Earticle

본 대부분의 기업들은 업무 프로세스가 전산화되어 있기 때문에 전산장애나 재해로 인해 시스템이 다운되면 기업의 신뢰성과 사업의 연속성에 치명적인 악영향을 미칠 수 있다. 이에 따라 기업에서는 데이터 손실이 없는 비즈니스 연속성에 대한 관심이 고도화되는 실정이다. 본 논문에서는 시스템 장애를 물리적 장애와 논리적 장애로 분리하여 정의하고 실시간 데이터백업에 필요한 IOPS 성능을 개선하기 위한 Flash SSD를 이용하는 CDP 시스템을 제안하였다. Flash SSD를 사용한 CDP의 IOPS 성능을 측정하기 위하여 실험 시스템을 구성하고, 블록크기의 변화에 따른 IOPS를 실험한 결과, Flash SSD를 이용한 CDP 시스템의 IOPS 성능이 S-ATA에 비해 약 50배 향상되는 것으로 확인하였다.

If the System is downed by computer or disasters, it can have a bad influence on the reliability of the corporation and business continuity because all companies are computerized. Accordingly, interest on the business continuity without the loss of data in the corporate is increasing. In this paper, system faults have been defined as physical faults and logical faults and CDP solution using Flash SSD has been proposed for enhancing IOPS which is needed for realtime-backup. In order to measure IOPS performance of the CDP using Flash SSD, we constructed an experimental system. From the results we can see that IOPS performance of CDP using Flash SSD is about 50 times more effective than that of the S-ATA.

Many studies on flash memory-based buffer replacement algorithms that consider the characteristics of flash memory have recently been developed. Conventional flash memory-based buffer replacement algorithms have the disadvantage that the operation speed slows down, because only the reference is checked when selecting a replacement target page and either the reference count is not considered, or when the reference time is considered, the elapsed time is considered. Therefore, this paper seeks to solve the problem of conventional flash memory-based buffer replacement algorithm by dividing pages into groups and considering the reference frequency and reference time when selecting the replacement target page. In addition, because flash memory has a limited lifespan, candidates for replacement pages are selected based on the number of deletions.

정보통신기술의 발달로 인하여 e-learning산업은 급속히 팽창하고 있으며, 이외 함께 제작 기술의 발전도 빠르게 이루어지고 있다. 기존의 동영상 강의 제작의 큰 비중을 차지하던 Microsoft 사의 WMV(Window Media Video) 형식에서 최근에는 다른 동영상 파일에 비해 용량은 작지만 화질이 매우 우수한 포맷인 FLV(Flash video)으로 전환되고 있는 추세이다. 플래쉬 기반으로 이루어진 Flv는 윈도우, 리눅스 이외에도 대부분의 OS 환경과 브라우저에서 구동이 되며, 사용자가 별도의 플레이어 설치나 코덱 설치 과정 없이 재생을 할 수 있어 차세대 표준형, 개방형 플랫폼으로 개발자들의 관심을 끌고 있으나, 아직까지는 기존의 WMV 형식과 FLV간의 용량이나 화질의 비교, 제작시간 등에 대한 정확한 비교 분석에 관한 데이터가 나와 있지 않은 실정이다. 따라서 본 논문은 제작 방식에 따라 WMV와 FLV간의 데이터 비교 분석을 통하여 각기 특성에 맞는 제작 플랫폼을 설정하는데 가이드를 제시하고자 한다.

Due to the development of information telecommunication technology, e-learning industry is rapidly expanding its scope along with its production technology. The recent trend of e-learning program is likely converted from Wmv(Window Media Video) of Microsoft to Flv(Flash video), which has less capacity but better quality than other image file. It has successfully drawn the users attention since Flv can operate at most OS environments and browsers let alone with window and Lenux without extra players and codec setup. However, there is no accurate data on comparative analysis between Wmv and Flv regarding capacity, quality and production time. Therefore, the study shows the comparative data analysis on Wmv and Flv so as to set out production platform up to its idiosyncrasy.

NAND flash memory-based embedded systems are becoming increasingly common. These embedded systems have to provide a fast boot time. In this paper, we have designed and proposed a flash file system for embedded systems that require fast booting. By using a Flash Image Area, which keeps the latest flash memory information such as types and status of all blocks, the file system mounting time can be reduced significantly. We have shown by experiments that our file system outperforms YAFFS and RFFS.

Most traditional database systems exploit a record-oriented model where the attributes of a record are placed contiguously in a hard disk to achieve high performance writes. However, for read-mostly data warehouse systems, the column-oriented database has become a proper model because of its superior read performance. Today, flash memory is largely recognized as the preferred storage media for high-speed database systems. In this paper, we introduce a column-oriented database model based on flash memory and then propose a new column-aware flash indexing scheme for the high-speed column-oriented data warehouse systems. Our index management scheme, which uses an enhanced $B^+$-Tree, achieves superior search performance by indexing an embedded segment and packing an unused space in internal and leaf nodes. Based on the performance results of two test databases, we concluded that the column-aware flash index management outperforms the traditional scheme in the respect of the mixed operation throughput and its response time.

