• The KIPS Transactions:PartC
• /
• v.8C no.1
• /
• pp.97-104
• /
• 2001

In this paper, we present a new hierarchical model for performance analysis of channel allocation and packet service protocol in wireless n network. The proposed hierarchical model consists of two parts : upper and lower layer models. The upper layer model is the structure state model representing the state of the channel allocation and call service. The lower layer model, which captures the performance of the system within a given structure state, is the wireless packet retransmission protocol model. These models are developed using SRN which is an modeling tool. SRN, an extension of stochastic Petri net, provides compact modeling facilities for system analysis. To get the performance index, appropriate reward rates are assigned to its SRN. Fixed point iteration is used to determine the model parameters that are not available directly as input. That is, the call service time of the upper model can be obtained by packet delay in the lower model, and the packet generation rates of the lower model come from call generation rates of the upper model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.179-183
• /
• 2008

연속적이고 무한히 발생되는 데이터 스트림을 관리하는 데이터 스트림 관리시스템(DSMS)은 연속질의를 이용하여 스트림을 처리한다. 연속질의는 질의 별로 독립적인 큐를 유지하기 때문에 질의 개수가 증가함에 따라서 메모리 비용이 증가되며, 잦은 메모리 할당으로 인한 시스템의 성능 저하를 갖는다. 이러한 문제점을 해결하기 위한 기존의 연구로 메모리 풀을 이용한 메모리 관리 기법이 있다. 하지만 페이지의 크기가 고정되어 있기 때문에 각 질의마다 필요로 하는 데이터 스트림의 최적의 크기에 적합하게 할당되지 못하여 메모리를 낭비하는 문제점이 있다. 본 논문에서는 이러한 문제를 해결하기 위해 연속질의 처리를 위한 큐 메모리 관리 기법을 제안한다. 제안기법은 큐 관리 테이블에서 관리하는 각각의 큐 메모리들을 타임스탬프를 가지고 일정한 기간을 주기로 큐 메모리의 사용량을 분석한다. 분석된 큐 메모리들은 이전의 큐 메모리의 사용량과 현재 사용된 큐 메모리의 사용량을 비교함으로써 상한 값과 하한 값을 구함으로써 현재 큐 메모리에서 가지고 있는 사용량을 추가할 것인지, 줄일 것인지를 판단하여, 메모리의 사용량을 최적화 함으로써 시스템의 메모리 가용성을 향상한다. 제안 기법은 성능평가를 통해 메모리의 가용성이 기존의 방식에 비하여 향상된 성능을 보인다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.6
• /
• pp.293-303
• /
• 2008

As the computing environment moves to the wireless and handheld system, the power efficiency is getting more important. That is the case especially in the embedded hand-held system and the power consumed by the memory system takes the second largest portion in overall. To save energy consumed in the memory system we can utilize low power mode of SDRAM. In the case of RDRAM, nap mode consumes less than 5% of the power consumed in active or standby mode. However hardware controller itself can't use this facility efficiently unless the operating system cooperates. In this paper we focus on how to minimize the number of active units of SDRAM. The operating system allocates its physical pages so that only a few units of SDRAM need to be activated and the unnecessary SDRAM can be put into nap mode. This work can be considered as a generalized and system-wide version of PAVM(Power-Aware Virtual Memory) research. We take all the physical memory into account, especially buffer cache, which takes an half of total memory usage on average. Because of the portion of buffer cache and its importance, PAVM approach cannot be robust without taking the buffer cache into account. In this paper, we analyze the RAM usage and propose power-aware page allocation policy. Especially the pages mapped into the process' address space and the buffer cache pages are considered. The relationship and interactions of these two kinds of pages are analyzed and exploited for energy saving.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.197-200
• /
• 2007

Dynamic power management reduces the power consumption of the system by switching system components into different power states, which have different power consumption levels. The main function of a power management is to decide when to perform state transitions. In this paper, a power management model based on stochastic processes is introduced. This model is developed using SRN (Stochastic Reward Nets), which has facilities to represent system queue and various modeling functions.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10c
• /
• pp.525-527
• /
• 1999

본 논문은 사이버 쇼핑몰의 일반적인 제약사항으로 모든 사용자들이 ID 등록을 해야만 물품 구매를 할 수 있는 불편을 해소하였다. 본 논문은 이러한 불편을 해소하고자 사이버 쇼핑몰 사용자 ID 등록 절차 대신에 각 사용자를 구분하는 카운트를 사용한다. 이 카운트를 상호 배타적으로 할당하는 것을 보장하기 위한 분산 상호 배제 록 알고리즘(Distributed Mutual Exclusion Lock Algorithm)을 사용한다. 록 알고리즘의 정확한 구현으로 사용자들이 모든 사이버 쇼핑몰에 자신의 ID 등록을 통해서 물품 구매를 하는 불편함을 해결할 수 있다. 록 알고리즘은 사용자가 웹 페이지에 접속하게 되면 각 사용자를 구분하는 카운트 값을 배타적으로 할당하게 된다. 이러한 기능을 실제 웹 환경에서 설계 및 구현되었다. 또한 실제 웹 환경에서 설계 및 구현된 내용을 실험하였다. 실험 결과 각 사용자들에 대한 배타적인 카운트 값이 할당됨을 확인할 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.6
• /
• pp.1289-1300
• /
• 2011

