公开(公告)号：KR1020040046520A

公开(公告)日：2004-06-05

申请号：KR1020020074464

申请日：2002-11-27

Applicant: 한국전자통신연구원

Inventor： 유한양 ,  최완

IPC: G06F17/40

Abstract: PURPOSE: A method for controlling a state and keeping consistency on a dual DBMS(DataBase Management System) is provided to make the state of the DBMS respectively operated in an active DBMS and a standby DBMS at one time point, and keep the consistency of the dual DBMS by improving the state control and an operation process. CONSTITUTION: An internal state of each DBMS is installed(301). At an initial operation, an initial state is fixed by receiving the state of a system including the DBMS, and the internal state of each DBMS and a matched DBMS. For a normal operation after initialization, the internal state of each DBMS is changed based on the state of the system and the state information transmitted/received between the matched DMBSs(307,308,315).

Abstract translation: 目的：提供一种用于控制双DBMS（DataBase管理系统）状态和保持一致性的方法，使DBMS的状态分别在一个时间点在活动DBMS和备用DBMS中运行，并保持一致 双DBMS通过改进状态控制和操作过程。 构成：安装每个DBMS的内部状态（301）。 在初始操作中，通过接收包括DBMS的系统的状态和每个DBMS的内部状态和匹配的DBMS来初始化状态是固定的。 对于初始化后的正常操作，每个DBMS的内部状态根据系统的状态和匹配的DMBS之间发送/接收的状态信息而改变（307,308,315）。

公开(公告)号：KR1020030027473A

公开(公告)日：2003-04-07

申请号：KR1020010060775

申请日：2001-09-28

Applicant: 한국전자통신연구원

Inventor： 박유미 ,  유한양 ,  최완 ,  조완섭

IPC: G06F17/30

Abstract: PURPOSE: A method for visualizing path expression in object-relational DBMS is provided to prepare various and sophisticated queries in a graphical query language. CONSTITUTION: If a client selects a corresponding database(S10) and selects a corresponding class(S20), a GQL(Graphical Query Language) system reads attribute information from a system catalogue of an ORDBMS(Object-Relational DBMS), and writes the read attribute information(S100). The GQL system displays the attribute information in a line of a class box(S30). If the client executes a double click of a query objected attribute(S200), a domain class of the attribute is decided, the GQL system reads the attribute information from an ORDB(Object-Relational DataBase), and creates new class box(S300). If the client clicks a wanted attribute in an attribute list of new class box, the GQL system reads information corresponded to a domain class of the attribute from the ORDB, and creates another class box(S400). If the client checks a path expression displayed as three class boxes, the client selects and inputs a wanted operator and condition value in an operator column and a condition value column out of three class boxes(S500). Thus, a query condition with respect to an attribute is prepared(S40). The GQL system converts a graphic query of a client into a text query, and transmits the query to the ORDBMS(S50).

Abstract translation: 目的：提供一种用于在对象关系DBMS中可视化路径表达式的方法，用于以图形查询语言准备各种和复杂的查询。 规定：如果客户端选择对应的数据库（S10）并选择相应的类（S20），则GQL（图形查询语言）系统从ORDBMS（对象关系DBMS）的系统目录中读取属性信息，并将读取 属性信息（S100）。 GQL系统在类框中显示属性信息（S30）。 如果客户端执行双击查询对象属性（S200），则决定属性的域类，GQL系统从ORDB（对象关系数据库）读取属性信息，并创建新的类框（S300） 。 如果客户端单击新类框的属性列表中的所需属性，则GQL系统从ORDB读取与该属性的域类相对应的信息，并创建另一个类框（S400）。 如果客户端检查显示为三个类框的路径表达式，则客户端在三个类框中的操作员列和条件值列中选择并输入所需操作符和条件值（S500）。 因此，准备关于属性的查询条件（S40）。 GQL系统将客户端的图形查询转换为文本查询，并将查询发送到ORDBMS（S50）。

