• Proceedings of the Korean Information Science Society Conference
• /
• 2010.11a
• /
• pp.165-166
• /
• 2010

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.4
• /
• pp.543-553
• /
• 1995

While RAID 5 disk arrays offer advantages of performance and reliability on wide variety of applications, they suffer from the performance degradation with write jobs. To reduce the effect of this problem, parity logging scheme was suggested. This scheme reduces disk busy time by transforming many small random accesses into large sequential accesses. This paper presents a new parity logging technique which prevents parity update image from redundant storing in the disk buffer and reduces parity maintenance overhead. A analytical model and several simulation results of our proposed scheme are presented.

• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.15-27
• /
• 2003

Recently, the volume of data is increasing rapidly in server for multimedia service, according to development of multimedia application environment. In recent research for storage technology the technology like of the SAN(Storage Area Network) advantages in scalibility of storage devices, and can read data from multiple disk arrays through RAID 0, 5. The RAID 0 and 5 translate to logical address to physical address using equation, but in case of adding disks at the system with equation -based mapping, the problem that we must rearrange the whole data in the previous disks happens. We use the mapping table to solve this problem in recent, but we can not load the whole mapping table in main memory because it occupies too large space. Therefore the extra I/Os are demanded to evaluate real physical address of data, so total performance of the system is degraded. In this paper, we propose the mapping method that supports the scalibility in RAID 0 or 5 system. The proposing method applies small metadata, so- called SZIT and simple equation, so it is possible that we make translate logical address to physical address rapidly and it is scalable in disk extending simultaneously Our suggesting method, if we add disks to the striping system for expanding of storage capacity, has an advantage of never stop service. So, SZlT-based mapping method can do online-disk-expanding in real-time service.

• International Journal of Contents
• /
• v.10 no.1
• /
• pp.62-67
• /
• 2014

Usually, cloud storage systems are developed based on DFS (Distributed File System) for scalability and reliability reasons. DFSs are designed to improve throughput than IO response time, and therefore, they are appropriate for batch processing jobs. Recently, cloud storage systems have been used for update intensive applications such as OLTP and so on. However, in DFSs, in-place update operations are not carefully considered. Therefore, when updates are frequent, I/O performance of DFSs are degraded significantly. DFSs with RAID techniques have been proposed to improve their performance and reliability. Their performance degradation caused by frequent update operations can be more significant. In this paper, we propose an in-place update method for DFS RAID exploiting a differential logging technique. The proposed method reduces the I/O costs, network traffic and XOR operation costs for RAID. We demonstrate the efficiency of our proposed in-place update method through various experiments.

• Journal of The Korean Association of Information Education
• /
• v.3 no.1
• /
• pp.22-32
• /
• 1999

In this paper, a von system which is based on current computer technologies is implemented. This von system organizes many personal computers in a local network into a configuration server, storage servers, and clients. To provide reliability, the data is stored using software RAID technology. Also, multimedia services are client-based for efficiency. The software components of this von system are the von configuration software, the von configuration service software, the von management software, and the von service software. Only the von service software runs in the client, while all the others run in the configuration server. The operating system on which the configuration server is run is the Windows NT such that the software executes like the UNIX-daemon process in the Windows NT service manager.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.2
• /
• pp.305-311
• /
• 2010

iATA (Internet Advanced Technology Attachment) is a block-level protocol developed to transfer ATA commands over TCP/IP network, as an alternative network storage solution to address insufficient storage problem in mobile devices. This paper employs RAID5 distributed storage servers concept into iATA, in which the idea behind is to combine several machines with relatively inexpensive disk drives into a server array that works as a single virtual storage device, thus increasing the reliability and speed of operations. In the case of one machine failed, the server array will not destroy immediately but able to function in a degradation mode. Meanwhile, information can be easily recovered by using boolean exclusive OR (XOR) logical function with the bit information on the remaining machines. We perform I/O measurement and benchmark tool result indicates that additional fault tolerance feature does not delay read/write operations with reasonable file size ranged in 4KB-2MB, yet higher data integrity objective is achieved.

