-
公开(公告)号：KR100261280B1
公开(公告)日：2000-07-01
申请号：KR1019980003300
申请日：1998-02-05
Applicant: 한국전자통신연구원
IPC: G06F13/00
Abstract: PURPOSE: A system for transmitting a file at high speed and a method for controlling the same are provided to utilize a page area, which stores communication data and page data in a main memory for reading the file, as a shared page, and thereby reduce a data copy process and processes of converting a data area and a kernel area. CONSTITUTION: A process(10) manages pages(11) in a main memory via a page table. A memory(40) stores a data file. A communication unit(30) is provided for data transmission. A shared page(20) sets a file stored in the memory(40), and reads in the stored file by a memory-mapped technique. The shared page(20) sends a read area to the communication unit(30), using a DMA(Direct Memory Access) function, before this shared page(20) reads new data.
Abstract translation: 目的：提供用于高速传输文件的系统及其控制方法，以利用将通信数据和页面数据存储在用于读取文件的主存储器中的页面区域作为共享页面，从而减少 数据复制过程和转换数据区域和内核区域的过程。 构成：过程（10）通过页表管理主存储器中的页面（11）。 存储器（40）存储数据文件。 提供用于数据传输的通信单元（30）。 共享页面（20）设置存储在存储器（40）中的文件，并通过存储器映射技术读入存储的文件。 共享页面（20）在该共享页面（20）读取新数据之前，使用DMA（直接存储器访问）功能向通信单元（30）发送读取区域。
-
2.发明公开중계자를 이용한 멀티미디어 서버의 버퍼 관리시스템 및 관리방법(BUFFER MANAGEMENT METHOD AND SYSTEM OF MULTIMEDIA SERVER WITH COORDINATOR) 无效多媒体服务器中间件缓冲区管理系统及管理方法（缓冲区管理方法及多媒体服务器与协调器系统）
公开(公告)号：KR1019980019466A
公开(公告)日：1998-06-05
申请号：KR1019980009515
申请日：1998-03-19
Applicant: 한국전자통신연구원
IPC: H04L12/407
Abstract: 본 발명은 멀티미디어 데이터를 상영하는 노드인 클라이언트(10)와 멀티미디어 데이터가 분산 저장된 디스크를 관리하는 파일 서버(20) 사이에 클라이언트(10)의 요구를 받아서 파일 서버(20)에 명령하고, 파일 서버(20)로부터 전송되는 데이터를 받아서 클라이언트(10)에게 넘겨주는 기능을 수행하는 중계자(30)를 설치하여, 파일 서버(20)를 구성하는 각각의 컴퓨터 파일 서버의 부하를 균일하게 유지시키고, 파일 서버(20)를 구성하는 컴퓨터들이 낮은 성능을 갖는 시스템으로도 구성이 가능하게 되어 경제적인 손실을 줄일 수 있는 중계자를 이용한 멀티미디어 서버의 버퍼 관리시스템 및 관리방법에 관한 것이다.
-
-
Patent Agency Ranking
公开(公告)号：KR100289051B1
公开(公告)日：2001-05-02
申请号：KR1019970065696
申请日：1997-12-03
Applicant: 한국전자통신연구원
IPC: G06F5/06
Abstract: PURPOSE: A device and method for managing a buffer of serial medium data storage is provided to supply a phase difference buffer changing method for decreasing the number of accesses to the lower disk by managing an access phase of a user who is using serial medium data and enhancing a reusing rate of a buffer block using the phase. CONSTITUTION: A data block is stored actually at a buffer area(201) in a main memory(101). A PCB(Phase Control Block)(202) stores phase managing information for managing a phase of a user. A PDB(Phase Difference Block)(203) stores information of a phase difference between adjacent users for changing a buffer. An SRB(Select Replacing Block)(204) stores information which sorts the PDB(203). A buffer manager stores new data by selecting a data block in a conventional main memory buffer area(101) as a sacrifice for appropriating a necessary buffer area(201) in accordance with a user's connection according to the SRB(204). New data refer to the phase of the user, and transmit information to a disk storage control information(205). A disk storage(102) stores the data block in a disk storage buffer area(206). The data block is transmitted from the main memory buffer area(101) to the buffer area(201) for temporarily storing a video data block through the main memory buffer area(101) of a serial medium data server and a video terminal(105) for receiving and outputting video data from the serial medium data server. If a data block is existed in the buffer area, an access to the disk storage(102) is not necessary.