公开(公告)号：KR1020020059080A

公开(公告)日：2002-07-12

申请号：KR1020000087588

申请日：2000-12-30

Applicant: 한국전자통신연구원

Inventor： 배명남 ,  최완

IPC: G06F17/30

Abstract: PURPOSE: A method for analyzing and transforming a dataset of an overlapped structure on a navigation based data model is provided to effectively control a cursor used for providing the successive approach to the dataset satisfying a condition in a storing system securing the reference integrity of an object identifier while residing in a main memory. CONSTITUTION: A data application program(101) carries out a query to a client terminal and the query is transferred to a data server(103) through the Internet. The data server including a cursor(108) approaches to the dataset embedded in the main memory(104) of a system. Thus, the data server receives a query from the data application program and the query is parsed by a query processor(107). The parsed query is analyzed by the data server and the result is transferred to the data application program by extracting from the main memory. In case that the result includes many objects, the result is divisionally transferred through the cursor. The cursor includes a details list(111), a condition valuation part(112), a catalogue(115), a dataset approach part(110) and a dataset extractor(109).

Abstract translation: 目的：提供一种用于在基于导航的数据模型上分析和转换重叠结构的数据集的方法，以有效地控制用于提供连续方法的游标，以满足在确保对象的参考完整性的存储系统中的条件的数据集 标识符驻留在主存储器中。 构成：数据应用程序（101）向客户终端执行查询，并且通过因特网将查询传送到数据服务器（103）。 包括光标（108）的数据服务器接近嵌入在系统的主存储器（104）中的数据集。 因此，数据服务器从数据应用程序接收查询，并且查询由查询处理器（107）解析。 解析的查询由数据服务器分析，结果通过从主存储器中提取而传送到数据应用程序。 在结果包含许多对象的情况下，结果将通过光标分割。 光标包括细节列表（111），条件估值部分（112），目录（115），数据集合部分（110）和数据集提取器（109）。

公开(公告)号：KR1020020047601A

公开(公告)日：2002-06-22

申请号：KR1020000076091

申请日：2000-12-13

Applicant: 한국전자통신연구원

Inventor： 한미경 ,  최완 ,  김해근 ,  최미선 ,  송은미 ,  정진미 ,  김영국

IPC: G06F12/16

Abstract: PURPOSE: A device and a method for the backup/recovery without processing a log on an MMDB(Main-Memory DataBase system) are provided to improve the performance of the entire system by directly copying a page to be updated to a shadow area of a main memory, editing the updated page and finally carrying out the backup for the updated page from the stable storage to a backup disk. CONSTITUTION: A data processor(21) carries out a transaction by using a shadow paging technique and performs the first backup transferring the pages changed by the present transaction to a backup buffer of the stable storage(24) from the shadow area according to the transaction result. A recovery processor(22) performs the second backup transferring the first backup pages to the backup disk(25) and recovers the pages listed up to a backup list by loading to the MMDB area when a system error is occurred. A main storage(23) stores the information for the MMDB area storing an original data of the database and the shadow area storing the contents changed by the present transaction. The stable storage stores the transaction pages processed by the data processor and the backup data to be transferred to the backup disk.

Abstract translation: 目的：提供用于备份/恢复的设备和方法，而不处理MMDB（主内存数据库系统）上的日志，以通过将要更新的页面直接复制到一个阴影区域来提高整个系统的性能 主存储器，编辑更新的页面，并最终将更新页面的备份从稳定存储器执行到备份磁盘。 构成：数据处理器（21）通过使用影子寻呼技术执行交易，并且根据交易执行第一次备份，以从阴影区域将当前事务改变的页面传送到稳定存储器（24）的备份缓冲器 结果。 恢复处理器（22）执行第二备份，将第一备份页面传送到备份盘（25），并且当发生系统错误时通过加载到MMDB区域来恢复直到备份列表列出的页面。 主存储器（23）存储存储数据库的原始数据的MMDB区域的信息和存储由本次交易改变的内容的阴影区域。 稳定存储将数据处理器处理的事务页面和要传输到备份磁盘的备份数据进行存储。

