公开(公告)号：KR1020110071218A

公开(公告)日：2011-06-29

申请号：KR1020090127723

申请日：2009-12-21

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  박창준 ,  류성원 ,  김항기 ,  이기석

IPC: G06Q50/10 ,  G06Q50/30

CPC classification number: G06Q50/10 ,  G06Q50/30

Abstract: PURPOSE: A dynamically updatable game server and a method thereof are provided to continuously offer a game service by dynamically updating game programs which are in executing in a plurality of servers. CONSTITUTION: A program input unit(530) inputs the new version of a game program. A program difference analyzer(540) analyzes the difference of a program source code between the new version of the game program and the executing game program. An update module generator(550) creates the updated module based on the analyzed difference. A dynamic update module replacement unit(560) replaces the update module into a specific module of the executing game program.

Abstract translation: 目的：提供一种动态更新游戏服务器及其方法，通过动态更新在多个服务器中执行的游戏程序来持续提供游戏服务。 构成：程序输入单元（530）输入游戏程序的新版本。 程序差异分析器（540）分析游戏程序的新版本和执行的游戏程序之间的程序源代码的差异。 更新模块发生器（550）基于所分析的差异创建更新的模块。 动态更新模块更换单元（560）将更新模块替换为执行游戏程序的特定模块。

公开(公告)号：KR1020020049767A

公开(公告)日：2002-06-26

申请号：KR1020000079050

申请日：2000-12-20

Applicant: 한국전자통신연구원

Inventor： 정효택 ,  김동관 ,  양영종

IPC: G06F17/30

Abstract: PURPOSE: An apparatus and a method for reconstructing a data structural according to an access frequency are provided to rapidly access to data with high access frequency by converting an existing data structural into a tree structure, and by providing a weight according to an access frequency of each data node. CONSTITUTION: A logical structure converter(11) converts indexes on data capable of performing an access through a URL in the form of a tree. A restriction condition processor(12) selects nodes and edges not to be changed in the data structural system in the form of the tree by the logical structure converter(11) and performs a grouping of the nodes and edges selected. In addition, the restriction condition processor(12) divides existing trees into sub trees. A weight setting device(13) sets a weight to each node in the sub trees divided through the processor(12). A hoffman code converter(14) converts each weight into hoffman codes and provides the codes to each node. A logical data structure generator(15) generates logical data according to hoffman code values provided through the code converter(14).

Abstract translation: 目的：提供一种根据访问频率重构数据结构的装置和方法，通过将现有数据结构转换为树结构，通过提供根据访问频率的访问频率的权重来快速访问具有高访问频率的数据 每个数据节点。 构成：逻辑结构转换器（11）通过URL以树的形式转换能够执行访问的数据的索引。 限制条件处理器（12）由逻辑结构转换器（11）选择树形式的数据结构系统中不改变的节点和边，并执行所选节点和边的分组。 此外，限制条件处理器（12）将现有树分为子树。 权重设定装置（13）对通过处理器（12）划分的子树中的每个节点设置权重。 霍夫曼代码转换器（14）将每个权重转换成霍夫曼代码并将代码提供给每个节点。 逻辑数据结构生成器（15）根据通过代码转换器（14）提供的霍夫曼代码值生成逻辑数据。

公开(公告)号：KR1020010056462A

公开(公告)日：2001-07-04

申请号：KR1019990057928

申请日：1999-12-15

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  이상덕 ,  양영종 ,  정효택

IPC: G06F17/18

Abstract: PURPOSE: An apparatus and method for reusing a previously used case for extracting demanding item of an objected-oriented system and a method for extracting demanding item are provided to reduce a time and a cost for developing a software by analyzing a functional requesting item of a software more exactly through a reuse of an existing case in a demanding item analyzing stage of a software developing process. CONSTITUTION: An actor input unit(301) receives an actor extracted from a problem area. A using case input unit(302) receives a using case newly extracted from a problem area. A using case summary description input unit(303) receives a simple explanation with respect to the extracted using case in accordance with a fixed input form. A using case detailed description input unit(304) describes an event flow of the interior of a using case. Another using case input unit(305) inputs a using case scenario in accordance with a fixed input form based on a using case description. A using case description unit(306) receives a sequence diagram, an activity diagram and a status diagram for a standardization of a using case. A using case searching unit(307) searches a similar using case in accordance with a demand by a user. A using case form managing unit(308) manages a version of a stored using case. All using cases, which are extracted or changed, are stored in a storing portion(309).

