• Korean Journal of Optics and Photonics
• /
• v.12 no.4
• /
• pp.339-344
• /
• 2001

We propose a method to represent binary data by modulating both the intensity and the phase of uniform plane waves with a twisted-nematic liquid crystal display for holographic digital data storage especially in a disk-shaped recording medium. As far as intensity modulation is concerned, our method is not different from the conventional method, because binary bit values 0 and 1 are represented by the dark (off) and lit (on) states of the liquid crystal display pixels, respectively (or vice versa). With our method, however, the on pixels are also controlled so that the beams passing through them can have one of two different phase delays. If the difference of the two phase delays is close to 180 degrees, we can reduce the dc component of the data image and thus improve the beam intensity uniformity at the holographic recording plane when Fourier plane holograms are recorded. The feasibility of our method is experimentally demonstrated. rated.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.9
• /
• pp.889-897
• /
• 2010

Recently, a cutting-edge large-scale image database system has demanded these attributes: search with alarming speed, performs with high accuracy, archives efficiently and much more. An image identifier (descriptor) is for measuring the similarity of two images which plays an important role in this system. The extraction method of an image identifier can be roughly classified into two methods: a local and global method. In this paper, the proposed image identifier, LFH(Local Feature's Histogram), is obtained by a histogram of robust and distinctive local descriptors (features) constrained by a district sub-division of a local region. Furthermore, LFH has not only the properties of a local and global descriptor, but also can perform calculations at a magnificent clip to determine distance with pinpoint accuracy. Additionally, we suggested a way to extract LFH via GPU (OpenGL and GLSL). In this experiment, we have compared the LFH with SIFT (local method) and EHD (global method) via storage capacity, extraction and retrieval time along with accuracy.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.2
• /
• pp.242-260
• /
• 1999

실시간 데이터압축 기법은 저장 시스템의 용량과 저장 시스템간의 대역폭을 동시에 증가시킴으로써 프로세서-메모리, 메모리-디스크간의 성능 격차를 줄이기 위한 새로운 대안으로 연구되어 왔다. 하지만 압축데이타를 복원하는데 소요되는 시간은 압축 기법의 장점을 상쇄시킬 만큼 큰 오버헤드로 작용한다. 본 논문에서는 이러한 복원 오베헤드를 줄이기 위해 선택적 압축 기법과 복원 오버헤드 감춤 기법들을 적용한 선택적 압축 메모리 시스템을 제안하고 선택적으로 압축된 데이터를 효과적으로 접근하기 위한 캐쉬 구조와 메모리 운용방법을 제시한다. 제안된 선택적 압축 메모리 시스템의 성능은 분석적 모델과 트레이스 구동 방식의 실험을 통해 평가된다. 실험 결과에 따르면 선택적 압축 메모리 시스템의 성능은 데이터의 압축율과 응용 프로그램의 데이터 접근 유형에 따른 압축 블록의 참조율 및 복원 오버헤드를 줄이기 위해 장착된 복원 버퍼의 접근 성공률에 따라 좌우됨을 알 수있다. 복원 버퍼는 대부분의 벤치마크 프로그램들에 대해서 기존의 복원 오버헤드를 70% 이상 줄여 주며 이 경우 일반적인 메모리 시스템에 비해 최대 20%까지의 성능 향상을 보인다. 뿐만 아니라 선택적 압축 기법은 평균 47% 의 데이터 이동시간의 감소와 10%의 온 칩 캐쉬 접근 실패 횟수의 감소 효과를 제공한다.

• 정보저장시스템학회:학술대회논문집
• /
• 2005.10a
• /
• pp.32-37
• /
• 2005

We developed a numerical simulator in order to study the Super-RENS/ROM (Super REsolution Near-Field Structure, Read Only Memory) using 3-dimensional FDTD (finite difference time domain) method. The simulation can be performed by three steps. In the first step, we utilized the vector-diffraction theory to calculate the characteristics of incident laser beam from the object-lens to the surface of the disk. At the second step, we fed the calculated result as an input for the main FDTD simulations on the optical layers in the disk structure. After performed the FDTD simulations, we took near-to-far field transformation for the reflected signal, from the surface of the disk to the detector. Finally, we can get reflected signal at the photo-diode. Using this developed simulator, we were able to study about the reading signal from various disk structures as a function of a laser beam position. We calculated reading signals for various pit sizes for Super-ROM structure, and it is found that the simple optical diffraction theory can not explain the reading mechanism of Super-ROM, and more complicated temperature dependent physics must be involved.

