The GPU environments can be virtualized on demand by cloud through the vCUDA technology. Of course, for those clouds not equipped with a GPU, a central processing unit (CPU) can still work as multithreads in parallel. Thus, in a cloud, AES can be parallelized using many cores of a GPU or multicores of a CPU with high efficiency and low cost. For typical cloud applications, such as web services, there are massive users and each one has short plaintext. If we simply parallelize AES in such an application, we cannot obtain better performance because of the GPU's extra data transferring cost. Thus, we coalesce the massive users' data and cut these data into same‐length slices for improving the performance of parallel AES as much as possible. So we design six parallel AES algorithms using GPU parallelism or CPU parallelism, which differ in parallel scope and whether data are coalesced or cut to slices. Specifically, they are coalescent and sliced GPU (GCS), coalescent and unsliced GPU, uncoalescent GPU, coalescent and sliced CPU, coalescent and unsliced CPU, and uncoalescent CPU. Moreover, we implement them on two representative platforms and evaluate their performance. Through comparing their performance, GCS has the best performance among these algorithms. In a cloud with Nvidia GPUs, GCS is a more powerful algorithm for massive users' data encrypting, relatively. Copyright © 2015 John Wiley & Sons, Ltd.

Graphic Processing Units (GPUs) have evolved to provide a massive computational power. In contrast to Central Processing Units, GPUs are so-called many-core processors with hundreds of cores capable of running thousands of threads in parallel. This parallel processing power can accelerate the simulation of communication systems. In this work, we utilize NVIDIA's Compute Unified Device Architecture (CUDA) to execute two different sphere decoders on a graphic card. Both flat fading and frequency selective channels are considered. We find that the execution of the soft-sphere decoder can be accelerated by factors of 6-8, and the fixed-complexity sphere decoder even by a factor of 50.

MSA Probs is a parallel algorithm developed to align multiple sequence alignment using a central processing unit (CPU). Whereas the CPU has some limitations, such as the inability to parallelize tasks in the processor (latency-oriented). To overcome these limitations, this paper proposes an improved version of MSA Probs that is compatible with a graphical Processing Unit (GPU). This idea helps in enhancing the performance of our algorithm (MSAprobs). To parallelize the sequential algorithm, Compute Unified Device Architecture (CUDA) or OpenCL is commonly used on GPUs. The NIVIDIA API is used to investigate the GPU's computing power. The results of using CPU only in MSAprobs versus the CPU and GPU are compared using two data sets from the Bali Base and OX Bench. The evaluation of the CPU and GPU is done using Threads 1,2 and 4. The results showed that by combining CPU and GPU, performance is improved and execution time is reduced.

There are two inherent obstacles to effectively using Graphics Processing Units (GPUs) for query processing in search engines: (a) the highly restricted GPU memory space, and (b) the CPU-GPU transfer latency. Previously, Ao et al. presented a GPU method for lists intersection, an essential component in AND-based query processing. However, this work assumes the whole inverted index can be stored in GPU memory and does not address document ranking. In this paper, we describe and analyze a GPU query processing method which incorporates both lists intersection and top-k ranking. We introduce a parameterized pruned posting list GPU caching method where the parameter determines how much GPU memory is used for caching. This method allows list caching for large inverted indexes using the limited GPU memory, thereby making a qualitative improvement over previous work. We also give a mathematical model which can identify an approximately optimal choice of the parameter. Experimental results indicate that this GPU approach under the pruned list caching policy achieves better query throughput than its CPU counterpart, even when the inverted index size is much larger than the GPU memory space.