Abstract translation: 目的：提供一种用于重用先前使用的用于提取面向对象系统的要求项的情况的装置和方法以及用于提取要求项的方法，以通过分析一个功能请求项来减少开发软件的时间和成本 软件更准确地通过在软件开发过程的苛刻的项目分析阶段重新使用现有案例。 构成：演员输入单元（301）接收从问题区域提取的演员。 用例输入单元（302）接收从问题区域新提取的使用情况。 使用案例概要描述输入单元（303）根据固定的输入形式接收关于所提取的使用情况的简单说明。 使用案例详细描述输入单元（304）描述了使用案例的内部的事件流。 另一使用案例输入单元（305）基于使用案例描述根据固定的输入表单输入使用案例场景。 使用案例描述单元（306）接收用于使用案例的标准化的序列图，活动图和状态图。 使用案例搜索单元（307）根据用户的需求搜索类似的使用情况。 使用案例表单管理单元（308）管理存储的使用案例的版本。 提取或改变的所有使用情况都存储在存储部分（309）中。

公开(公告)号：KR1020000007233A

公开(公告)日：2000-02-07

申请号：KR1019980026444

申请日：1998-07-01

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  정효택 ,  양영종

IPC: G06F15/16

Abstract: PURPOSE: An apparatus and method for clustering an object are provided, which formulates a system of clustering to objects and moves automatically for performing an object distribution and a performance improvement of a target system. CONSTITUTION: An object clustering apparatus comprises: a clustering performing unit (101) for performing a clustering to a class to generate a component; a combine ratio computing unit (102) for computing a combine ratio between classes of the component generated in the clustering performing unit (101); a domain mean combine ratio computing unit (103) for computing a mean combine ratio to a domain; a combine ratio magnitude comparing unit (104) for comparing the combine ratio between classes of the combine ratio computing unit (102) with the domain mean combine ratio of the domain mean combine ratio computing unit (103); a class extracting unit (105) for extracting a class having the domain mean combine ratio; a terminal class extracting unit (106) for extracting a terminal class; a class extracting unit (107) for extracting a class; and a subdomain generating unit (108) for generating a subdomain by the component. Therefore, the automation of the object distribution is performed.

Abstract translation: 目的：提供一种用于聚类对象的装置和方法，其形成对对象的聚类系统和自动移动以执行目标系统的对象分布和性能改进。 构成：对象聚类装置包括：聚类执行单元（101），用于对类执行聚类以生成分量; 组合比计算单元（102），用于计算在聚类执行单元（101）中生成的分量的类别之间的组合比; 用于计算与域的平均组合比的域平均组合比计算单元（103）; 组合比值比较单元（104），用于将组合比计算单元（102）的类之间的组合比与域平均组合比计算单元（103）的域均值组合比进行比较; 用于提取具有域均值组合比的类的类提取单元（105）; 终端类提取单元，用于提取终端类; 用于提取类的类提取单元（107）; 以及用于通过所述部件生成子域的子域生成单元（108）。 因此，执行对象分发的自动化。

公开(公告)号：KR101349603B1

公开(公告)日：2014-01-15

申请号：KR1020100108247

申请日：2010-11-02

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  양영종 ,  전인걸 ,  김원태 ,  박승민

IPC: G06F9/44

Abstract: 실시간 시스템에서 실행중인 소프트웨어를 동적으로 업데이트 하는 방법 및 이를 위한 장치가 개시된다. 실시간 시스템에서 소프트웨어 업데이트 장치는 수신된 복수개의 주기적 태스크 및 비주기적 태스크 중 비주기적 태스크의 마감시간을 계산하고, 계산된 마감시간을 비주기적 태스크에 할당하는 마감시간 계산부, 주기적 태스크 및 마간시간이 할당된 비주기적 태스크를 스케줄링하여 업데이트를 수행하는 제 1 스케줄링부, 업데이트 전 모드와 업데이트 후 모드사이의 오프셋을 계산하는 오프셋 계산부 및 스케줄링된 주기적 태스크 및 상기 업데이트된 태스크를 스케줄링하고, 상기 계산된 오프셋에 기초하여 태스크를 출력하는 제 2 스케줄링부를 포함하여 구성할 수 있다. 따라서, 실시간 응용 프로그램의 오류 수정이나 기능 수정을 위해실시간 시스템을 정지할 필요가 없으므로, 지속적인 실시간 서비스 제공이 가능하여 사용자들의 만족도 향상과 기업의 경제적 손실 방지를 기대할 수 있다.

