• 제목/요약/키워드: hit ratio

검색결과 280건 처리시간 0.031초

문서의 인위적 요약과 통계적 알고리즘의 비교 및 분석 (Comparison and analysis of artificial summary and statistical algorithm of document)

  • 김유식;유준현;박순철
    • 대한전자공학회:학술대회논문집
    • /
    • 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅲ
    • /
    • pp.1255-1258
    • /
    • 2003
  • Today with the sheep of information which is produced the variety is increasing geometrical progression. To recently the internet being supplied quickly, will reach and the computer users whom it uses increase and the documents which have become digital anger are increasing. From the dissertation which it sees directness it extracts a weight with possibility work and it uses it summarizes a statistics algorithm technique and a sentence. The summary literature course which the summary and the person due to a statistics algorithm summarize an agreement ratio it compares and it compares. And being more accurate like this statistical base summary method more little more, the good hit rate is high and it proposes the document summary algorithm method which is good.

  • PDF

RAID 5의 성능향상을 위한 쓰기 캐쉬의 동적 적응 반출기법 (New Dynamic Adaptive Threshold Destage Algorithms for Cached RAID 5)

  • 이혁;최상방
    • 대한전자공학회:학술대회논문집
    • /
    • 대한전자공학회 2000년도 하계종합학술대회 논문집(3)
    • /
    • 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.

  • PDF

팔 유형 특성에 따른 소매 적합성 연구 (A Study on the Sleeve Fitness According to Characteristics of Arm Types)

  • 이정란
    • 한국의류학회지
    • /
    • 제27권2호
    • /
    • pp.219-228
    • /
    • 2003
  • The purpose of this study were to examine the characteristics of arm types proposed in the previous study and to present sleeve patterns according size and shape of 4 arm types. 31 direct and indirect items were measured. The subjects for this study were 132 females of twenties. The results were as follows. 1. All of the direct measurements showed significant difference by the arm types. especially at the front armhole part. Such a.: the items of arm gradients, armhole sizes belong to the indirect measurements made clear difference also. 2. By the discriminant analysis, 9 items including underarm length, front armhole girth, upper arm girth, degree of elbow point-wrist point discriminated arm type by the hit ratio of 90.4% Item related to shape of arm rather than size made a sharp distinction. 3. Through the examination of existing patterns and wearing tests on the subjects among 4 groups, sleeve patterns classified by the arm type were suggested.

멀티미디어 파일 시스템을 위한 효율적인 버퍼 관리 방안 (An Efficient Buffer Management Scheme Multimedia File Systems)

  • 허성관;낭종호
    • 한국정보과학회논문지:시스템및이론
    • /
    • 제26권3호
    • /
    • pp.271-281
    • /
    • 1999
  • 본 논문에서는 멀티미디어 파일 시스템을 위한 새로운 버퍼관리기법을 제안한고 이를 실제 구현을 통하여 성능을 분석하였다. 제안한 관리 기법에서는 다양한 멀티미디어 파일의 참조 특성을 반영하여 참조되는 파일에 일정량의 버퍼를 할당해주고, 새로운 데이터에대한 대치는 그 파일에게 할당된 버퍼에서만 수행되도록 하였다. 또한 이런 버퍼 할당을 데이터 소비율에 맞게 동적으로 조정할 수 있는 방법과 새로운 파일 참조를 제어하는 채택 제어 방법도 설계/구현하였다. FreeBSD를 수정하여 구현한 시스템에서의 실험에 의하면 제안한 버퍼 관리 기법이 FrddBSD의 경우보다 다양한 데이터 소비율을 필요로 하는 참조인 경우에 적은 수의 버퍼를 이용하여 더 높은 적중률(hit ratio)을 제공할 수 있음을 알 수 있었다. 이런 향상된 적중률은 멀티미디어 상영시 니터(jitter)를 줄이는데 기여할 수 있다.

네트워크 경로에 기초한 웹 캐쉬 알고리즘 (A Network-path based Web Cache Algorithm)

  • 민경훈;장혁수
    • 한국정보처리학회논문지
    • /
    • 제7권7호
    • /
    • pp.2161-2168
    • /
    • 2000
  • Since most of the existing web cache structures are static, they cannot support the dynamic request change of he current WWW users well. Users re generally using multiple programs in several different windows with rapid preference change within a relatively short period of time. We develop a network-path based algorithm. It organizes a cache according to the network path of the requested URLs and build a network cache farm where caches are logically connected with each other and each cache has its own preference over certain network paths. The algorithm has been implemented and tested in a real site. The performance results show that the new algorithm outperforms the existing static algorithms in the hit ratio and response time dramatically.

