• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.12
• /
• pp.1476-1484
• /
• 1999

TV 방송 분야에서 다양한 정보와 상호 작용성을 제공하기 위해서 최근 기존 방송 내용인 A/V 스트림 외 부가정보 방송이 시도되고 있다. 데이타 방송에 대한 기존 연구는 대부분 고정된 내용의 데이타를 방송하는 환경을 가정하고 있어서 그 결과가 방송 내용의 변화가 많은 환경에 부적합하다. 본 논문에서는 데이타에 대한 접근이 반복되지 않을 가능성이 높고 사용자 접근 확률을 예상하기 어려운 상황에서 응답 시간을 개선하는 방안으로 수신 데이타를 무조건 캐쉬에 반입하고 교체가 필요한 경우 다음 방송 시각이 가장 가까운 페이지를 축출하는 사용자 단말 시스템에서의 캐슁 정책을 제안하였다. 제안된 캐쉬 관리 정책은 평균적인 캐쉬 접근 실패 비용을 줄임으로써 사용자 응답 시간을 개선하며, 서로 다른 스케줄링 기법을 사용하는 다양한 방송 제공자가 공존하는 환경에서 보편적으로 효과를 가져올 수 있다.Abstract Recently, many television broadcasters have tried to disseminate digital multimedia data in addition to the traditional content (audio-visual stream). The broadcast data need to be cached by a client system, to provide a reasonable response time for a user request. Previous studies assumed the dissemination of a fixed set of items, and the results are not suitable when broadcast items are frequently changed. In this paper, we propose a novel cache management scheme that chooses the replacement victim based on the remaining time to the next broadcast instance. The proposed scheme reduces response time, where it is hard to predict the probability distribution of user accesses. The caching policy we present here significantly reduces expected response time by minimizing expected cache miss penalty, and can be applied without difficulty to different scheduling algorithms.

• Journal of KIISE:Databases
• /
• v.29 no.2
• /
• pp.148-157
• /
• 2002

Caching in meta-search engines can improve the response time of users' request. We describe the cache scheme in our meta-search engine in terms of its architecture and operational flow. In particular, we propose a popularity-driven cache algorithm that utilizes popularities of queries to determine cached data to be purged. The popularity is a value that represents the normalized occurrence frequency of user queries. This paper presents how to collect popular queries and how to calculate query popularities. An empirical performance evaluation of the popularity-driven caching with the traditional schemes (i.e., least recently used (LRU) and least frequently used (LFU)) has been carried out on a collection of real data. In almost all cases, the proposed replacement policy outperforms LRU and LFU.

• Journal of KIISE:Databases
• /
• v.36 no.1
• /
• pp.44-49
• /
• 2009

In recent years, with the development of portable terminals, various searching services on large data have been provided in portable terminals. In order to search large data, most applications for information retrieval use indexes such as B-trees or R-trees. However, only a small portion of the data set is accessed by users, and the access frequencies of each data are not uniform. The existing indexes such as B-trees or R-trees do not consider the properties of the skewed access patterns. And a cache stores the frequently accessed data for fast access in memory. But the size of memory used in the cache is restricted. In this paper, we propose a new index based on disk, called J-tree, which considers user's search patterns. The proposed index is a balanced tree which guarantees uniform searching time on all data. It also supports fast searching time on the frequently accessed data. Our experiments show the effectiveness of our proposed index under various settings.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.11
• /
• pp.643-657
• /
• 2004

The chordal ring based CC-NUMA system contains many links to transmit transactions between a local node and a remote node because of its structural characteristics. However, the inclination that the transactions concentrate on the ring link increases both the traffic of the ring link and the response time, which degrades the overall performance of the chordal ring based CC-NUMA system. In this paper we suggest a new remote cache replacement policy that considers both the number of total links and the number of ring links to traverse for the transactions. Our proposed replacement policy can balance data between the ring link and the chordal link properly because it reflects the characteristics of chordal ring based CC-NUMA system well.

• The KIPS Transactions:PartD
• /
• v.8D no.3
• /
• pp.221-232
• /
• 2001

Due to rapid advance of wireless communication technology, demand on data services in mobile environment is gradually increasing. Caching at a mobile client could reduce bandwidth consumption and query response time, and yet a mobile client must maintain cache consistency. It could be efficient for the server to broadcast a periodic cache invalidation report for cache consistency in a cell. In case that long period of disconnection prevents a mobile client from checking validity of its cache based solely on the invalidation report received, the mobile client could request the server to check cache validity. In doing so, some schemes may be more efficient than others depending on the number of available channels and the mobile clients involved. In this paper, we propose new cache consistency schemes, effects, efficient especially (1) when channel capacity is enough to deal with the mobile clients involved or (2) when that is not the case, and evaluate their performance.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.8
• /
• pp.719-729
• /
• 2000

