• Journal of Internet Computing and Services
• /
• v.13 no.3
• /
• pp.83-91
• /
• 2012

Recently, column-oriented storage has become a progressive model for high-speed database systems because of its superior I/O performance. In this paper, we analysis traditional raw-oriented storage model and then propose a new column-aware storage management model using flash memory drive and assist drive to improve the effective performance of the high-speed column-oriented database system. Our storage management scheme called column-aware polarization improves the performance of update operation by dividing and compressing table columns into active-columns or inactive-columns, and balancing congested update operations using a assist drive in high workload periods. The results obtained from experimental tests show that our scheme improves the update throughput of column-oriented storage by 19 percent, and the response time by up to 49 percent.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.7 no.1
• /
• pp.28-35
• /
• 2007

This paper presents a high-speed flash analog-to-digital converter (ADC) for ultra wide band (UWB) receivers. In this flash ADC, the interpolating technique is adopted to reduce the number of the amplifiers and a linear and wide-bandwidth interpolating amplifier is presented. For this ADC, the transistor size for the cascaded stages is inversely scaled to improve the trade-off in bandwidth and power consumption. The active inductor peaking technique is also employed in the pre-amplifiers of comparators and the track-and-hold circuit to enhance the bandwidth. Furthermore, a digital-to-analog converter (DAC) is embedded for the sake of measurements. This chip has been fabricated in $0.13{\mu}m$ 1P8M CMOS process and the total power consumption is 113mW with 1V supply voltage. The ADC achieves 4-bit effective number of bits (ENOB) for input signal of 200MHz at 5-GSample/sec.

• Journal of information and communication convergence engineering
• /
• v.8 no.4
• /
• pp.377-382
• /
• 2010

With the advent of flash memory based new storage device (SSD), there is considerable interest within the computer industry in using flash memory based storage devices for many different types of application. The dynamic index structure of large text collections has been a primary issue in the Information Retrieval Applications among them. Previous studies have proven the three approaches to be effective: In- Place, merge-based index structure and a combination of both. The above-mentioned strategies have been researched with the traditional storage device (HDD) which has a constraint on how keep the contiguity of dynamic data. However, in case of the new storage device, we don' have any constraint contiguity problems due to its low access latency time. But, although the new storage device has superiority such as low access latency and improved I/O throughput speeds, it is still not well suited for traditional dynamic index structures because of the poor random write throughput in practical systems. Therefore, using the experimental performance evaluation of various index maintenance schemes on the new storage device, we propose an efficient index structure for new storage device that improves significantly the index maintenance speed without degradation of query performance.

• KIISE Transactions on Computing Practices
• /
• v.24 no.3
• /
• pp.129-137
• /
• 2018

NVMe(Non-Volatile Memory Express) SSD(Solid State Drive) is a high-performance storage that makes use of flash memory as a storage cell, PCIe as an interface and NVMe as a protocol on the interface. It supports multiple I/O queues which makes it feasible to process parallel-I/Os on multi-core environments and to provide higher bandwidth than SATA SSDs. Hence, NVMe SSD is considered as a next generation-storage for data-center and cloud computing system. However, in the virtualization system, the performance of NVMe SSD is not fully utilized due to the bottleneck of the software I/O stack. Especially, when it uses I/O stack of the hypervisor or the host operating system like Xen and KVM, I/O performance degrades seriously due to doubled-I/O stack between host and virtual machine. In this paper, we propose a new I/O engine, called Direct-AIO (Direct-Asynchronous I/O) engine, that can access NVMe SSD directly for I/O performance improvements on QEMU emulator. We develop our proposed I/O engine and analyze I/O performance differences between the existed I/O engine and Direct-AIO engine.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2009.11a
• /
• pp.30.1-30.1
• /
• 2009

정보화가 급속히 진전됨에 따라 보다 많은 양의 정보를 전송, 처리, 저장하게 되면서 이를 위해 대용량, 고속, 비휘발성의 특징을 갖는 차세대 메모리의 개발이 절실히 요구되고있다. 이 중 저항 변화 메모리(ReRAM)는 일반적으로 TiO2, Al2O3, NiO2, HfO2, ZrO2 등의 전이금속산화물을 이용한 MIM 구조로서 적당한 전기 신호를 가하면 저항이 높아서 전도되지 않는 상태(Offstate)에서 저항이 낮아져 전도가 가능한 상태(On state)로 바뀌는 메모리 특성을가진다. ReRAM은 비휘발성 메모리이며 종래의 비휘발성 기억소자인 Flash memory 보다 access time 이105 배 이상 빠르고, 5V 이하의 낮은 전압에서도 동작이 가능하다. 또한 구조가 간단하여 공정 단순화가 가능하고 소자의 집적화도 쉽다는 점 등 많은 장점들이 있어서 Flash memory를 대체할 수 있는 유력한 후보로 여겨지고 있다. 본연구에서는 DC-magnetron Sputtering 방법으로 전이금속 박막을 증착하고, Dry furnace로 산화시켜 전이금속산화물 박막을 제작한 후 저항변화 특성을 연구하였다. 두 개의 전이금속산화물 박막을 dual-layer로 형성시켜 저항변화특성을 관찰하였으며 또한, 전이금속산화물 박막의 조성을 달리 하여 저항변화를 관찰 하였다. 전이금속산화물 박막의 전기적 특성을 알아보기 위해 Si(100) wafer 위에 Pt를 이용 MIM 형태로 capacitor 시편을 제작 하여, probe station으로 I-V 측정을 하였고 조성 및 표면 분석을 위해서는 AES와 AFM을, 미세구조를 분석을 위해서는 TEM과 SEM 을 사용하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.5
• /
• pp.2449-2456
• /
• 2013