  • PDF

도산예측을 위한 유전 알고리듬 기반 이진분류기법의 개발 (A GA-based Binary Classification Method for Bankruptcy Prediction)

  • 민재형;정철우
    • 한국경영과학회지
    • /
    • 제33권2호
    • /
    • pp.1-16
    • /
    • 2008
  • The purpose of this paper is to propose a new binary classification method for predicting corporate failure based on genetic algorithm, and to validate its prediction power through empirical analysis. Establishing virtual companies representing bankrupt companies and non-bankrupt ones respectively, the proposed method measures the similarity between the virtual companies and the subject for prediction, and classifies the subject into either bankrupt or non-bankrupt one. The values of the classification variables of the virtual companies and the weights of the variables are determined by the proper model to maximize the hit ratio of training data set using genetic algorithm. In order to test the validity of the proposed method, we compare its prediction accuracy with ones of other existing methods such as multi-discriminant analysis, logistic regression, decision tree, and artificial neural network, and it is shown that the binary classification method we propose in this paper can serve as a premising alternative to the existing methods for bankruptcy prediction.

Mitigating Cache Pollution Attack in Information Centric Mobile Internet

  • Chen, Jia;Yue, Liang;Chen, Jing
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제13권11호
    • /
    • pp.5673-5691
    • /
    • 2019
  • Information centric mobile network can significantly improve the data retrieving efficiency by caching contents at mobile edge. However, the cache pollution attack can affect the data obtaining process severely by requiring unpopular contents deliberately. To tackle the problem, we design an algorithm of mitigating cache pollution attacks in information centric mobile network. Particularly, the content popularity distribution statistic is proposed to detect abnormal behavior. Then a probabilistic caching strategy based on abnormal behavior is applied to dynamically maintain the steady-state distribution for content visiting probability and achieve the purpose of defense. The experimental results show that the proposed scheme can achieve higher request hit ratio and smaller latency for false locality content pollution attack than the CacheShield approach and the baseline approach where no mitigation approach is applied.

Optimizing Caching in a Patch Streaming Multimedia-on-Demand System

  • Bulti, Dinkisa Aga;Raimond, Kumudha
    • Journal of Computing Science and Engineering
    • /
    • 제9권3호
    • /
    • pp.134-141
    • /
    • 2015
  • In on-demand multimedia streaming systems, streaming techniques are usually combined with proxy caching to obtain better performance. The patch streaming technique has no start-up latency inherent to it, but requires extra bandwidth to deliver the media data in patch streams. This paper proposes a proxy caching technique which aims at reducing the bandwidth cost of the patch streaming technique. The proposed approach determines media prefixes with high patching cost and caches the appropriate media prefix at the proxy/local server. Herein the scheme is evaluated using a synthetically generated media access workload and its performance is compared with that of the popularity and prefix-aware interval caching scheme (the prefix part) and with that of patch streaming with no caching. The bandwidth saving, hit ratio and concurrent number of clients are used to compare the performance, and the proposed scheme is found to perform better for different caching capacities of the proxy server.

A Web Cache Replacement Technique of the Divided Scope Base that Considered a Size Reference Characteristics of Web Object

  • Seok, Ko-Il
    • 한국콘텐츠학회:학술대회논문집
    • /
    • 한국콘텐츠학회 2003년도 춘계종합학술대회논문집
    • /
    • pp.335-339
    • /
    • 2003
  • We proposed a Web cache replacement technique of a divided scope base that considered a size reference characteristics of a Web object for efficient operation of a Web base system and, in this study, analyzed performance of the replacement technique that proposed it though an experiment. We analyzed a reference characteristics of size to occur by a user reference characteristics through log analysis of a Web Base system in an experiment. And we divide storage scope of a cache server as its analysis result and tested this replacement technique based n divided scope. The proposed technique has a flexibility about a change of a reference characteristics of a user. Also, experiment result, we compared it with LRU and the LRUMIN which were an existing replacement technique and confirmed an elevation of an object hit ratio.

  • PDF

Design of a DI model-based Content Addressable Memory for Asynchronous Cache

  • Battogtokh, Jigjidsuren;Cho, Kyoung-Rok
    • International Journal of Contents
    • /
    • 제5권2호
    • /
    • pp.53-58
    • /
    • 2009
  • This paper presents a novel approach in the design of a CAM for an asynchronous cache. The architecture of cache mainly consists of four units: control logics, content addressable memory, completion signal logic units and instruction memory. The pseudo-DCVSL is useful to make a completion signal which is a reference for handshake control. The proposed CAM is a very simple extension of the basic circuitry that makes a completion signal based on DI model. The cache has 2.75KB CAM for 8KB instruction memory. We designed and simulated the proposed asynchronous cache including CAM. The results show that the cache hit ratio is up to 95% based on pseudo-LRU replacement policy.