• 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
• /
• 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.

• Journal of the Institute of Convergence Signal Processing
• /
• v.19 no.3
• /
• pp.125-132
• /
• 2018

KVN(Korean VLBI Network) introduced the Mark6 system of the MIT haystack observatory as a recorder for data recording up to 32 Gbps for VLBI(Very Long Baseline Interferometry) observation. The Mark6 recorder can record data at up to 32Gbps when two systems and 64 HDDs are installed. However, because of the unique recording method that is characteristic of Mark6, we are introducing a large amount of data into a virtual file system, or using a general RAID method, which causes data loss when reading and transferring files at the highest recording speed or file system have. The Flexbuff system, a software recorder developed by JIVE(Joint Institute for VLBI ERIC), can be configured to operate as a data recorder through RAID configuration and network upgrades. In particular, when installed in the Mark6 system, it can record VLBI data at a maximum speed of 32 Gbps with less loss of data compared to the existing Mark6 by utilizing Mark6 resources well. In this paper, we propose that the existing Mark6 system can be operated as Mark6-Flexbuff by installing jive5ab software, and it is verified through experiment that it can be effectively used for VLBI observation operation through data recording test.

• 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.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.10
• /
• pp.413-420
• /
• 2013

Redundant arrays of independent disks (RAID) without a power-fail-safe component in small and mid-sized business suffers from intolerably long resynchronization time after a unclean power-failure. Data blocks and a parity block in a stripe must be updated in a consistent manner, however a data block may be updated but the corresponding parity block may not be updated when a power goes off. Such a partially modified stripe must be updated with a correct parity block. However, it is difficult to find which stripe is partially updated (inconsistent). The widely-used traditional parity resynchronization manner is a intolerably long process that scans the entire volume to find and fix inconsistent stripes. This paper presents a fast resynchronization scheme with a negligible overhead for small and mid-sized RAIDs. The proposed scheme is integrated into a software RAID driver in a Linux system. According to the performance evaluation, the proposed scheme shortens the resynchronization process from 200 minutes to 5 seconds with 2% overhead for normal I/Os.

• The KIPS Transactions:PartA
• /
• v.13A no.6 s.103
• /
• pp.517-524
• /
• 2006

In cluster computers, it is essential to Implement the single I/O space(SIOS) supporting integrated I/O substructure to efficiently process I/O intensive applications. SIOS service provides with global I/O address space to directly access peripherals and hard disks in its own or remote nodes from any node in the cluster computer In this thesis, we propose the implementation method of SIOS in Linux clusters by using only freewares. This method is implemented at device driver level that uses Enhanced Network Block Device(ENBD) and file system level that uses S/W RAID and NFS. The major strengths of this method are easiness of implementation and almost no cost due to using freewares. In addition, since freewares used are open sources, it is possible to apply this method to other platforms with only slight modification. Moreover, experiments show that I/O throughputs are up to 5.5 times higher in write operations and approximately 2.3 times higher in read operations than those of CDD method that uses the device driver developed at kernel level.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.3
• /
• pp.811-820
• /
• 1997

Disk arrays are using to enhance data trandfer rate and I/O performance in multimedia applications which need a high-performance storage device with large storage capacity and high-speed network.As performance varies with configuration and data layout scheme,disk array characteristic variables must be approrpriately deter-mined in desibning disk array archetecture for a speciffic applicatoin. In this paper,in order to design a disk array architecturte as a storage server of a small-scale VOD system,we evaluate performance of a disk array to chose the number of disks in the array,disk array cinfiguration,a degree of declustering for a given data block size of continous media file system and I/D request size through simulation.Simulation result shows that RAID level 5 with 5 disks ios a suitable candidate for the disk array architecture which privides MPEG-2 files with a rate of 6 Mbps,Moreover,we whow that stripe unit is 64 KB and a layout scheme is contigous placement.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2662-2670
• /
• 1997

High speed network makes possible to transfer not only the test data but also multimedia data such as audio, image, and moving pictures and etc. In a multimedia applications, as a multimedia storage system, it is necessary to use a disk array to store and retrieve data by real time. It is important feature to various disk array parameters as a storage system on a real VOD system, such as configuration method of each disks and allocation method of multimedia data. In this paper, various parameters for the disk array are decided to be used for the video-on-demand system application by simulations in two ways. The real and simulation measurement are compared and analyzed on the performance. Simulation results shows that RAID level 5 and 256KB as striping unit and 185KB as data requests size per seconds are suitable parameter for the disk array architecture which provides MPEG-1 files with a rate of 1.5Mbps in two measurements of real and simulation.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.12-14
• /
• 1998

본 논문은 RAID5가 갖는 쓰기 문제를 극복하기 위하여 지금까지의 연구 중에서 가장 우수한 성능을 나타내는 패리티 로깅 기법의 단점을 개선한 새로운 패리티 로깅 기법을 제안한다. 제안된 기법은 패리티 로깅기법에서 가장 큰 오버헤드인 패리티 로그의 크기를 줄이기 위하여 로그 데이터를 압축하는 방법을 사용한다. 로그 데이터의 압축은 로그 데이터의 저장에 필요한 비휘발성 버퍼의 크기와 로그 데이터 접근시간을 감소시켜 전체적인 성능을 향상시킬수 있다. 시뮬레이션을 통한 성능분석 결과에서 제안된 기법은 기존의 패리티 로깅 기법에 비하여 디스크 접근시간에서 우수한 성능을 나타냄을 보인다.

• Journal of applied mathematics & informatics
• /
• v.40 no.5_6
• /
• pp.925-931
• /
• 2022

In this article, we compare the reliability and the hazard function between a baseline distribution and the corresponding transmuted-G distribution. Some examples based on existing transmuted-G distributions in literature are used. Three tests of parameter significance are utilized to test the importance of a transmuted-G distribution over the baseline distribution, and real data is used in an application of the inference about the importance of transmuted-G distributions.