The column-oriented database storage is a very advanced model for large-volume data analysis systems because of its superior I/O performance. Traditional data storages exploit row-oriented storage where the attributes of a record are placed contiguously in hard disk for fast write operations. However, for search-mostly datawarehouse systems, column-oriented storage has become a more proper model because of its superior read performance. Recently, solid state drive using MLC flash memory is largely recognized as the preferred storage media for high-speed data analysis systems. In this paper, we introduce fast column-oriented data storage model and then propose a new storage management scheme using a cross compressed replication for the high-speed column-oriented datawarehouse system. Our storage management scheme which is based on two MLC SSD achieves superior performance and reliability by the cross replication of the uncompressed segment and the compressed segment under high workloads of CPU and I/O. Based on the results of the performance evaluation, we conclude that our storage management scheme outperforms the traditional scheme in the respect of update throughput and response time of the column segments.

• Archives of design research
• /
• v.14 no.4
• /
• pp.217-226
• /
• 2001

In 21st century, with the rapid development of Information technologies, a new form of postgraduate courses on multimedia has emerged. The curriculum of that course is consisted of a combination of design related fields and computer science. In this thesis, I analyzed problems of design education come from multimedia departments of 2-year colleges. This form of departments result in the reduction of design related courses and that is also reduction of basic design rather than multimedia. But, for the desirable multimedia design education, the functional aspects of multimedia must be harmonized with the artistic aspects. For these reason, I developed a course of study on multimedia design. It consists of theoretical studies about design principles, actual training on flash application, and individual projects accomplishments. And on the basis of this results, case studies of 3 subjects were done in a semester.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.10
• /
• pp.1167-1175
• /
• 2016

This paper proposes an efficient page-level mapping algorithm that reduces the erase count in the FTL for flash memory systems. By maintaining the weight for each write request in the request buffer, the proposed algorithm estimates the degree of temporal locality for each incoming write request. To exploit temporal locality deliberately for determination of hot request, the degree of temporal locality should be much higher than the reference point determined experimentally. While previous LRU algorithm treats a new write request to have high temporal locality, the proposed algorithm allows write requests that are estimated to have high temporal locality to access hot blocks to store hot data intensively. The pages are more frequently updated in hot blocks than warm blocks. A hot block that has most of invalid pages is always selected as victim block at Garbage Collection, which results in delayed erase operation and in reduced erase count. Experimental results show that erase count is reduced by 9.3% for real I/O workloads, when compared to the previous LRU algorithm.

• Journal of the Korean Society of Safety
• /
• v.14 no.2
• /
• pp.26-31
• /
• 1999

In a compartment fire, convective heat transfer dominates spread of the fire and smoke movement before flash-over occurs, and natural convection is very important in particular when there are no openings. The finite volume method with SIMPLE algorithm was applied to a square cavity similar to a compartment without an opening. The objectives of this study are to evaluate the method and to simulate natural convection from a hot body in the cavity. The results without the hot body showed an excellent agreement with those of previous studies. Streamlines, isotherms and Nusselt numbers were computed for different Rayleigh numbers.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10a
• /
• pp.351-355
• /
• 2006

최근 플래시 메모리는 여러 장점들 때문에 다양한 휴대기기에서 많이 사용되고 있다. 반면 내구성에 약점을 갖고 있는 플래시 메모리의 특성 때문에 최대한 소거 동작을 적게 하여 오랫동안 사용하는 FTL 알고리즘을 개발하는 연구가 필요하다. 이러한 FTL 알고리즘을 실험하고 평가하기 위해서 트레이스 데이터를 연구에 활용하는 일이 많아지면서, 쉽게 트레이스 데이터에 대한 분석도구를 개발하였다. 우리는 트레이스 데이터를 다양한 그래프로 그려주고 통계치를 산출해주는 도구를 개발하였고, 이를 바탕으로 트레이스 분석 작업을 쉽게 할 수 있도록 하였다. 마찬가지로 이러한 도구는 버퍼 교체정책을 실험하고 평가하는 일에도 사용 될수 있다. 그리고 각 그래프를 설명하면서 트레이스에 데이터 대한 설명과 함께 분석을 통하여 버퍼교체 알고리즘 및 FTL 알고리즘에 어떻게 활용 할 수 있는지 설명하였다.