• Journal of Computing Science and Engineering
• /
• v.12 no.1
• /
• pp.24-35
• /
• 2018

Deep convolutional neural network (DCNN) is an advanced technology in image recognition. Because of extreme computing resource requirements, DCNN implementation with software alone cannot achieve real-time requirement. Therefore, the need to implement DCNN accelerator hardware is increasing. In this paper, we present a field programmable gate array (FPGA)-based hardware accelerator design of DCNN targeting handwritten Hangul character recognition application. Also, we present design optimization techniques in SDAccel environments for searching the optimal FPGA design space. The techniques we used include memory access optimization and computing unit parallelism, and data conversion. We achieved about 11.19 ms recognition time per character with Xilinx FPGA accelerator. Our design optimization was performed with Xilinx HLS and SDAccel environment targeting Kintex XCKU115 FPGA from Xilinx. Our design outperforms CPU in terms of energy efficiency (the number of samples per unit energy) by 5.88 times, and GPGPU in terms of energy efficiency by 5 times. We expect the research results will be an alternative to GPGPU solution for real-time applications, especially in data centers or server farms where energy consumption is a critical problem.

• Smart Media Journal
• /
• v.3 no.1
• /
• pp.33-38
• /
• 2014

Even though the performance of microprocessor is improved continuously, the performance improvement of computing system becomes hard to increase, in order to some drawbacks including increased power consumption. To solve the problem, general-purpose computing on graphics processing units(GPGPUs), which execute general-purpose applications by using specialized parallel-processing device representing graphics processing units(GPUs), have been focused. However, the characteristics of applications related with graphics is substantially different from the characteristics of general-purpose applications. Therefore, GPUs cannot exploit the outstanding computational resources sufficiently due to various constraints, when they execute general-purpose applications. When designing GPUs for GPGPU, memory system is important to effectively exploit the GPUs since typically general-purpose applications requires more memory accesses than graphics applications. Especially, external memory access requiring long latency impose a big overhead on the performance of GPUs. Therefore, the GPU performance must be improved if hierarchical memory architecture which can reduce the number of external memory access is applied. For this reason, we will investigate the analysis of GPU performance according to hierarchical cache architectures in executing various benchmarks.

• KIISE Transactions on Computing Practices
• /
• v.22 no.3
• /
• pp.127-132
• /
• 2016

Several methods have been proposed to apply artificial neural networks to text compression in the past. However, the networks and targets are both limited to the small size due to hardware capability in the past. Modern GPUs have much better calculation capability than CPUs in an order of magnitude now, even though CPUs have become faster. It becomes possible now to train greater and complex neural networks in a shorter time. This paper proposed a method to transform the distribution of original data with a probabilistic neural predictor. Experiments were performed on a feedforward neural network and a recurrent neural network with gated-recurrent units. The recurrent neural network model outperformed feedforward network in compression rate and prediction accuracy.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.11
• /
• pp.529-536
• /
• 2015

When rendering both three-dimensional objects and photo images together, the non-photorealistic rendering results are in visual discord since the two contents have their own independent color distributions. This paper proposes a non-photorealistic rendering technique which renders both three-dimensional objects and photo images such as cartoons and sketches. The proposed technique computes the color distribution property of the photo images and reduces the number of colors of both photo images and 3D objects. NPR is performed based on the reduced colormaps and edge features. To enhance the natural scene presentation, the image region segmentation process is preferred when extracting and applying colormaps. However, the image segmentation technique needs a lot of computational operations. It takes a long time for non-photorealistic rendering for large size frames. To speed up the time-consuming segmentation procedure, we use GPGPU for the parallel computing using the GPU. As a result, we significantly improve the execution speed of the algorithm.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.55-63
• /
• 2019

Recently, the GPU computing method has been utilized to improve the performance of the physics simulation field. In particular, in the case of a deformed object simulation requiring a large amount of computation, a GPU-based parallel processing algorithm is required to guarantee real-time performance. We have studied the parallel structure design method to improve the performance of the mass spring simulation method which is one of the methods of implementing the deformation object simulation. We used OpenGL's GLSL, a graphics library that allows direct access to the GPU, and implemented the GPGPU environment using an independent pipeline, the compute shader. In order to verify the effectiveness of the parallel structure design method, the mass - spring system was implemented based on CPU and GPU. Experimental results show that the proposed method improves computation speed by about 6,000% compared to the CPU Environment. It is expected that the lightweight simulation technology can be effectively applied to the augmented reality and the virtual reality field by using the design method proposed later in this research.