公开(公告)号：KR100463837B1

公开(公告)日：2004-12-29

申请号：KR1020020078425

申请日：2002-12-10

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  차정은 ,  김철홍 ,  정효택 ,  송문섭 ,  양영종

IPC: G06F9/45

CPC classification number: G06F8/36

Abstract: In an apparatus for extracting enterprise beans from a servlet source code, a servlet source code input device receives the servlet source code and a Java class. A Java class extraction device extracts Java classes from the servlet source code. A Java class clustering device groups the Java classes on the basis of interrelated classes. A session bean extraction device and an entity bean extraction device extract a session bean and an entity bean from the clustered classes, respectively. An enterprise bean relationship information extraction device extracts relationship information between enterprise beans from the extracted session bean and entity bean.

Abstract translation: 在从servlet源代码中提取企业bean的设备中，servlet源代码输入设备接收servlet源代码和Java类。 Java类提取设备从servlet源代码中提取Java类。 Java类聚类设备根据相关类对Java类进行分组。 会话bean提取设备和实体bean提取设备分别从聚簇类中提取会话bean和实体bean。 企业bean关系信息提取装置从提取的会话bean和实体bean中提取企业bean之间的关系信息。

公开(公告)号：KR100428715B1

公开(公告)日：2004-04-27

申请号：KR1020010001943

申请日：2001-01-12

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  양영종 ,  정효택

IPC: G06F9/44

CPC classification number: G06F8/30 ,  G06F8/315

Abstract: An enterprise java bean generating apparatus and method automatically produces codes based on design information to thereby reduce a development time, a development cost and errors capable of occurring in an enterprise java bean based software development process. The apparatus includes a class diagram receiving unit for receiving the class diagram, an inheritance relationship removing unit for eliminating an inheritance relationship existing among classes in the class diagram, an enterprise java bean extracting unit for extracting the enterprise java beans based on the class diagram whose inheritance relationship was eliminated, an enterprise java bean interface generating unit for producing a remote interface and a home interface of the enterprise java bean, an enterprise java bean grouping unit for grouping enterprise java beans related to each other, and a facade pattern applying unit for applying a facade pattern to the grouped enterprise java beans to thereby unify external interfaces.

Abstract translation: 企业java bean生成设备和方法基于设计信息自动生成代码，从而减少在基于企业javabean的软件开发过程中能够发生的开发时间，开发成本和错误。 该装置包括用于接收类图的类图接收单元，用于消除类图中的类之间存在的继承关系的继承关系移除单元，用于基于类图提取企业java bean的企业java bean提取单元， 继承关系被消除，用于产生企业java bean的远程接口和home接口的企业java bean接口产生单元，用于分组彼此相关的企业java bean的企业java bean分组单元，以及用于产生相互关联的企业java bean的门面模式应用单元 将正面模式应用于分组的企业Java Bean，从而统一外部接口。

公开(公告)号：KR1020000007234A

公开(公告)日：2000-02-07

申请号：KR1019980026445

申请日：1998-07-01

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  이상덕 ,  김채규

IPC: G06F15/16

Abstract: PURPOSE: The matrixes that presents numerically the strength of combinations between classes are utilized in the application implementation stage and fails to present a systematic methods in relation to the application developing methodology. Also the matrixes are not quantitative and not suitable for the practical work of application development. CONSTITUTION: The device includes a Relationship Value Applier(201) which applies values between classes by the standard of relationship between classes within the class diagram when a class diagram is inputted from outside, Message Measurer(202) which produces the number of the kinds of messages by measuring the kinds of messages transmitted between classes within the class diagram. Class Dependency Measurer(203) which measures interdependency with the direction by values produced by the Relationship Value Applier(201) and Message Measurer(202), and Combination Measurer(204) which measures the degree of combination between classes by using values produced by Class Dependency Measurer(203) and outputs the result. This device presents the strength of combinations between classes by numbers before the program code is generated in developing object-oriented applications. Consequentially, the system works more effectively by collecting objects with stronger combinations in the same module.