Prefetching aims at reducing memory latency by fetching, in advance, data that are likely to be requested by the processor in a near future. The effectiveness of prefetching is determined by how accurate the prediction on the needed instructions and data is. Most previous studies on prefetching were limited to proposing a particular prefetch scheme and its performance evaluation, paying little attention to theoretical aspects of prefetching. This paper focuses on the theoretical aspects of instruction prefetching. For this purpose, we propose a clairvoyant prefetch model that makes use of perfect history information. Based on this theoretical model, we analyzed upper limits on the prefetch prediction accuracies of the SPEC benchmarks. The results show that the prefetch prediction accuracy is very high when there is no cache. However, as the size of the instruction cache increases, the prefetch prediction accuracy drops drastically. For example, in the case of the spice benchmark, the prefetch prediction accuracy drops from 53% to 39% when the cache size increases from 2Kbyte to 16Kbyte (assuming 16byte block size). These results indicate that as the cache size increases, most localities are captured by the cache and that instruction prefetching based on the information extracted from the references that missed in the cache suffers from prediction inaccuracies

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10b
• /
• pp.73-75
• /
• 2004

최근에 무선 인터넷과 이동 컴퓨팅 기술이 발전하고, 휴대 전화, PDA와 같은 이동 단말기가 보편화됨에 따라 사용자의 위치 정보를 활용하는 위치 기반 서비스(LBS: Location Based Service)가 다양한 분야에서 제공되고 있다. 모바일 환경에서 위치 기반 서비스를 제공하기 위해서는 모바일 장치에서 대용량의 공간 데이타를 신속하게 처리하고 효과적으로 관리하기 위한 공간 MMDBMS(Main Memory Database Management System)가 필요하다. 이에 본 논문에서는 기존의 PC용 MMDBMS인 HSQLDB를 확장하여 모바일 장치에서 공간 데이타를 효율적으로 관리할 수 있는 공간 MMDBMS를 설계 및 구현하였다. 공간 MMDBMS는 OGC의 공간 데이타 모델을 따르며, 공간 데이타 특성에 적합한 압축 기법인 산술 코딩 기법을 제공하고, 모바일 장치에 적합한 MBR 압축 및 해슁 기법을 이용한 공간 인덱스를 지원한다. 그리고, 모바일 장치의 낮은 성능의 프로세서에서 공간 데이타 디스플레이 기능을 제공하고, 모바일 장치와 디스크 기반 GIS인 ZEUS 사이에서 공간 데이타 import/export의 성능 향상을 위한 데이타 캐쉬 기능을 지원한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.2
• /
• pp.288-299
• /
• 1993

This paper proposes the new bus protocol in the tightly coupled multiprocessor system. The bus protocol uses the pipelined data transfer and block transfer scheme to increase the bus bandwidth, The bus also has the independent transfer lines for the address and data respectively, and it can transfer the data up to maximum 264 Mbytes /sec. This paper also models the multiprocessor system where each processor boards have the private cache. Simulation evaluates the bus and system performance according to hit ratio of the reference data in cache memory, In the case of using this bus, the bus is evaluated not to be saturated when up to 10 processor boards are connected to the bus. As for up to 4 memory interleavng, the performance increases linearly.

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.163-173
• /
• 2004

Recently, location-based services are becoming very Popular in mobile environments. In this paper, we propose a new association based prefetch algorithm (called by STAP) that efficiently supports information service based on the large quantity of spatial database in mobile environments. We apply the spatial-temporal relations that are meaningful for location-based queries in mobile environments. Moreover, STAP considers user's mobility and the weight of spatial data. The relation of services is a new aspect not considered in previous cache politics. So STAP is the first prefetch algorithm considering the spatial-temporal relations and thus the cache policy begins to gain a new dimension. We evaluate the performance of STAP and prove the efficiency of STAP.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10c
• /
• pp.256-258
• /
• 2002

데이타베이스 공유 시스템(Database Sharing System : DSS)은 고성능 트랜잭션 처리를 위해 다수 개의 노드들을 연동하며, 각 노드는 디스크 계층에서 데이타베이스를 공유한다. DSS를 구성하는 노드들이 고장날 경우, 데이타베이스를 정확한 상태로 복구하기 위한 회복 기법이 필요하다. DSS에서의 데이터베이스 회복 과정은 여러 노드에 분산된 로그 레코드의 병합 작업을 포함하며, 병합된 로그 레코드를 이용한 REDO 작업을 수행하여야 하므로 일반적인 단일 데이타베이스 시스템에 비해 많은 시간이 소요된다. 본 논문에서는 Oracle 9i에서 개발된 캐쉬 연합(cache fusion) 기법을 개선한 버전 관리 기법을 제안한다. 제안한 기법은 DSS를 구성하는 단일 노드의 고장 시 로그 병합 과정이 필요 없으므로 빠른 회복을 지원할 수 있으며, Oracle 9i에서 발생하는 빈번한 디스크 저장 오버헤드를 줄일 수 있다는 장점을 갖는다.