• ETRI Journal
• /
• v.45 no.2
• /
• pp.318-328
• /
• 2023

Recently, embedded systems, such as mobile platforms, have multiple processing units that can operate in parallel, such as centralized processing units (CPUs) and neural processing units (NPUs). We can use deep-learning compilers to generate machine code optimized for these embedded systems from a deep neural network (DNN). However, the deep-learning compilers proposed so far generate codes that sequentially execute DNN operators on a single processing unit or parallel codes for graphic processing units (GPUs). In this study, we propose PartitionTuner, an operator scheduler for deep-learning compilers that supports multiple heterogeneous PUs including CPUs and NPUs. PartitionTuner can generate an operator-scheduling plan that uses all available PUs simultaneously to minimize overall DNN inference time. Operator scheduling is based on the analysis of DNN architecture and the performance profiles of individual and group operators measured on heterogeneous processing units. By the experiments for seven DNNs, PartitionTuner generates scheduling plans that perform 5.03% better than a static type-based operator-scheduling technique for SqueezeNet. In addition, PartitionTuner outperforms recent profiling-based operator-scheduling techniques for ResNet50, ResNet18, and SqueezeNet by 7.18%, 5.36%, and 2.73%, respectively.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.47 no.5
• /
• pp.56-70
• /
• 2010

Rendering technique generating 2 dimensional image to give reality and high performance graphical processor for efficient processing of massive data are necessary to support realistic 3 dimensional graphical image. Recently, graphical hardwares have evolved rapidly. This enables high quality rendering effect that we were unable to process in realtime. Improving shading technique enabled us to render realistic images but still much time is required for this process. Multiple operational units are being integrated in a graphical processor for effective floating point operation using massive data to process almost real looking images. In this paper, we have designed and implemented a special functional unit to support high quality 3 dimensional computer graphic image on programmable integrated shader processor. We have done evaluation through functional level simulation of designed special functional unit. Hardware resource usage rate and execution speed are measured implementing directly on FPGA Virtex-4(xc4vlx200).

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.4
• /
• pp.119-128
• /
• 2019

Due to the nature of the 3D graphics processor system, many mathematical calculations are required and parallel processing research using GPU (Graphics Processing Unit) is being performed for high-speed processing. In this paper, we propose a 3D graphics processor using MAMS, a parallel processor that does not use cache memory, to solve the GPU problem of increasing bandwidth caused by cache memory miss and the problem that 3D shader processing speed is not constant. The 3D graphics processor using MAMS proposed in this paper designed Vertex shader, Pixel shader, Tiling and Rasterizing structure using DirectX command analysis, the FPGA(Xilinx Virtex6@100MHz) board for MAMS was constructed and designed using Verilog. We compared the processing time of the developed FPGA (100Mhz) and nVidia GeForce GTX 660 (980Mhz), the processing time using GTX 660 was not constant and suing MAMS was constant.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.19 no.3
• /
• pp.29-42
• /
• 2016

Remote sensing allows acquisition of information across a large area without contacting objects, and has thus been rapidly developed by application to different areas. Thus, with the development of remote sensing, satellites are able to rapidly advance in terms of their image resolution. As a result, satellites that use remote sensing have been applied to conduct research across many areas of the world. However, while research on remote sensing is being implemented across various areas, research on data processing is presently insufficient; that is, as satellite resources are further developed, data processing continues to lag behind. Accordingly, this paper discusses plans to maximize the performance of satellite image processing by utilizing the CUDA(Compute Unified Device Architecture) Library of NVIDIA, a parallel processing technique. The discussion in this paper proceeds as follows. First, standard KOMPSAT(Korea Multi-Purpose Satellite) images of various sizes are subdivided into five types. NDVI(Normalized Difference Vegetation Index) is implemented to the subdivided images. Next, ArcMap and the two techniques, each based on CPU or GPU, are used to implement NDVI. The histograms of each image are then compared after each implementation to analyze the different processing speeds when using CPU and GPU. The results indicate that both the CPU version and GPU version images are equal with the ArcMap images, and after the histogram comparison, the NDVI code was correctly implemented. In terms of the processing speed, GPU showed 5 times faster results than CPU. Accordingly, this research shows that a parallel processing technique using CUDA Library can enhance the data processing speed of satellites images, and that this data processing benefits from multiple advanced remote sensing techniques as compared to a simple pixel computation like NDVI.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.5
• /
• pp.1177-1186
• /
• 2014

Computer-generated hologram (CGH) is to mathematically model optical phenomenon with digital computer. Because it requires huge amount of computational power, a fast and high performance technique is needed. In this paper, we proposed two parallelizations for CGH calculation. The first is to parallelize CGH algorithm in a GPU (general processing unit) and the second is to parallelize multiple GPUs. The proposed algorithm was implemented in GTX780 Ti GPU. It calculates a $1,024{\times}1,024$ hologram with 10K object points for about 24ms.