In recent computer systems with high-performance, users execute programs needing large memory and programs intensively accessing files simultaneously. Such a large memory requirement makes virtual memory systems access swap spaces in disk, and intensive file accesses require file systems to access file system partitions in disk. Executing the two kinds of programs at once incurs large disk seeks between swap spaces and file system partitions frequently. To solve the problem, this paper proposes a new scheme called SAF to create several swap spaces in a file system partition, where pages to be paged out are stored. When a page is paged out, the scheme stores the page to one of the swap spaces close to a disk location where the most recently accessed file is located. The chosen swap space in the file system partition is closer to the disk location than the traditional swap space, so that our scheme can reduce the large disk seek time spent to move to the traditional swap space in paging out a page. The experiment of our scheme implemented in FreeBSD 6.2 shows that SAF reduces the execution time of several benchmarks over FreeBSD ranging from 14% to 42%.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.5
• /
• pp.74-82
• /
• 2011

SSD has a limitation of number of erase/write cycles and does not allow in-place update unlike the hard disk because SSD is composed of an array of NAND flash memory. Thus, FTL is used to effectively manage SSD of having different characteristics from traditional disk. FTL has page, block, log-block mapping method. Among then, when log-block mapping method such as BAST and FAST is used, the performance of SSD is degraded because frequent merge operations cause lots of pages to be copied and deleted. This paper proposes a data allocation and replacement method based on access frequency by allocating PRAM as checking area of access frequency, log blocks, storing region of hot data in SSD. The proposed method can enhance the performance and lifetime of SSD by storing cold data to flash memory and storing log blocks and frequently accessed data to PRAM and then reducing merge and erase operations. Besides, a data replacement method is used to increase utilization of PRAM which has limitation of capacity. The experimental results show that the ratio of erase operations of the proposed method is 46%, 38% smaller than those of BAST and FAST and the write performance of the proposed method is 34%, 19% higher than those of BAST and FAST, and the read performance of the proposed method is 5%, 3% higher than those of BAST and FAST, respectively.

• Journal of KIISE:Databases
• /
• v.36 no.5
• /
• pp.374-385
• /
• 2009

Flash memory becomes increasingly popular as data storage for various devices because of its versatile features such as non-volatility, light weight, low power consumption, and shock resistance. Flash memory, however, has some distinct characteristics that make today's disk-based database technology unsuitable, such as no in-place update and the asymmetric speed of read and write operations. As a result, most traditional disk-based database systems may not provide the best attainable performance on flash memory. To maximize the database performance on flash memory, some approaches have been proposed where only the changes made to the database, i.e., logs, are written to another empty place that has born erased in advance. In this paper, we propose an efficient log management scheme for flash-based database systems. Unlike the previous approaches, the proposed approach stores logs in specially allocated blocks, called log blocks. By evenly distributing logs across log blocks, the proposed approach can significantly reduce the number of write and erase operations. Our performance evaluation shows that the proposed approaches can improve the overall system performance by reducing the number of write and erase operation compared to the previous ones.

• Journal of Advanced Navigation Technology
• /
• v.14 no.5
• /
• pp.767-774
• /
• 2010

Due to advantages of NAND flash memory such as non-volatility, low access latency, low energy consumption, light weight, small size and shock resistance, it has become a better alternative over traditional magnetic disk drives, and has been widely used. Traditional DBMSs including mobile DBMSs may run on flash memory without any modification by using Flash Translation Layer (FTL), which emulates a random access block device to hide the characteristics of flash memory such as "erase-before-update". However, most existing FTLs are optimized for file systems, not for DBMSs, and traditional DBMSs are not aware of them. Also, traditional DBMSs do not consider the characteristics of flash memory. In this paper, we propose a flash-conscious storage system for DBMSs that utilizes flash memory as a main storage medium, and carefully put the characteristics of flash memory into considerations. The proposed flash-conscious storage system exploits log records to avoid costly update operations. It is shown that the proposed storage system outperforms the state.

• The Journal of Society for e-Business Studies
• /
• v.8 no.2
• /
• pp.77-98
• /
• 2003

A web-based information system, that is a dominant type of information systems, suffers from the "web crisis" in development and maintenance of the system. To cope with the problem, a technology of software clustering to web-based application, which is one of web engineering, is strongly needed. In this paper, we propose a Form Clustering Algorithm along with an application example, which are used for internal-system reengineering to web-based information system. A Form Clustering Algorithm focuses on Page-model which is the feature of the web among the various web-based information system's structural model. Specially, we applying distance matrix to navigation model of graph form for easily analyzing, and web log analysis for identifying core function object that have a highly loading. Also, we create web software structure that can be used to maximize reusability and assign hardware effectively through 2-phase clustering step. Form Clustering Algorithm might be used at web-based information system development and maintenance for reusable web component development and hardware assignment, respectively.