Abstract translation: 目的：在应用程序实现阶段，利用数学上呈现组合的强度的矩阵，并没有提出与应用程序开发方法有关的系统方法。 矩阵也不是定量的，不适用于应用开发的实际工作。 构成：所述装置包括关系值应用器（201），当从外部输入类图时，通过类图中的类之间的关系标准在类之间应用值，消息量测器（202）产生数量种类 消息通过测量在类图中的类之间传输的消息的种类。 类别依赖性测量器（203），其测量与关系值应用器（201）和消息测量器（202）产生的值与方向的相互依赖性，以及组合测量器（204），其通过使用由Class产生的值来测量类之间的组合程度 依赖度量器（203）并输出结果。 该设备在开发面向对象的应用程序中生成程序代码之前，以数字形式显示了类之间的组合的强度。 因此，通过在同一个模块中收集更强的组合对象，系统更有效地工作。

公开(公告)号：KR1020120047352A

公开(公告)日：2012-05-14

申请号：KR1020100108247

申请日：2010-11-02

Applicant: 한국전자통신연구원

Inventor： 김동관 ,  양영종 ,  전인걸 ,  김원태 ,  박승민

IPC: G06F9/44

Abstract: PURPOSE: A software update method and an apparatus for the same are provided to supply a real-time service without stopping a real-time system for modification of a real-time application program. CONSTITUTION: A first scheduling unit(103) schedules a periodic task and an aperiodic task. The first scheduling unit performs update. An offset calculating unit(104) calculates time interval for mode change. A second scheduling unit(105) schedules the periodic task and the scheduled periodic task. The second scheduling unit outputs a task based on the calculated offset.

Abstract translation: 目的：提供一种软件更新方法及其装置，以提供实时服务，而不用停止实时系统来修改实时应用程序。 构成：第一调度单元（103）安排周期性任务和非周期性任务。 第一个调度单元执行更新。 偏移计算单元（104）计算模式改变的时间间隔。 第二调度单元（105）调度周期性任务和调度的周期性任务。 第二调度单元基于计算的偏移输出任务。

公开(公告)号：KR100326925B1

公开(公告)日：2002-06-27

申请号：KR1019980044346

申请日：1998-10-22

Applicant: 한국전자통신연구원

Inventor： 김흥남 ,  김동관 ,  양영종 ,  정효택

IPC: G06F9/44

Abstract: 본 발명은 객체지향 프레임워크의 브라우징 장치 및 그 방법에 관한 것으로서, 외부와의 인터페이스를 담당하는 그래픽 사용자 인터페이스 관리기(100), 외부로부터의 사용자가 원하는 입력자료를 처리하는 프레임워크 처리기(200), 처리된 외부 입력에 따라 프레임워크 정보저장소에 저장된 자료들을 접근하는 정보저장소 관리기(300)를 구성하여 상기 그래픽 사용자 인터페이스 관리기의 입력부를 통해 외부로부터 입력된 자료들을 분석하여 프레임워크 처리기를 구성하는 프레임워크 구성단위처리부(210), 프레임워크 내부정보 처리부(220), 프레임워크 소스코드 처리부(230) 중 적절한 모듈을 선택하여 상기 입력을 처리하고 각 처리부에서 처리된 결과를 정보저장소에 입력하며, 정보저장소 관리기는 프레임워크 정보저장소를 접근하여 저장된 자료를 추 출하며 그래픽 사용자 인터페이스 관리기(100)의 출력부를 통해 출력함으로써, 객체지향 프레임워크 구조를 쉽게 이해할 수 있도록 지원하고, 프레임워크 재사용을 통한 어플리케이션 개발을 용이하게 할 수 있으며 소프트웨어 개발 시 개발 기간의 단축과 비용을 절감할 수 있고, 소프트웨어 유지 보수 측면에서도 효율성을 높일 수 있는 효과를 가진다.