The lifetime of a solid-state drive (SSD) is limited because of the number of program and erase cycles allowed on its NAND flash blocks. Data cannot be overwritten in an SSD, leading to an out-of-place update every time the data are modified. This result in two copies of the data: the original copy and a modified copy. This phenomenon is known as write amplification and adversely affects the endurance of the memory. In this study, we address the issue of reducing wear leveling through efficient handling of write requests. This results in even wearing of all the blocks, thereby increasing the endurance period. The focus of our work is to logically divert the write requests, which are concentrated to limited blocks, to the less-worn blocks and then measure the maximum number of write requests that the memory can handle. A memory without the proposed algorithm wears out prematurely as compared to that with the algorithm. The main feature of the proposed algorithm is to delay out-of-place updates till the threshold is reached, which results in a low overhead. Further, the algorithm increases endurance by a factor of the threshold level multiplied by the number of blocks in the memory.

In flash memory, many previous garbage collection methods only merge blocks statically and do not consider the contents of buffer. These schemes may cause more unnecessary block erase operations and page copy operations. However, since flash memory has the limitation of maximum rate and life cycle to delete each block, an efficient garbage collection method to evenly wear out the flash memory region is needed. This paper proposes a memory compaction scheme based on block-level buffer for flash memory. The proposed scheme not only merges the data blocks and the corresponding log block, but also searches for the block-level buffer to find the corresponding buffer blocks. Consequently, unnecessary potential page copying operations and block erasure operations could be reduced, thereby improving the performance of flash memory and prolonging the lifetime of flash memory.

In an Internet of Things (IoT)-configured system, each device executes on-chip software. Recent IoT devices require fast execution time of complex services, such as analyzing a large amount of data, while maintaining low-power computation. As service complexity increases, the service requires high-performance computing and more space for embedded space. However, the low performance of IoT edge devices and their small memory size can hinder the complex and diverse operations of IoT services. In this paper, we propose a remote on-demand software code execution unit using the cloudification of on-chip code memory to accelerate the program execution of an IoT edge device with a low-performance processor. We propose a simulation approach to distribute remote code executed on the server side and on the edge side according to the program's computational and communicational needs. Our on-demand remote code execution unit simulation platform, which includes an instruction set simulator based on 16-bit ARM Thumb instruction set architecture, successfully emulates the architectural behavior of on-chip flash memory, enabling embedded devices to accelerate and execute software using remote execution code in the IoT environment.

Determination of the correct point of ignition is important in the investigation. It depends on the interpretation of the situation and the fire pattern, and it is similar to the flow pattern in the flooring due to the radiation of the high temperature after the flashover. There are many difficulties in determining the ignition of the pattern, and the confusion are increased. There are bottom after Flashover, even on NFPA921 like the pour pattern of the method. It is said that the splatter of the oil pattern should not be confused with the use of actual cotton swab. Previous works on the bottom combustion pattern and the oil fire pattern after Flashover are analysed, and floating combustion pattern and the oil fire pattern are suggested.

스마트폰, 태블릿 PC, 울트라북등 휴대기기 사용의 증가로 인해 NAND-형 Flash Memory의 수요도 지속적으로 증가하고 있다. 그렇기 때문에 NAND-형 Flash Memory에서 발생할 수 있는 고장을 진단하기 위한 알고리즘 연 구가 매우 중요하다. Flash Memory는 셀 배열구조에 따라서 NOR-형 Flash Memory와 NAND-형 Flash Memory로 구분이 된다. NOR-형 Flash Memory는 다양한 테스트 알고리즘과 BIRA(Built-in Redundancy Analysis)알고리즘, 진단 알고리즘 등 다양한 알고리즘 연구가 진행되어 왔다. 그러나 NOR-형 Flash Memory에 서 연구되었던 이러한 알고리즘들을 이용하여 NAND-형 Flash Memory를 위한 고장을 확인하는 것이 매우 어렵 다. 따라서 본 논문에서는 NAND-형 Flash Memory에서 발생할 수 있는 고장 확인이 가능한 정확하고 효율적인 진단 알고리즘을 제안한다. 그래서 NAND-형 Flash Memory를 위한 고장 진단을 가능하게 한다.