• Journal of Internet Computing and Services
• /
• v.19 no.6
• /
• pp.53-62
• /
• 2018

As the performance of computer hardware has been increased in recent years, more realistic computer generated objects can be created and presented in personal computers and portable digital devices as well. For this reason, digital contents, including computer graphics, require virtual objects that are more realistic and representable in real-time on various devices, thus it requires more computational costs. In order to support the production of contents including computer graphics, games, and animations on multi-platform, Unity or unreal engines are mainly used. To represent more realistic behavior of virtual objects in a simulation, a virtual object must collide with other virtual objects and present the plausible interaction, as in the real world. However, such dynamic simulation requires a large amount of computational cost, and most portable devices cannot provide these dynamic simulations in real-time. In this paper, we proposed a GPGPU computation based dynamic cloth simulation to represent collision and response with spherical object in real-time. We believe that the proposed method can be useful for readily producing realistic digital contents.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.3
• /
• pp.23-32
• /
• 2008

The computing power of graphics processing units(GPU) has already surpassed that of CPU and the gap between their powers is getting wider. Thus, research on GPGPU which applies GPU to general purpose becomes popular and shows great success especially in the field of parallel data processing. Since the implementation of cryptographic algorithm using GPU was started by Cook et at. in 2005, improved results using graphic libraries such as OpenGL and DirectX have been published. In this paper, we present skills and results of implementing block ciphers using CUDA library announced by NVIDIA in 2007. Also, we discuss a general method converting source codes of block ciphers on CPU to those on GPU. On NVIDIA 8800GTX GPU, the resulting speeds of block cipher AES, ARIA, and DES are 4.5Gbps, 7.0Gbps, and 2.8Gbps, respectively which are faster than the those on CPU.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.5
• /
• pp.656-662
• /
• 2018

Pattern matching algorithm is widely used in many application fields such as bio-informatics, intrusion detection, etc. Among many string matching algorithms, KMP (Knuth-Morris-Pratt) algorithm is commonly used because of its fast execution time when using large texts. However, the processing speed of KMP algorithm is also limited when the text size increases significantly. In this paper, we propose a high throughput parallel KMP algorithm considering CPU-GPU memory hierarchy based on OpenCL in GPGPU (General Purpose computing on Graphic Processing Unit). We focus on the optimization for the allocation of work-times and work-groups, the local memory copy of the pattern data and the failure table, and the overlapping of the data transfer with the string matching operations. The experimental results show that the execution time of the optimized parallel KMP algorithm is about 3.6 times faster than that of the non-optimized parallel KMP algorithm.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.11 no.1
• /
• pp.27-35
• /
• 2015

Increasing the number of GPU (Graphic Processor Unit) cores, the studies on High Performance Computing Platform using GPU have actively been made in recent. This trend has led to the development of GPGPU (General Purpose GPU) and CUDA (Compute Unified Device Architecture) Framework. In this paper, we explain the many benefits of the GPU based system, and propose the ICIDF(Identify Compute-Intensive Data set and Function) methodology to apply GPU technology to legacy military information system for performance improvement. To demonstrate the efficiency of this methodology, we applied this method to AES CPU based program obtained from the Internet web site. Simply changing the data structure made improved the performance of AES program. As a result, the performance of AES based GPU program is improved gradually up to 10 times. Depending on the developer's ability, additional performance improvement can be expected. The problem to be solved is heat issue, but this problem has been much improved by the development of the cooling technology.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.6
• /
• pp.1035-1043
• /
• 2022

With the spread of the Internet of Things (IoT), cloud computing, and big data, the need for high-speed encryption for applications is emerging. GPU optimization can be used to validate cryptographic analysis results or reduced versions theoretically obtained by the GPU in a reasonable time. In this paper, PIPO lightweight encryption implemented in various environments was implemented on GPU. Optimally implemented considering the brute force attack on PIPO. In particular, the optimization implementation applying the bit slicing technique and the GPU elements were used as much as possible. As a result, the implementation of the proposed method showed a throughput of about 19.5 billion per second in the RTX 3060 environment, achieving a throughput of about 122 times higher than that of the previous study.