• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1713-1719
• /
• 2022

As the noise canceler with a small processing delay increases the sampling frequency, a better-quality output can be obtained. For a single buffer, processing delay occurs because it is impossible to write new data while the processor is processing the data. When synthesizing with anti-noise and output signal, this processing delay creates additional buffering overhead to match the phase. In this paper, we propose an accelerator structure that minimizes processing delay and increases processing speed by alternately performing read and write operations using the Symmetric Even-Odd-buffer. In addition, we compare the structural differences between the two methods of noise cancellation (Fast Fourier Transform noise cancellation and adaptive Least Mean Square algorithm). As a result, using an Symmetric Even-Odd-buffer the processing delay was reduced by 29.2% compared to a single buffer. The proposed Symmetric Even-Odd-buffer structure has the advantage that it can be applied to various canceling algorithms.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.11
• /
• pp.5271-5289
• /
• 2016

On large-scale data analysis platforms deployed on cloud infrastructures over the Internet, the instability of the data transfer time and the dynamics of the processing rate require a more sophisticated data distribution scheme which maximizes parallel efficiency by achieving the balanced load among participated computing elements and by eliminating the idle time of each computing element. In particular, under the constraints that have the real-time and limited data buffer (in-memory storage) are given, it needs more controllable mechanism to prevent both the overflow and the underflow of the finite buffer. In this paper, we propose an auto regulated data provisioning model based on receiver-driven data pull model. On this model, we provide a synchronized data replenishment mechanism that implicitly avoids the data buffer overflow as well as explicitly regulates the data buffer underflow by adequately adjusting the buffer resilience. To estimate the optimal size of buffer resilience, we exploits an adaptive buffer resilience control scheme that minimizes both data buffer space and idle time of the processing elements based on directly measured sample path analysis. The simulation results show that the proposed scheme provides allowable approximation compared to the numerical results. Also, it is suitably efficient to apply for such a dynamic environment that cannot postulate the stochastic characteristic for the data transfer time, the data processing rate, or even an environment where the fluctuation of the both is presented.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.2
• /
• pp.407-415
• /
• 1996

This paper introduces two schemes to improve the frame buffer access bandwidth. The first scheme suggests a rasterizer called SBUFRE that has Span Z Buffer and Span Z＆ Color Buffer within a rasterizer. The second scheme suggests a ZDRAM that has Z-value comparator within the DRAM. These schemes are to convert read- modify-write Z buffer compare into single write only operation that improves approximately 50% frame buffer access bandwidth.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.396-398
• /
• 2023

Statista 통계 조사에 따르면 데이터의 규모는 매년 증가할 것으로 예상하고 빅데이터 처리 프레임워크의 관심이 높아지고 있다. 빅데이터 처리 프레임워크 Spark는 Shuffle 과정에서 노드 간 데이터 전송이 일어난다. 이때 분산 처리한 데이터를 네트워크로 전송하기 위해 객체를 바이트 스트림으로 변환하여 메모리 buffer에 담는 직렬화 작업이 필요하다. 그러나 바이트 스트림을 buffer에 담는 과정에서 바이트 스트림의 크기가 메모리 buffer보다 클 경우, 메모리 할당 과정이 추가로 발생하여 전체적이 Spark의 성능 저하로 이어질 수 있다. 이에 본 논문에서는 Spark 환경에서 분산 처리 성능 향상을 위한 직렬화 buffer 최적화 시스템을 제안한다. 제안하는 방법은 Spark Driver가 Executor에게 작업을 할당하기 전 직렬화된 데이터 크기 측정과 직렬화 옵션 설정을 통해 Executor에게 적절한 buffer를 할당할 수 있다. 향후 제안하는 방법의 검증을 위해 실제 Spark 클러스터 환경에서 성능 평가가 필요하다.

• Journal of Ceramic Processing Research
• /
• v.13 no.spc2
• /
• pp.336-340
• /
• 2012

Wide band gap p-type hydrogenated amorphous silicon oxide (a-SiO:H) buffer layer has been used at the interface of transparent conductive oxide (TCO) and hydrogenated amorphous silicon (a-Si:H) p-type layer of a p-i-n type a-Si:H solar cell. Introduction of 5 nm thick buffer layer improves in blue response of the cell along with 0.5% enhancement of photovoltaic conversion efficiency (η). The cells with buffer layer show higher open circuit voltage (Voc), fill factor (FF), short circuit current density (Jsc) and improved blue response with respect to the cell without buffer layer.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.6
• /
• pp.1375-1385
• /
• 1996