公开(公告)号：KR1020020043350A

公开(公告)日：2002-06-10

申请号：KR1020000072828

申请日：2000-12-04

Applicant: 한국전자통신연구원

Inventor： 김원영 ,  최완

IPC: H04L12/437

Abstract: PURPOSE: A technique for detecting an asynchronous deadlock in a transaction processing system is provided to detect a deadlock while minimizing the overhead of a general transaction using a synchronous control technique. CONSTITUTION: A deadlock processor obtains a list(WTL) of transactions in a wait state transaction(Tw)(701). Then the deadlock processor obtains a WFG(Wait-For Graph) for each wait-state transaction(Tw) of the list(WTL)(702). The deadlock processor checks whether a specific transaction is in the wait state(702a). In the case that the specific transaction is in the wait state, the deadlock processor takes the semaphore of the lock header of the waitingREQ of the transaction(702b) and checks whether the waitingREQ of the transaction exists in a lock header waitingList(702c). In the case that the waitingREQ of the transaction does not exist in the lock header waitingList, the deadlock processor releases the semaphore(702e) and processes the next transaction. If the waitingREQ of the transaction exists in the lock header waitingList, the deadlock processor analyzes the obtained WFG and detects a cycle(703). In the case that a cycle exists(704), the deadlock processor appoints one of the transactions related to the cycle as a sacrifice transaction and withdraws it(705).

Abstract translation: 目的：提供一种用于检测事务处理系统中的异步死锁的技术，以便在使用同步控制技术最小化一般事务的开销的同时检测死锁。 构成：死锁处理器获得等待状态事务（Tw）（701）中的事务的列表（WTL）。 然后，死锁处理器获得用于列表（WTL）（702）的每个等待状态事务（Tw）的WFG（等待图）。 死锁处理器检查特定事务是否处于等待状态（702a）。 在特定事务处于等待状态的情况下，死锁处理器接收事务的等待REQ的锁定头的信号（702b），并检查事务的等待REQ是否存在于锁头等待列表（702c）中。 在事务的waitingREQ不存在于锁头等待列表中的情况下，死锁处理器释放信号量（702e）并处理下一个事务。 如果事务的等待REQ存在于锁头等待列表中，则死锁处理器分析所获得的WFG并检测周期（703）。 在存在循环（704）的情况下，死锁处理器将与循环相关的一个交易指定为牺牲交易并将其退出（705）。

公开(公告)号：KR100314879B1

公开(公告)日：2002-02-28

申请号：KR1019980017630

申请日：1998-05-15

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 최완 ,  김상욱 ,  김진호 ,  이승선

IPC: G06F17/30

Abstract: PURPOSE: A method for managing a lock for an MDBMS(Main-memory DataBase Management System) is provided to support a concurrency control effectively by setting a unit of locking adapted to an MDBMS and supplying new lock managing method without hashing. CONSTITUTION: A partition which is an object to be locked by a transaction is accessed and a latch is set(S1). A lock header is checked(S2) and if a lock is set in a partition, it is checked whether a lock is obtained in the partition previously by a requested transaction(S3). If a lock is obtained in the partition previously, it is checked whether a strength of the obtained lock mode is more than a mode strength of a requested lock(S4). If a strength of the obtained lock mode is not more than a mode strength of a requested lock, it is checked whether the obtained lock is shared with other transaction(S6). If the obtained lock is shared with other transaction, the lock mode is set as a writing mode(S7). In addition, the latch is released and a waiting state is maintained until a time capable of converting the lock(S8). If the lock is converted(S9), the process is returned(S10).

公开(公告)号：KR100319742B1

公开(公告)日：2002-02-19

申请号：KR1019980052249

申请日：1998-12-01

Applicant: 주식회사 케이티 ,  한국전자통신연구원

Inventor： 한미경 ,  최완

IPC: G06F17/30

Abstract: 1. 청구범위에 기재된 발명이 속한 기술분야
본 발명은 중복 데이터 일치성 제어 방법에 관한 것임.
2. 발명이 해결하려고 하는 기술적 과제
본 발명은 메모리 데이터베이스뿐만 아니라 분산되어 있는 프로세서들의 디스크 데이터베이스간에 실시간성을 지원하면서 중복 데이터에 대한 일치성을 제공할 수 있는 중복 데이터 일치성 제어 방법 및 그 기록매체를 제공하는데 그 목적이 있음.
3. 발명의 해결방법의 요지
본 발명은, 주 메모리 데이터베이스와 주 디스크 데이터베이스의 중복 데이터를 일치시키는 제 1 단계; 변경 정보를 서브시스템의 메모리 데이터베이스와 서브시스템을 위한 디스크 데이터베이스에 대한 일치성 제어를 수행하는 각 프로세스로 전송하는 제 2 단계; 중복 데이터에 대한 상기 서브시스템들의 메모리 데이터베이스의 일치성 제어를 수행하는 제 3 단계; 및 중복 데이터에 대한 상기 서브시스템들의 디스크 데이터베이스의 일치성 제어를 수행하는 제 4 단계를 포함한다.
4. 발명의 중요한 용도
본 발명은 분산 데이터베이스 관리시스템(DBMS) 등에 이용됨.