Graphics Processing Units (GPUs) are used as general purpose parallel accelerators in a wide range of applications. They are found in most computing systems, and mobile devices are no exception. The recent availability of programming APIs such as OpenCL for mobile GPUs promises to open up new types of applications on these devices.However, producing high performance GPU code is extremely difficult. Subtle differences in device characteristics can lead to large performance variations when different optimizations are applied. As we will see, this is especially true for a mobile GPU such as the ARM Mali GPU which has a very different architecture than desktop-class GPUs. Code optimized and tuned for one type of GPUs is unlikely to achieve the performance potential on another type of GPUs.Auto-tuners have traditionally been an answer to this performance portability challenge. For instance, they have been successful on CPUs for matrix operations, which are used as building blocks in many high-performance applications. However, they are much harder to design for different classes of GPUs, given the wide variety of hardware characteristics.In this paper, we take a different perspective and show how performance portability for matrix multiplication is achieved using a compiler approach. This approach is based on a recently developed generic technique that combines a high-level programming model with a system of rewrite rules. Programs are automatically rewritten in successive steps, where optimizations decision are made. This approach is truly performance portable, resulting in high-performance code for very different types of architectures such as desktop and mobile GPUs. In particular, we achieve a speedup of 1.7x over a state-of-the-art auto-tuner on the ARM Mali GPU.

Hundreds of genome-wide association studies have been performed over the last decade, but as single nucleotide polymorphism (SNP) chip density has increased so has the computational burden to search for epistasis [for n SNPs the computational time resource is O(n(n-1)/2)]. While the theoretical contribution of epistasis toward phenotypes of medical and economic importance is widely discussed, empirical evidence is conspicuously absent because its analysis is often computationally prohibitive. To facilitate resolution in this field, tools must be made available that can render the search for epistasis universally viable in terms of hardware availability, cost and computational time. By partitioning the 2D search grid across the multicore architecture of a modern consumer graphics processing unit (GPU), we report a 92× increase in the speed of an exhaustive pairwise epistasis scan for a quantitative phenotype, and we expect the speed to increase as graphics cards continue to improve. To achieve a comparable computational improvement without a graphics card would require a large compute-cluster, an option that is often financially non-viable. The implementation presented uses OpenCL--an open-source library designed to run on any commercially available GPU and on any operating system. The software is free, open-source, platformindependent and GPU-vendor independent. It can be downloaded from http://sourceforge.net/projects/epigpu/.

For a phase-sensitive optical time-domain reflectometer (ϕ-OTDR) distributed sensor system, space-frequency analysis can reduce the false alarm by analyzing the frequency distribution compared with the traditional difference value method. We propose a graphics processing unit (GPU)-based parallel computing method to perform multichannel fast Fourier transform (FFT) and realize the real-time space-frequency analysis. The experiment results show that the time taken by the multichannel FFT decreased considerably based on this GPU parallel computing. The method can be completed with a sensing fiber up to 16 km long and an entry-level GPU. Meanwhile, the GPU can reduce the computing load of the central processing unit from 70% down to less than 20%. We carried out an experiment on a two-point space-frequency analysis, and the results clearly and simultaneously show the vibration point locations and frequency components. The sensor system outputs the real-time space-frequency spectra continuously with a spatial resolution of 16.3 m and frequency resolution of 2.25 Hz.

In the phase-field simulation of dendrite growth during the solidification of an alloy, the computational cost becomes extremely high when the diffusion length is significantly larger than the curvature radius of a dendrite tip. In such cases, the adaptive mesh refinement (AMR) method is effective for improving the computational performance. In this study, we perform a three-dimensional dendrite growth phase-field simulation in which AMR is implemented via parallel computing using multiple graphics processing units (GPUs), which provide high parallel computation performance. In the parallel GPU computation, we apply dynamic load balancing to parallel computing to equalize the computational cost per GPU. The accuracy of an AMR refinement condition is confirmed through the single-GPU computations of columnar dendrite growth during the directional solidification of a binary alloy. Next, we evaluate the efficiency of dynamic load balancing by performing multiple-GPU parallel computations for three different directional solidification simulations using a moving frame algorithm. Finally, weak scaling tests are performed to confirm the parallel efficiency of the developed code.

Implementation of Image Enhancement Algorithms and Recursive Ray Tracing using CUDA