• Journal of IKEEE
• /
• v.21 no.3
• /
• pp.276-279
• /
• 2017

In this paper, SIMT structure based GPGPU (General Purpose Computing on Graphics Processing Units) is used for accelerating the Rasterizer which constitutes the screen of the display device in pixel unit. The GPU has a large number of ALUs, and the processing is very fast because of parallel processing. Therefore, in this paper, we implemented a rasterizer that generates a 3D graphics model using a CPU that performs operations sequentially and a GPU that performs operations in parallel. We confirmed that proposed rasterizer in this paper is 1.45 times better than rasterizer using Intel CPU when generating one frame.

• Journal of information and communication convergence engineering
• /
• v.9 no.3
• /
• pp.266-270
• /
• 2011

The advancement of technology enables smart phones or handheld devices to render complex 3D graphics. However, the processing power and memory of smart phones remain very limited to render high polygon and details 3D models especially on games which requires animation, physic engine, or augmented reality. In this paper, several techniques will be introduced to speed up the computation and reducing the number of vertices of the 3D meshes without losing much detail.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.181-182
• /
• 2008

본 논문에서는 연산 최적화 알고리듬 중 PSO(Particle Swarm Optimization) 알고리듬을 NVIDIA사(社)에서 제공한 CUDA(Compute Unified Device Architecture)를 이용하여 새롭게 구현하였다. CUDA는 CPU가 아닌 GPU(Graphic Processing Unit)의 다양한 병렬 처리 능력을 사용해 복잡한 컴퓨팅 문제를 해결하는 소프트웨어 개발을 가능케 하는 기술이다. 이 기술을 연산 최적화 알고리듬 중 PSO에 적용함으로써 알고리듬의 수행 속도를 개선하였다. CUDA를 적용한 PSO 알고리듬의 검증을 위해 언어 기반으로 프로그래밍하고 다양한 Test Function을 통해 시뮬레이션 하였다. 그리고 기존의 PSO 알고리듬과 비교 분석하였다. 또한 알고리듬의 성능 향상으로 여러 가지 최적화 분야에 적용 할 수 있음을 보인다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.541-542
• /
• 2021

클라우드 컴퓨팅 환경에서는 고성능 컴퓨팅을 지원하기 위해 사용자에게 GPU(Graphic Processing Unit)가 할당된 가상머신을 제공하여 사용자가 고성능 응용을 실행할 수 있도록 지원한다. 일반적인 컴퓨팅 환경에서 한 명의 사용자가 GPU를 독점해서 사용하기 때문에 자원 경쟁으로 인한 문제가 상대적으로 적게 발생하지만 독립적인 여러 사용자가 컴퓨팅 자원을 공유하는 클라우드 환경에서는 자원 경쟁으로 인해 서로 성능 영향을 미치는 문제를 발생시킨다. 본 논문에서는 여러 개의 가상머신이 단일 GPU를 공유하는 RPC(Remote Procedure Call) 기반 GPU 가상화 환경에서 다수의 가상머신이 GPGPU(General Purpose computing on Graphics Processing Units) 작업을 수행할 때 GPU 메모리 입력 경쟁으로 인해 발생하는 커널 함수의 실행 지연 문제를 분석한다.

• Nuclear Medicine and Molecular Imaging
• /
• v.43 no.5
• /
• pp.459-467
• /
• 2009

Purpose: The maximum likelihood-expectation maximization (ML-EM) is the statistical reconstruction algorithm derived from probabilistic model of the emission and detection processes. Although the ML-EM has many advantages in accuracy and utility, the use of the ML-EM is limited due to the computational burden of iterating processing on a CPU (central processing unit). In this study, we developed a parallel computing technique on GPU (graphic processing unit) for ML-EM algorithm. Materials and Methods: Using Geforce 9800 GTX+ graphic card and CUDA (compute unified device architecture) the projection and backprojection in ML-EM algorithm were parallelized by NVIDIA's technology. The time delay on computations for projection, errors between measured and estimated data and backprojection in an iteration were measured. Total time included the latency in data transmission between RAM and GPU memory. Results: The total computation time of the CPU- and GPU-based ML-EM with 32 iterations were 3.83 and 0.26 see, respectively. In this case, the computing speed was improved about 15 times on GPU. When the number of iterations increased into 1024, the CPU- and GPU-based computing took totally 18 min and 8 see, respectively. The improvement was about 135 times and was caused by delay on CPU-based computing after certain iterations. On the other hand, the GPU-based computation provided very small variation on time delay per iteration due to use of shared memory. Conclusion: The GPU-based parallel computation for ML-EM improved significantly the computing speed and stability. The developed GPU-based ML-EM algorithm could be easily modified for some other imaging geometries.