• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.04a
• /
• pp.83-86
• /
• 2014

분산 파일시스템의 클라이언트 측에 SSD 장치를 캐시장치로 사용하여 분산파일시스템에 부족한 랜덤 입출력 성능을 향상시키고, Back-end 데이터 서버의 부하를 줄일 수 있다. 본 논문은 국내에서 개발된 분산파일시스템인 MAHA-FS의 클라이언트 측에 읽기 캐시로 SSD 장치를 지원함으로써 캐시 히트시에 읽기 성능을 향상 시킬 수 있음과 더불어 읽기 캐시의 기능 추가로 인한 쓰기 성능의 저하가 없음을 보여준다. 본 논문에서 제안한 SSD 캐시를 이용하여 분산파일시스템의 활용 분야을 넓힐 수 있을 것으로 기대한다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.12
• /
• pp.433-442
• /
• 2014

In this paper, we introduce the client side cache of distributed file system for enhancing read performance by eliminating the network latency and decreasing the back-end storage burden. This performance enhancement can expand the fields of distributed file system to not only cloud storage service but also high performance storage service. This paper shows that the distributed file system with client side SSD cache can satisfy the requirements of VDI(Virtual Desktop Infrastructure) storage. The experimental results show that full-clone is more than 2 times faster and boot time is more than 3 times faster than NFS.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.351-352
• /
• 2014

분산 파일시스템은 서로 분산된 여러 서버들을 가지고 파일 시스템을 구성함으로써 높은 확장성과 고가용성을 지원한다. HDFS는 대용량 데이터 저장장치로 처리되고 있지만 실시간 파일 접근에 관한 고려는 부족하다. 파일을 읽을 때 네임노드와 데이터 노드는 상호 작용을 하지만 엄청난 대용량의 데이터 그리고 동시작업량이 많을 때 접근수행속가 급격하게 감소하게 된다. 따라서 실시간 클라우드 서비스 환경에서 HDFS 파일 접근 수행속도를 향상시키기 위한 연구가 이슈이다. 본 논문에서는 HDFS의 위에 분산 캐시를 둔 새로운 캐시시스템을 제안한다.

• The KIPS Transactions:PartA
• /
• v.9A no.2
• /
• pp.163-170
• /
• 2002

Previous buffer cache schemes for continuous media file system only exploited the sequentiality of continuous media accesses and didn't consider looping references. However, in some video applications like foreign language learning, users mark the scene as loop area and then application automatically playbacks the scene several times. In this paper, we propose a novel buffer cache scheme for continuous media file system that sequential and looping references exist together. Proposed scheme increases the cache hit ratio by detecting reference pattern of files and appling an appropriate replacement policy to each file.

• Journal of Internet Computing and Services
• /
• v.4 no.2
• /
• pp.11-19
• /
• 2003

The technique which stores cache data in EXT2 or UFS designed for general purpose is not suitable for satisfying the speed required for web cache due to the general purpose file system. This study shows that there is the better solution by optimizing the file system using the characteristics of web file. It is impossible that the suggested RawCFS changes the size of cached object and the access authentication, and this results from the existence of up-to-dated object in the original server. This file system is proved in the capability test that it is faster than the technique by 40% which stores in each file by object unit. This can be used in the design of high end web server such as shoppingmall or Internet Broadcasting station which should provide objects like image or HTML as well as cache server to the client for the fast service.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.8
• /
• pp.868-872
• /
• 2010

Many embedded Java systems are equipped with limited memory, which can constrain the code cache size provided for Java just-in-time compilation, affecting the Java performance. This paper proposes expanding the limited code cache when it is full, by saving the machine code for some methods in the code cache into the file system of the permanent storage and reloading it to the code cache when they are re-invoked later. This is applying the client ahead-of-time compilation during the execution time for the purpose of enlarging the code cache. Our experimental results indicate that the proposed execution method can improve the performance by as much as 1.6 times compared to the conventional method, when the code cache size is reduced by half.

• The Journal of the Korea Contents Association
• /
• v.14 no.11
• /
• pp.28-38
• /
• 2014

In this paper, we propose the distributed cache management scheme to efficiently access small files in Hadoop Distributed File Systems(HDFS). The proposed scheme can reduce the number of metadata managed by a name node since many small files are merged and stored in a chunk. It is also possible to reduce the file access costs, by keeping the information of requested files using the client cache and data node caches. The client cache keeps small files that a user requests and metadata. Each data node cache keeps the small files that are frequently requested by users. It is shown through performance evaluation that the proposed scheme significantly reduces the processing time over the existing scheme.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06b
• /
• pp.366-370
• /
• 2008

NAND 플래시 메모리는 특성상 쓰기 횟수가 제한적이라는 단점을 가지고 있어 쓰기 연산이 빈번히 발생하게 되면 NAND 플래시 메모리의 수명이 줄어든다. 이러한 문제점을 해결하기 위해 NAND 플래시 메모리의 특성을 고려한 지연 쓰기 기법이 연구되고 있다. 하지만 지연 쓰기를 하기 때문에 쓰기 횟수는 줄어들지만 캐시 적중률이 낮아진다. 이러한 문제해결을 위해 본 논문에서는 NAND 플래시 메모리 기반 파일 시스템을 위한 더블 캐시 정책을 제안한다. 더블 캐시는 실질적인 캐시인 Real Cache와 요구 페이지의 패턴을 관찰하기 위한 Ghost Cache로 구성된다. 이 정책은 Real Cache에서의 지연 쓰기를 하지 않고, Ghost Cache 공간에서 dirty페이지와 clean페이지를 활용하여 효율적인 지연 쓰기가 가능하도록 설계함으로써 쓰기 횟수를 줄이고, 적중률을 높인다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.2
• /
• pp.247-251
• /
• 2010

Using SCM in storage systems introduce new potentials for improving I/O performance and reliability. In this paper, we study the use of SCM as a buffer cache that guarantees transactional unit writes. Our proposed method can improve storage system reliability and performance at the same time and can recover the storage system immediately upon a system crash. The Proposed method is based on the LINUX JBD(Journaling Block Device), thus reliability is equivalent to JBD. In our experiments, the file system that adopts our method shows better I/O performance even while guaranteeing high reliability and shows fast file system recovery time (about 0.2 seconds).

• Journal of KIISE
• /
• v.43 no.5
• /
• pp.524-530
• /
• 2016

The desire to access data faster and the growth of next-generation memories such as non-volatile memories, contribute to the development of research on memory file systems. It is recommended that memory mapped file I/O, which has less overhead than read-write I/O, is utilized in a high-performance memory file system. Memory mapped file I/O, however, brings a page table overhead, which becomes one of the big overheads that needs to be resolved in the entire file I/O performance. We find that same overheads occur unnecessarily, because a page table of a file is removed whenever a file is opened after being closed. To remove the duplicated overhead, we propose the mapping cache, a technique that does not delete a page table of a file but saves the page table to be reused when the mapping of the file is released. We demonstrate that mapping cache improves the performance of traditional file I/O by 2.8x and web server performance by 12%.