• Proceedings of the IEEK Conference
• /
• 2000.06c
• /
• pp.47-50
• /
• 2000

In this paper, we propose a new destage algorithms, called the Dynamic Adaptive Threshold which determines turn-on and turn-off thresholds dynamically depending on the current write cache occupancy level and the differential rate of the host write requests. For performance evaluation, the proposed algorithm is compared with the wellknown High-Low Water Mark (HLWM) algorithm. Performance tests are fulfilled with our cached RAID 5 simulator. The simulation results show that the proposed algorithm outperforms the HLWM algorithm in terms of response time of host reads and write cache hit ratio under various workloads.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04a
• /
• pp.4-6
• /
• 2004

클러스터 컴퓨터에서 접속된 모든 노드들의 디스크들을 통합 사용하기 위한 SIOS의 구현은 사용자 레벨과 파일 시스템 레벨, 그리고 디바이스 드라이버 레벨로 분류할 수 있다. 본 연구에서 제안하는 방법은 현재 공개되어 있는 소프트웨어 라이브러리만을 이용하여 리눅스 클러스터에서 SIOS를 구현하는 방법으로서, 확장 네트워크 블록 디바이스(ENBD: Enhanced Network Block Device)를 이용한 디바이스 드라이버 레벨의 하위 계층과 S/W RAID 및 NFS를 이용한 파일 시스템 레벨의 상위 계층으로 구성된다. 이 방법의 주요 장점은 현재 공개되어 있는 소프트웨어 라이브러리를 이용하기 때문에 구현이 용이하고 비용이 들지 않는다는 점이다. 그리고 하위 계층으로서 디바이스 드라이버 레벨의 ENBD를 이용하기 때문에 파일 시스템을 변경하지 않기 때문에 이전의 응용 프로그램에 대한 호환성이 높다. 또한, 상위 계층에서는 파일 시스템 레벨의 S/W RAID와 NFS를 이용함에 따라 디스크 배열 방식의 조정이 비교적 자유롭다. 또 다른 장점은 하위 계층과 상위 계층이 서로 독립적이기 때문에, 클러스터의 사용 목적에 따라 각 계층을 다양한 방법으로 변경할 수 있다는 것이다. Bonnie 벤치마크를 이용한 성능 측정 결과에 따르면, ENBD를 이용하여 RAID-5로 구성한 경우에 오버헤드가 높은 NFS를 사용했음에도 불구하고 비용이 많이 드는 다른 방법과 대등한 성능을 보였으며, 부분적으로는 더 높은 성능과 확장성을 가지는 것으로 나타났다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1553-1557
• /
• 2005

High performance and reliability of the storage system to handle a very large amount of data has been become very important. Many techniques have been applied on the various application systems to establish very large capacity storage that satisfy the requirement of high I/O speed and physical or logical failure protection. We applied RAID and LVM to construct a storage system for the global data center which needs a very reliable large capacity storage system. The storage system is successfully established and equipped on the latest Linux application server.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.55-57
• /
• 2001

본 논문에서는 최근 이슈가 되고 있는 Fibre Channel과 같은 네트워크를 이용한 공유 디스크 시스템에서 다수의 호스트가 소프웨어 RAID\ulcorner 4 흑은 5를 이용할 때, 디스크 공유에 의해 발생하는 패리티 블록 일관성 문제를 다룬다. 본 논문에서는 XDWRITE을 이용해 Software HAID에서의 일관성 문제 해결과 디스크 1/0수 감소에 대해 설명한다. 또한, 버퍼캐쉬에서 파일 블록 쓰기 발생시 원본의 복사본을 이용하여 패리티 계산을 위한 디스크 읽기의 수를 감소 시키는 방법에 대해 제안한다. 본 논문에서 제안한 방식은 기존의 방식인 보다 1.5배에서 2배 디스크 1/0의 수를 감소시키는 것을 알 수 있다.