-
公开(公告)号：KR1020010018215A
公开(公告)日：2001-03-05
申请号：KR1019990034078
申请日：1999-08-18
Applicant: 한국전자통신연구원
IPC: G11B20/10
Abstract: PURPOSE: An allocating method of dynamic buffers is provided to increase reusing rate of buffer blocks and to reduce access frequency of a lower disk device by saving/managing approach tendency of a user and deciding the size of buffer blocks therewith. CONSTITUTION: A dynamic buffer allocator registers CRD(current user request distribution information, 203) and BCT(buffer control information,204) if the two are not registered when a user requests a video. Video access data of the user are added to the CRD in real time. The dynamic buffer allocator calculates the current access probability of user with access frequency, saves or corrects the buffer size data of registered video on the BCT. Difference of video data block size influences reusing rate probability of used data blocks. If the video data block is large, a single data block saved in a main memory buffer area(101) is reused. PRD(past user request distribution information,202) is used as initial data when the system is started. The current video server state is applied exactly to buffer management information as the CRD is added continuously. The dynamic buffer allocator always keeps the access data of users and predicts the access tendency of users to decide the buffer block size allocated to each video. A buffer manager responds to data block request of users and reads data of a lower disk device if the request data block is not on a specific buffer area.
Abstract translation: 目的：提供动态缓冲区的分配方法，以增加缓冲区块的重用率，并通过保存/管理用户的访问趋势并确定缓冲区块的大小来降低下磁盘设备的访问频率。 构成：当用户请求视频时，如果两者未被注册，则动态缓冲区分配器注册CRD（当前用户请求分配信息203）和BCT（缓冲器控制信息，204）。 将用户的视频访问数据实时添加到CRD。 动态缓冲区分配器计算具有访问频率的用户的当前访问概率，保存或更正BCT上注册视频的缓冲区大小数据。 视频数据块大小的差异影响使用数据块的重用速率概率。 如果视频数据块大，则存储在主存储器缓冲区（101）中的单个数据块被重新使用。 当系统启动时，PRD（过去的用户请求发布信息，202）被用作初始数据。 当连续添加CRD时，当前的视频服务器状态恰好适用于缓冲管理信息。 动态缓冲区分配器始终保持用户的访问数据，并预测用户决定分配给每个视频的缓冲区块大小的访问趋势。 如果请求数据块不在特定缓冲区上，缓冲管理器响应用户的数据块请求并读取下磁盘设备的数据。
-
公开(公告)号：KR1020000038076A
公开(公告)日：2000-07-05
申请号：KR1019980052938
申请日：1998-12-03
Applicant: 한국전자통신연구원
IPC: G06F12/00
Abstract: PURPOSE: A method for zero-copy message passing is disclosed to eliminate memory duplication overhead caused by message-passing and to minimize the decline in performance by sharing a memory, which is used for passing process, demo and network card message passing, in a network environment. CONSTITUTION: A method for zero-copy message passing is composed of transmission, reception, and direction. When sending message to a remote host, message is copied from processor(201)'s local area to an area where memory(203)'s DMA is possible. A network card(204) transmits data to network, therefore eliminating overhead through zero-copy. A memory area(203) for all processes of message passing and network card receives designated area allocated by boot time. The memory area receives the physical address of start point and size that are returned by operation system and informs the general processor of the information. Each process finds the area where common memory can enter within its imaginary memory, maps the physical memory area and imaginary memory area using a map. Therefore, when writing in its imaginary memory space, it's actually writing in common memory.
Abstract translation: 目的：公开了一种零复制消息传递方法，以消除由消息传递引起的内存复制开销，并通过共享用于通过过程，演示和网卡消息传递的内存来最小化性能下降， 网络环境。 构成：零拷贝消息传递的方法由传输，接收和方向组成。 当向远程主机发送消息时，将消息从处理器（201）的本地区域复制到存储器（203）的DMA可能的区域。 网卡（204）向网络发送数据，从而通过零拷贝消除开销。 用于消息传递和网卡的所有处理的存储区域（203）接收由引导时间分配的指定区域。 存储区域接收操作系统返回的起始点和大小的物理地址，并向通用处理器通知该信息。 每个进程找到公共内存可以在其虚拟内存中进入的区域，使用地图映射物理内存区域和虚拟内存区域。 因此，当写入其虚拟的存储空间时，它实际上是在共同的内存中写入。
-
公开(公告)号：KR100349328B1
公开(公告)日：2002-08-21
申请号：KR1019990034078
申请日：1999-08-18
Applicant: 한국전자통신연구원
IPC: G11B20/10
Abstract: 본발명은비디오서버에서의사용자요구분포정보를이용하여하부디스크디바이스사용횟수를줄이는동적버퍼할당장치및 방법에관한것으로, 비디오데이터를사용하는사용자의접근패턴을기록, 관리하고, 이를이용하여버퍼블록의재사용률을높임으로서하부디스크디바이스접근횟수를줄이기위한수요분포에따른버퍼할당방법을제공하는것을과제로한다. 상기과제를해결하기위한본 발명의구성은비디오데이터를사용하는사용자의접근정보를항상유지하고이 정보를이용하여사용자의접근경향을예측하여각 비디오에할당하는버퍼블록의크기를결정하는것을특징으로한다. 버퍼관리기는사용자의위상에따라버퍼를관리하여사용자의데이터블록요구에응답하고만약예측이틀려요구데이터블록이버퍼영역에없는경우에하부디스크디바이스를접근하여데이터를읽는다.
-
-
-
-
-
-
-
Search Results
8
records
(took 0.016 seconds)