As there has been much demands for processing multimedia data, a storage manager for multimedia data makes much effects on system performance. Because the size of multimedia data is usually very large, disk I/O for the data consumes much time and causes the system performance to be decreased. Therefore, it makes a better effect on system performance that a multimedia data storage manager decreases its disk I/O by the buffer management of multimedia data. This paper proposes a buffer management technique which allocates the buffer to be equal to its corresponding segment which consists of physically continuous disk page set and is disk management unit for multimedia data in many multimedia data storage manager. As the size of buffer varies, it also proposes a buffer replacement policy which consider not only reference behavior of buffer but also buffer size The proposed multimedia data buffer management technique is implemented in KORED/STORM which is a storage manager for multimedia data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1248-1250
• /
• 2015

MySQL 데이터베이스에서 사용하는 Non-clustered Secondary Index는 디스크 접근 시 Random한 입출력을 유발하여 디스크 장치의 성능을 저하시키는 문제점을 가지고 있다. 이를 해결하기 위해 MySQL의 Storage Engine은 Insert Buffer를 사용하여 Random한 디스크 접근을 방지한다. Benchmark를 통해 성능 평가를 진행한 결과 Insert Buffer를 사용하는 것만으로 성능이 개선되는 것을 확인하였다. 또한, 현재의 Insert Buffer를 수정하여 더 큰 성능 향상을 이끌어낼 수 있는 부분을 발견하여 이에 대한 아이디어를 제시, 간단한 구현을 통해 추가적인 성능 향상을 보였다.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2001.10a
• /
• pp.263-268
• /
• 2001

In this paper, we present a new heuristic algorithm fur buffer allocation in FMS (Flexible Manufacturing System). It is conducted by using a genetic algorithm and simulation. First, we model the system by using a simulation software, \"Arena\". Then, we apply a genetic algorithm to achieve an optimal solution. VBA blocks, which are kinds of add-in functions in Arena, are used to connect Arena with the genetic algorithm. The system being modeled has seven workstations, one loading/unloading station, and three AGVs (Automated Guided Vehicle). Also it contains three products, which each have their own machining order and processing times. We experimented with two kinds of buffer allocation problems with a proposed heuristic algorithm, and we will suggest a simple heuristic approach based on processing times and workloads to validate our proposed algorithm. The first experiment is to find a buffer profile to achieve the maximum throughput using a finite number of buffers. The second experiment is to find the minimum number of buffers to achieve the desired throughput. End of this paper, we compare the result of a proposed algorithm with the result of a simple buffer allocation heuristic based on processing times and workloads. We show that the proposed algorithm increase the throughput by 7.2%.t by 7.2%.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.228-231
• /
• 2008

For a satellite data communication, the technology of frame synchronization is widely used between a sender and a receiver. Last year, we suggested zero-loss frame synchronization [1] using pattern search and using bits threshold search algorithm that is based on SIMD technology [2,3]. This algorithm could solve both of hardware and software drawbacks, which are frame loss and low processing performance. However, this algorithm didn't optimize the processing of output data, synchronized data, which caused overhead to the memory allocation and the memory copy. Consequently, the performance of the frame synchronizer application was degraded. In this paper, we enhance previous work using a circular buffer in order to optimize the output data processing. The performance comparison with the previous algorithm shows that the enhanced proposed approach dramatically outperforms in the output data processing speed.

• Progress in Superconductivity and Cryogenics
• /
• v.8 no.3
• /
• pp.37-40
• /
• 2006

Oxide buffer layers for YBCO coated conductors were fabricated using MOD processing and development of microstructure and texture were investigated. A $CeO_2$ buffer layers were formed on RABiTS tape. Acetate-based precursor solution was employed to synthesize the precursor solution. Subsequently, the precursor solution was stabilized and modified with triethanolamine. $CeO_2$ precursor gel film was coated and annealed in $Ar/H_2$ atmosphere at high temperature. An annealed $CeO_2$ film shows mixed orientation with high (001) texturing. It was shown that (111) texture of $CeO_2$ layers were enhanced by multiple coating. This degradation was attributed to development of microcracks in the multiply coated $CeO_2$ films. Also discussed are the synthesis and the characterization of $La_2Zr_2O_7$ (LZO) buffer layers on RABiTS tape. A biaxially textured LZO buffer layer was fabricated with MOD processing method using metal alkoxide based precursor solution. It was shown that the LZO film were epitaxially grown on RABiTS tape and crack-free & uniform surface was obtained after annealing in $Ar/H_2$ atmosphere.