• 반도체디스플레이기술학회지
• /
• 제7권3호
• /
• pp.17-21
• /
• 2008

Application of flash memory in mobile and ubiquitous related devices is rapidly being increased due to its low price and high performance. In addition, some notebook computers currently come out into market with a SSD(Solid State Disk) instead of hard-drive based storage system. Regarding this trend, applications need to increase the storage capacity using multiple flash memory chips for larger capacity sooner or later. Flash memory based storage subsystem should resolve the performance bottleneck for writing in perspective of speed and lifetime according to its physical property. In order to make flash memory storage work with tangible performance, reclaiming of invalid regions needs to be controlled in a particular manner to decrease the number of erasures and to distribute the erasures uniformly over the whole memory space as much as possible. In this paper, we study the performance of flash memory recycling algorithms and demonstrate that the proposed algorithm shows acceptable performance for flash memory storage with multiple chips. The proposed cleaning method partitions the memory space into candidate memory regions, to be reclaimed as free, by utilizing threshold values. The proposed algorithm handles the storage system in multi-layered style. The impact of the proposed policies is evaluated through a number of experiments.

• 정보저장시스템학회논문집
• /
• 제7권1호
• /
• pp.7-12
• /
• 2011

The flash memory has been remarked as the next generation media of portable and desktop computers' storage devices. Their features include non-volatility, low power consumption, and fast access time for read operations, which are sufficient to present flash memories as major data storage components for desktop and servers. The purpose of our study is to upgrade a traditional mirroring scheme based on SSD storages due to the relatively slow or freezing characteristics of write operations, as compared to fast read operations. For this work, we propose a new storage management scheme called Memory Mirror-Switching based on traditional mirroring scheme. Our Mirror-Switching scheme improves flash operation performance by switching write-workloads from flash memory to RAM and delaying write operations to avoid freezing. Our test results show that our scheme significantly reduces the write operation delay and storage freezing.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제20권10호
• /
• pp.525-533
• /
• 2014

SCST(The generic SCSI target subsystem for Linux)는 개발자가 SCSI 타겟 스토리지를 구성할 수 있게 하는 오픈소스 소프트웨어이며, 스토리지 네트워크 구성을 위해 iSCSI, FC, SRP 등의 다양한 SCSI 네트워크 프로토콜을 지원한다. 본 논문에서는 SCST를 이용한 스토리지 가상화 방법을 제안하고 고성능 스토리지인 All Flash Array를 4Gb 파이버 채널, 10Gb 이더넷, 40Gb 인피니밴드를 통해 가상화하며, 가상화된 All Flash Array의 성능을 고찰한다. 실험 결과 40Gb 인피니밴드를 통해 가상화된 All Flash Array가 세 경우 중 가장 높은 성능을 보였다. 40Gb 인피니밴드를 통해 가상화된 All Flash Array는 SCSI 타겟 시스템에 연결된 All Flash Array에 비해 순차/임의 읽기의 경우에 각각 78%, 79%의 성능을 보였으며, 순차/임의쓰기의 경우에는 83%, 88%의 성능을 보임을 확인하였다.

• 대한임베디드공학회논문지
• /
• 제6권5호
• /
• pp.263-272
• /
• 2011

The complexity of flash memory based storage systems is high due to diverse host interfaces and other design choices such as mapping granularity, flash memory controller execution models and so on. Thus, it is possible that the actual performance after implementation is not consistent with the target performance. This paper demonstrates that the performance prediction of flash memory based storage systems is possible through performance modeling that takes into account various design parameters. In the performance modeling, the FTL, which is the core element of flash memory based storage systems, is modeled as a set of (copy-on-write) logs and their interactions. Also, the flash memory controller is modeled based on the classification proposed in the design of the Ozone flash controller. In this study, the performance model has been implemented using Simulink and experimental results are presented and analyzed.

• International Journal of Internet, Broadcasting and Communication
• /
• 제10권3호
• /
• pp.26-34
• /
• 2018

In this paper, we propose a scheduling scheme that can efficiently serve I/O requests having deadlines in flash storage. The I/O requests with deadlines, namely, real-time requests, are assumed to be issued for streaming services of continuous media. Since a Web-based streaming server commonly supports downloads of HTMLs or images, we also aim to quickly process non-real-time I/O requests, together with real-time ones. For this purpose, we adopt the well-known rate-reservation EDF (RR-EDF) algorithm for determining scheduling priorities among mixed I/O requests. In fact, for the use of an EDF-style algorithm, overhead of task's switching should be low and predictable, as with its application of CPU scheduling. In other words, the EDF algorithm is inherently unsuitable for scheduling I/O requests in HDD storage because of highly varying latency times of HDD. Unlike HDD, time for reading a block in flash storage is almost uniform with respect to its physical location. This is because flash storage has no mechanical component, differently from HDD. By capitalizing on this uniform block read time, we compute bandwidth utilization rates of real-time requests from streams. Then, the RR-EDF algorithm is applied for determining how much storage bandwidth can be assigned to non-real-time requests, while meeting deadlines of real-time requests. From this, we can improve the service times of non-real-time requests, which are issued for downloads of static files. Because the proposed scheme can expand flexibly the scheduling periods of streams, it can provide a full usage of slack times, thereby improving the overall throughput of flash storage significantly.