公开(公告)号：KR100311218B1

公开(公告)日：2001-11-15

申请号：KR1019980049838

申请日：1998-11-19

Applicant: 한국전자통신연구원

Inventor： 한미경 ,  최완

IPC: G06F17/30

Abstract: 1. 청구범위에 기재된 발명이 속한 기술분야
본 발명은 데이터베이스 관리시스템에서 온라인상의 서브시스템 추가/삭제를 지원하기 위한 방법과 그를 실현시키기 위한 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록매체에 관한 것임.
2. 발명이 해결하려고 하는 기술적 과제
본 발명은 교환기시스템 운용 중 서브시스템의 추가/삭제 요구가 발생할 경우 동작중인 교환기시스템에 대해 패키지 교체를 위한 프로세서의 절체과정없이 서비스를 지속적으로 수행하면서 새로 추가/삭제된 서브시스템의 변경된 형상 정보를 반영하여 중복되어있는 데이터가 존재할 경우 데이터변경에 대한 일치성 제어관리를 지원하는 데이터베이스 관리시스템에서 온라인상의 서브시스템 추가/삭제를 지원하기 위한 방법과 그를 실현시키기 위한 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록매체를 제공하는데 그 목적이 있음.
3. 발명의 해결방법의 요지
본 발명은, DBMS가 서브시스템 추가/삭제에대한 형상정보 변경요청을 수신하여 중복 릴레이션 변경금지모드로 하는 단계; DBMS의 형상관련 릴레이션과 시스템 카탈로그내 중복테이블에 서브시스템의 형상정보를 추가/삭제하여 저장수단으로 백업하는 단계; 서브시스템에 대한 로그파일을 생성/삭제한후 중복릴레이션 변경금지를 해제하고, 서브시스템 추가/삭제 요청에대한 처리결과를 전송하는 단계; 및 저장수단 내의 중복릴레이션에 대한 프로세서들의 정보를 포함하는 관련파일을 대체하고, 상기 서브시스템에 대한 데이터베이스를 추가/삭제하는 단계를 포함한다.
4. 발명의 중요한 용도
본 발명은 데이터베이스 관리시스템이 온라인상에서 서브시스템을 추가/삭제가 가능하도록 그 기능을 지원하는데 이용됨.

公开(公告)号：KR1020010064251A

公开(公告)日：2001-07-09

申请号：KR1019990062401

申请日：1999-12-27

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 이병선 ,  최완 ,  전경표

IPC: G06F9/455

Abstract: PURPOSE: A realtime software simulation method using a relative timer is provided to reflect a difference of program performing velocities between a target real time system and a general computer on a test by using a relative timer. CONSTITUTION: If an application program appended kind of a timer being classified in a real time property is inputted(S51), a value of the timer is changed by reflecting a weight value being included in a controller in accordance with the kind of the timer of the program(S52). The application program is performed on an executing time system of a general computer based on the changed timer value(S53). If the timer is performed while the program is performed, a changed timer value is applied by operating a relative timer(S54). The performing result of the application program is analyzed and certified by a developer of the program(S55, S56). It is detected whether the real time property is achieved(S57). If the real time property objected in the target system is achieved, the used weight value is fixed as a weight value of the relative timer controller(S58). If the real time property objected in the target system is not achieved, it is detected whether an adding of new time kind or a re-classification are necessary(S59). If an adding of new time kind or a re-classification are necessary, the corresponding weight value is embodied(S61). A timer kind parameter of the corresponding program is changed in accordance with a timer kind newly appended or adjusted(S62). If the re-classification of the timer kind is not necessary, a weight value of the relative timer controller is modified without modifying the application(S60).