The increasing usage of mobile devices such as SmartPhone, TabletPC, and Ultrabook, the demand for NAND-type Flash Memory is also constantly growing. Therefore, high speed and miniaturization able, NAND-type Flash Memory’s research for diagnosis of possible malfunction is very important. The Flash Memory is divided into NOR-type Flash Memory, and NAND-type Flash Memory. A lot of study such as Test Algorithm, BISR(Built-In Self Repair) Algorithm and Diagnostic Algorithm, etc. has been progressed in NOR-type Flash Memory. However, it is very difficult to detect for NAND-type Flash Memory’s fault, with the former studies of algorithms for NOR-type Flash Memory. In this paper, an efficient and accurate problem diagnosable algorithm that can identify, the type of possible failure of the cell in NAND-type Flash Memory, is proposed. So it allows error diagnosing for NAND-type Flash Memory.

본 연구는 시간-보간법을 적용한 FLASH analog-to-digital converter (ADC)에 관한 것이다. 시간-보간 법은 기존의 FLASH ADC에서 요구되는 전압영역 비교기의 개수를 줄일 수 있으며 이 따른 전력 소모 및 칩 면적 의 절약을 기대할 수 있다. 본 연구에서는 5-bit, 즉 31개의 양자화 레벨을 갖는 ADC를 설계 및 구현하였으며, 16개의 양자화 레벨은 기존의 전압영역 비교기 방식을 유지하고, 나머지 15개의 양자화 레벨은 시간영역 비교기 를 통하여 처리되도록 구성하여, 기존 5-bit FLASH ADC 대비 전압영역 비교기의 숫자를 48.4% 줄일 수 있었다. 시제품은 14 nm Fin Field-effect transistor (FinFET) 공정으로 제작되었으며 구현면적은 0.0024 mm2, 전력 소모는 0.8 V 전원전압에서 0.82 mW로 측정되었으며, 400 MS/s의 변환속도 21 MHz 정현파 입력에 대하여 ADC는 28.03 dB의 신호-대-잡음비 (SNDR), 즉 4.36 유효비트(ENOB)의 성능을 보였다.

A time-interpolation technique has been applied to the conventional FLASH analog-to-digital converter (ADC) to increase a number of quantization level, thus it reduces not only a power dissipation, but also minimize an active chip area. In this work, we demonstrated 5-bit ADC which has 31 quantization levels consisting of 16 conventional voltage-mode comparators and 15 time-mode comparators. As a result, we have achieved about 48.4% voltage-mode comparator reductions. The ADC is fabricated in a 14nm fin Field-effect transistor (FinFET) process with an active die area of 0.0024 mm2 while consuming 0.82 mW through a 0.8 V supply. At 400-MS/s conversion rate, the ADC performs 28.03 dB SNDR (4.36 ENOB) at 21MHz input frequency.

In this paper, computer simulation was used for researching into the estimate of flash over result from compartment fire and the characteristics operation of sprinkler RTI. Computing simulation, we analyzed and verified the path of the flame in compartment fire and the adaptation of sprinkler system concerned with sprinkler RTI.

강원도 지역은 산지지형이 많고 여름철 장마나 국지적인 집중호우에 의한 토석류의 발생빈도가 다른 지역에 비해 많다. 토석류와 돌발홍수의 발생규모는 누적강우량이 많을수록 피해규모도 커지게된다. 따라서 강우발생에 따른 돌발홍수와 토석류 발생기준에 대해서 정립할 필요가 있다. 본 연구에서는 토석류 유발강우곡선과 돌발홍수 유발 강우곡선을 산정하여 돌발홍수와 토석류유발강우량에 대해서 연구하였다. 토석류 및 산사태 유발강우량은 유효시간에 따라 강우강도, 유효 평균강우강도, 누적강우량을 산정 하였다. 그리고 돌발홍수기준도를 작성하기 위해서는 첨두홍수량, 월류홍수량, 한계유출량을 산정하여 돌발홍수곡선을 산정하였다. 마지막으로 돌발홍수 유발강우량과토석류 유발강우량을 연계하여 산정된 두 회귀 곡선을 연계하여 각 등급별로 돌발홍수와 토석류의유발강우량 FADG(Flash flood and Debris flow Guidance) 곡선을 제시하였다.

Gangwon Province in Korea is relatively vulnerable to landslides and debris flows since it is mountainous and has a climatic condition under which summer monsoon and localized heavy rain occur frequently. Scale generation of flash floods and debris flow become greater in the case of cumulative rainfall, causing a greater scale of damage. As it is critical to identify a threshold of rainfall amount triggering flash food, debris flows and landslide, this study develops a flash flood-triggering rainfall curve and a debris flow-triggering rainfall curve. Debris flows and landslides caused by rainfall were estimated as rainfall intensity, both effective average rainfall and accumulated rainfall, depending on the effective time. In order to estimate the amount of flash flood-triggering rainfall, we estimated peak discharge, overtopping discharge, and threshold runoff. As a result, by making a linkage between the flash flood-triggering rainfall and the debris flow-triggering rainfall, we presented the two rainfall regression curves, FADG (Flash flood and Debris flow Guidance) curves for flash flood and debris flow, as a tool to monitor triggering rainfall.