• 정보저장시스템학회논문집
• /
• 제6권2호
• /
• pp.41-46
• /
• 2010

Recently, Flash Memory Drives are one of the best media to support portable and desktop computers' storage devices. Their features include non-volatility, low power consumption, and fast access time for read operations, which are sufficient to present flash memories as major database storage components for desktop and server computers. However, we need to improve traditional storage management schemes based on HDD(Hard Disk Drive) and RAID(Redundant Array of Independent Disks) due to the relatively slow or freezing characteristics of write operations of SSDs, as compared to fast read operations. In order to achieve this goal, we propose a new storage management scheme called Hetero-Mirroring based on traditional HDD mirroring scheme. Hetero-Mirroring-based storage management improves RAID-1 operation performance by balancing write-workloads and delaying write operations to avoid SSD freezing.

• 정보통신설비학회논문지
• /
• 제10권4호
• /
• pp.115-120
• /
• 2011

Recently flash memory is widely used in various mobile devices as storage medium. Nonvolatile memory can be divided into two categories: NAND- and NOR-type flash memory. NOR flash memory is mainly used to store instruction codes for operation; while NAND for data storage. However, NAND does show more economical benefits, that is, it is approximately 30~40% cheaper than NOR flash. Therefore it can be useful to improve NAND flash performance by replacing NOR flash with NAND flash combining with various buffer systems.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.157-159
• /
• 2007

The conventional hard disk has been the dominant database storage system for over 25 years. Recently, hybrid systems which incorporate the advantages of flash memory into the conventional hard disks are considered to be the next dominant storage systems to support databases for desktops and server computers. Their features are satisfying the requirements like enhanced data I/O, energy consumption and reduced boot time, and they are sufficient to hybrid storage systems as major database storages. However, we need to improve traditional index node management schemes based on B-Tree due to the relatively slow characteristics of hard disk operations, as compared to flash memory. In order to achieve this goal, we propose a new index node management scheme called FNC-Tree. FNC-Tree-based index node management enhanced search and update performance by caching data objects in unused free area of flash leaf nodes to reduce slow hard disk I/Os in index access processes.

• 한국항행학회논문지
• /
• 제14권5호
• /
• pp.767-774
• /
• 2010

NAND 플래시 메모리는 높은 입출력 성능, 내장애성, 저전력 소모 등 여러 가지 장점을 가지고 있어서 하드디스크를 대체할 수 있는 새로운 저장 장치로 관심을 받고 있다. 전통적인 DBMS들은 FTL (Flash Translation Layer)를 이용하면 전혀 수정 없이 플래시 메모리 위에서 동작한다. 그러나, 대부분의 FTL은 DBMS가 아닌 파일 시스템에 최적화 되어있다. 또한, 전통적인 DBMS는 플래시 메모리의 특징 (erase-before-write) 을 고려하지 않고 있다. 이 논문에서는 플래시 메모리를 이차 저장장치로 사용하는 DBMS를 위한 플래시 고려하는 스토리지 시스템을 제안한다. 제안하는 플래시 고려 스토리지 시스템은 비용이 높은 변경 연산을 피하기 위해 로그 레코드를 이용한다. 마지막으로, 실험을 통해서 제안하는 스토리지 시스템이 기존에 제안된 플래시를 고려하는 DBMS에 비해 우수함을 보인다.

• Journal of Computing Science and Engineering
• /
• 제3권3호
• /
• pp.181-194
• /
• 2009

Flash memory is increasingly being used in a wide range of storage applications because of its low power consumption, low access latency, small form factor, and high shock resistance. However, the current platforms for flash memory software development do not meet the ever-increasing requirements of flash memory applications. This paper presents three different hardware platforms for flash memory/NVRAM (non-volatile RAM) software development that overcome the limitations of the current platforms. The three platforms target different types of host system and provide various features that facilitate the development and verification of flash memory/NVRAM software. In this paper, we also demonstrate the usefulness of the three platforms by implementing three different types of storage system (one for each platform) based on them.