Abstract translation: 目的：提供使用相对定时器的实时软件仿真方法，以通过使用相对定时器来反映目标实时系统和通用计算机之间执行速度差异的测试。 构成：如果输入应用程序附加类别的定时器被分类为实时属性（S51），则通过根据定时器的种类反映控制器中包括的权重值来改变定时器的值 程序（S52）。 基于改变的定时器值在通用计算机的执行时间系统上执行应用程序（S53）。 如果在执行程序时执行定时器，则通过操作相对定时器来应用改变的定时器值（S54）。 应用程序的执行结果由程序开发人员分析和认证（S55，S56）。 检测是否实现实时性（S57）。 如果达到在目标系统中反对的实时属性，则所使用的重量值被固定为相对定时器控制器的权重值（S58）。 如果没有实现在目标系统中反对的实时属性，则检测是否需要添加新的时间种类或重新分类（S59）。 如果需要添加新的时间种类或重新分类，则体现相应的权重值（S61）。 相应程序的定时器类型参数根据新附加或调整的定时器类型而改变（S62）。 如果不需要定时器种类的重新分类，则在不修改应用的情况下修改相对定时器控制器的权重值（S60）。

公开(公告)号：KR1020010047384A

公开(公告)日：2001-06-15

申请号：KR1019990051585

申请日：1999-11-19

Applicant: 한국전자통신연구원

Inventor： 배명남 ,  한미경 ,  박유미 ,  최완 ,  전경표 ,  김상욱 ,  김윤호

IPC: G06F5/00

Abstract: PURPOSE: A method for controlling an extendable hashing directory dividing time of a hash index structure and a method for inserting, deleting and searching an entry using the method are provided to delay a directory separation by using an overflow node being connected to a leaf node. CONSTITUTION: It is decided whether a directory generated as an entry is inserted into an extendable hash at first is "NULL"(S301). If the entry is not "NULL"(S302), new directory is created and an index of a directory made by a bit as much as the directory size is calculated using a hashing function(S303). A bucket being connected to the calculated directory index becomes a home bucket(S304). It is judged whether an overflow bucket exists in the home bucket(S305). If the entry is "NULL"(S302), the above (S303) stage is repeated. If an overflow bucket exists in the home bucket, it is judged whether the last bucket of the overflow bucket being connected by a chain may insert an entry(S306). If the last bucket of the overflow bucket can not insert an entry, new bucket is allocated and connected to the last bucket of the overflow bucket(S307), and the entry is inserted into the last bucket and the process is complected(S308). If an overflow bucket does not exist in the home bucket, it is judged whether the home bucket being connected to the calculated directory index may insert an entry(S309). If the home bucket can insert an entry, an entry is inserted into the home bucket, and the process is completed(S310). If the home bucket can not insert an entry, it is detected whether a local depth is less than a global depth(S311). If a local depth is less than a global depth, two buckets are allocated and the entry in the home bucket is re-hashed(S312), it is decided whether the connected home entry may insert an entry(S313). If the home entry may insert an entry, an entry is inserted into the home bucket(S314), and the process is completed. If the home entry can not insert an entry, new bucket is allocated and an entry is inserted and connected to the home bucket(S315).

Abstract translation: 目的：提供一种用于控制散列索引结构的可扩展散列目录分割时间的方法以及使用该方法插入，删除和搜索条目的方法，以通过使用连接到叶节点的溢出节点来延迟目录分离。 构成：决定是否将作为条目生成的目录首先插入到可扩展散列中是“NULL”（S301）。 如果条目不是“NULL”（S302），则创建新的目录，并且使用散列函数计算由目录大小的位数所产生的目录的索引（S303）。 连接到所计算的目录索引的桶成为家庭桶（S304）。 判断在家用桶中是否存在溢出桶（S305）。 如果条目为“NULL”（S302），则重复上述（S303）阶段。 如果家庭桶中存在溢出桶，则判断通过链接连接的溢出桶的最后一桶是否可以插入条目（S306）。 如果溢出桶的最后一个桶不能插入一个条目，则新的桶被分配并连接到溢出桶的最后一个桶（S307），该条目被插入到最后一个存储桶中，并且处理被复制（S308）。 如果家庭桶中不存在溢出桶，则判断连接到所计算的目录索引的家庭桶是否可以插入条目（S309）。 如果家庭桶可以插入条目，则将入口插入到家庭桶中，并且处理完成（S310）。 如果家庭桶不能插入条目，则检测本地深度是否小于全局深度（S311）。 如果本地深度小于全局深度，则分配两个桶，并重新进行家庭桶中的条目（S312），则确定连接的家庭条目是否可以插入条目（S313）。 如果家庭条目可以插入条目，则将入口插入到家庭桶（S314）中，并且处理完成。 如果家庭条目不能插入条目，则分配新的桶并且插入条目并连接到家庭桶（S315）。