• Transactions of the Society of Information Storage Systems
• /
• v.2 no.2
• /
• pp.138-143
• /
• 2006

We developed a numerical simulator in order to study the Super-RENS/ROM (Super REsolution Near-Field Structure, Read Only Memory) using 3-dimensional FDTD (finite difference time domain) method. The simulation can be performed by three steps. In the first step, we utilized the vector-diffraction theory to calculate the characteristics of incident laser beam from the object-lens to the surface of the disk. At the second step, we fed the calculated result as an input for the main FDTD simulations on the optical layers in the disk structure. After performed the FDTD simulations, we took near-to-far field transformation for the reflected signal, from the surface of the disk to the detector. Finally, we can get reflected signal at the photo-diode. Using this developed simulator, we were able to study about the reading signal from various disk structures as a function of a laser beam position. We calculated reading signals for various pit sizes for Super-ROM structure, and it is found that the simple optical diffraction theory can not explain the reading mechanism of Super-ROM, and more complicated temperature dependent physics must be involved.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.204-207
• /
• 2008

NAND 플래시 메모리는 하드 디스크에 비해 작고, 빠르며, 저 전력 소모 등과 같은 장점을 가지고 있어 대체 저장 매체로 주목받고 있다. 그러나 제자리 갱신이 불가능한 특징을 가지고 있어 B-트리를 사용하면 갱신이 빈번하게 발생하여 읽기 연산에 비해 상대적으로 느린 쓰기 연산과 소거 연산이 빈번해져 시스템의 성능이 저하 된다. 이러한 성능 저하를 피하기 위해 $\mu}$-트리가 제안되었으나, 고정된 페이지 레이아웃 구조를 가지고 있어 노드 분할과 트리 신장이 빈번하게 일어난다. 본 논문에서는 NAND 플래시 메모리 상에서 B-트리 구현 시 발생하는 추가적인 쓰기 연산의 횟수를 줄이기 위해 갱신이 일어나는 단말 노드에 로그 노드를 할당하여, 갱신되는 내용을 저장한다. 따라서 부모 노드의 내용이 변경 되는 것을 늦추어 추가적인 쓰기 연산을 줄이게 되며, 순차적인 키 값의 삽입이나 일정 노드에 대한 빈번한 갱신은 로그 노드가 단말 노드로 전환되어 추가적인 쓰기 연산을 줄이게 된다. 이러한 방법으로 추가적인 쓰기 연산을 줄임으로써 시스템의 성능을 향상시키는 NAND 플래시 메모리를 위한 새로운 B-트리 구조를 제안한다.

• The KIPS Transactions:PartD
• /
• v.10D no.6
• /
• pp.949-960
• /
• 2003

Main memory database systems use the different implementation techniques to sturucture and organize the user dta and system catalogs, since traditional database systems are optimized for the characteristics of disk storage environment. We present, in this paper, the design considerations for our main memory database system $ALTIBASE^{TM}$ that is currently applied to the time-critical applications. We focus on the design issues of storage manager in $ALTIBASE^{TM}$. The major components are introduced, and features and characteristics of transaction management and recovery method are described. We also present the database replication mechanism and its conflict resolution mechanism for high availability and performance. In order to evaluate our transaction performance, we show various experimental reports as measured by the TPS.

• Journal of KIISE:Computer Systems and Theory
• /
• v.32 no.4
• /
• pp.177-185
• /
• 2005

NAND-type flash memory is becoming increasingly popular as a large data storage for mobile computing devices. Since flash memory is an order of magnitude more expensive than magnetic disks, data compression can be effectively used in managing flash memory based storage systems. However, compressed data management in NAND-type flash memory is challenging because it supports only page-based I/Os. For example, when the size of compressed data is smaller than the page size. internal fragmentation occurs and this degrades the effectiveness of compression seriously. In this paper, we present an efficient flash compression layer (FCL) for NAND-type flash memory which stores several small compressed pages into one physical page by using a write buffer Based on prototype implementation and simulation studies, we show that the proposed scheme offers the storage of flash memory more than $140\%$ of its original size and expands the write bandwidth significantly.

• Journal of Korea Multimedia Society
• /
• v.2 no.4
• /
• pp.367-377
• /
• 1999

Nowadays, the number of database applications which need fast transaction processing are increasing. One way to improve the performance of transaction processing is to reside the whole database in main memory As semiconductor memory becomes cheaper and chip densities increase, the research to improve transaction throughput rates of transaction processing system, using main memory databases, has begun In this thesis, how to implement a high performance transaction processing system based on main memory databases, new concurrency control scheme, recovery scheme and storage structure is presented. The objective of the proposed schemes is to improve the transaction processing system performance measured by transaction throughput and response times.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.377-378
• /
• 2007

본 논문에서는 저전력 임베디드 시스템을 위한 효율적인 다중 NAND 플래쉬 파일 시스템을 제안한다. 기존에 제안되었던 하드디스크를 비롯한 저장 장치들과는 달리 NAND 플래쉬 메모리는 특정 블록에 쓰기 연산을 하기 전에 해당 블록은 이미 소거된 상태이어야 한다. 또한 이러한 소거의 횟수는 각 블록마다 제한적이다. 이러한 문제를 해결하기 위해서 소거 횟수 평준화 기법이 많이 사용되고 있고 관련하여 많은 연구가 진행되고 있다. 본 논문에서는 소거 횟수에 임계치를 설정하여 연산하는 방법을 제안한다. 또한 기존에는 단일 플래쉬 메모리만을 고려하고 있으나 본 논문에서는 다중 플래쉬 메모리 구조를 고려한다.