Abstract:
본 발명은 파일 레벨 스트라이핑 장치 및 방법에 관한 것으로, 물리 블록 넘버(physical block number)를 통해 액세스되며, 정보를 실제적으로 저장하는 다수의 디스크; 다수의 디스크를 논리적으로 연결하여 하나의 커다란 논리 볼륨을 형성하고, 논리 볼륨을 관리하기 위해 필요한 정보를 다수의 디스크에 기록하고 관리하는 볼륨 관리자; 볼륨 관리자를 통해 제공되며, 논리 블록 넘버(logical block number)를 통해 액세스되는 논리 볼륨; 볼륨 관리자에 의해 제공되는 논리 볼륨을 하나의 저장 장치로 인식하며, 제공된 논리 볼륨 상에 파일들을 생성하고, 생성된 파일에 대한 입/출력을 수행하기 위해 논리 볼륨에 대하여 논리 블록 넘버를 적용하는 파일 시스템을 구비한다. 또한, 파일 생성 인터페이스에 파일 레벨 스트라이핑 여부를 지시하는 옵션을 추가하고, 아이노드 구조에 해당 파일을 위해 마지막으로 물리 블록 할당이 발생한 디스크 ID를 기록하기 위한 필드를 추가하여 이를 기반으로 볼륨 관리자가 물리 블록 할당을 수행함으로써, 전체 논리 볼륨뿐만 아니라 각 파일을 위한 데이터 블록들이 논리 볼륨에 참여하는 전체 디스크에 균등하게 분배되어 대용량 파일에 대한 입/출력의 성능을 향상시킬 수 있는 효과가 있다.
Abstract:
본 발명은 XML 문서 검색장치 및 그 방법에 관한 것으로서, 외부 장치를 통해 데이터베이스 질의어와 XML 질의어가 밀결합된 형태의 통합질의어를 입력받고, 상기 통합질의어를 분석하여 그 통합질의어에 대한 구문트리를 생성하는 구문분석모듈과, 상기 구문분석모듈에서 생성된 구문트리를 이용하여 통합 XML 질의어에서 요구하는 질의를 수행하는 질의처리모듈과, 상기 질의처리모듈의 질의 수행결과를 데이터베이스 엔진에서 처리하기에 적합한 형태로 변환하는 결과처리모듈과, 상기 구문분석모듈, 질의처리모듈 및 결과처리모듈을 제어하고, 데이터베이스 엔진과의 인터페이스를 수행하는 검색관리모듈을 포함하는 XML 문서 검색장치에 의해 XML 문서를 검색함으로써, 대량의 XML 데이터와 기존 데이터를 용이하게 연동하여 사용할 수 있고, XML 질의 � ��리 성능을 최적화할 수 있다는 특징이 있다.
Abstract:
본 발명은 각 논리적 디스크 장치들에 대한 입출력 처리를 다중 제어기들로 분산시키는 분산 공유 RAID 시스템에서 입출력 병목현상을 해결하고, 다중화를 통한 성능 향상 및 데이터 공유로 인해 발생할 수 있는 데이터 훼손이나 손실 등을 방지할 수 있는 다중 RAID 제어기를 통한 데이터 분산 공유 RAID 제어 시스템에 관한 것이다. 본 발명은 다수의 호스트와 논리적 디스크 장치에 대하여 다중 RAID 제어기를 통해 데이터를 분산 공유하는 RAID 제어 시스템에 있어서, 각 RAID 제어기는 그 공유 논리적 디스크장치의 모든 데이터들을 일정 단위로 나누어 각각 일정 분량의 데이터를 분담하고, 자신의 분담 데이터 영역에 대한 락 서버로서 동작하는 것을 특징으로 한다.
Abstract:
PURPOSE: A wideband I/O(Input/Output) device having a DMA(Direct Memory Access) function and a method thereof are provided to transmit/receive data with other computer through the network by directly accessing a memory of the computer if it is confirmed that the data is prepared in a temporary data storing memory. CONSTITUTION: A data reception processor(410) receives a DMA instruction and the I/O request data from other computer connected through the network. An internal MCU(Micro Controller Unit)(420) fixes a DMA processing order by analyzing the DMA instruction from the data reception processor and newly generates the DMA processing instruction. A DMA main controller(460) stores the input request data to a reception data FIFO(First Input First Output)(470) and outputs an output request data transmission instruction by referring to the DMA processing instruction contents orderly stored in a DMA instruction FIFO(450). A data transmission processor(492) transmits the output request data stored in a transmission data FIFO(490) to other computer through the network if the output request data transmission instruction is inputted from the DMA main controller.
Abstract:
The present invention is a data transfer protocol control system with a host bus interface that includes a transmitting/receiving command DMA, a transmitting data DMA and a receiving data DMA for controlling data transfer protocol with a host bus interface considering characteristic, usage frequency, simultaneous processing functions of the command DMA and the data DMAs. A host interface bus is efficiently used and bus usage ratio is distributed properly to support transfer flow properly and improve the entire system performance. The data transfer protocol control system with a host bus interface includes a transmitting/receiving command DMA for instructing the command DMA request buffer to read and write command message data, a transmitting data DMA for instructing the transmitting data DMA request buffer to read the command message data, a receiving data DMA for instructing the receiving data DMA request buffer to write the command message data and a data transfer protocol control device for putting read information, write information and message data on a host bus, receiving message data and a transfer response signal and delivering the message data through the response buffer of the corresponding DMA.
Abstract:
PURPOSE: A method for operating/managing a distributed/shared RAID(Redundant Array of Inexpensive Disks) system based on a SAN(Storage Area Network) is provided to reduce a load of a web server and a diagnosis/management program, and enhance the security reliability by managing/controlling a system through a server based on the web. CONSTITUTION: If a management object is downloaded by using the node information obtained through a heartbeat message, a current state of the server is judged(S402). If the current state of the server is in-service, a download process is terminated. In case that the download of the management object is needed, the state of the server is changed to going-in-service, and the state of all nodes of the in-service is changed to a standby state by broadcasting a standby message to all servers and controllers(S406). Thus, the discordance during the download of the management object is removed by making the standby nodes process only a restricted message.
Abstract:
PURPOSE: A RAID(Redundant Array of Inexpensive Disks) control system for sharing data through a multi-RAID controller is provided to solve the I/O(Input/Output) bottleneck on the shared RAID system distributing the I/O process for respective logical disks to the multi-controllers, and prevent the data loss or damage caused by the performance enhancement and the data share. CONSTITUTION: The system comprises a structure that the multi-RAID controller(220) shares each logical disk(230). For the interface between the hosts(210) or the logical disks, and the multi-RAID controller, an optical switch(240) of a high speed. To transmit the block data between the multi-RAID controllers fast, an SCI(Scalable Coherence Interface) switch(250) is used. To keep data coherence, each RAID controller manages an access authority for the management data of the RAID control by using a lock server.
Abstract:
PURPOSE: A device and a method for managing an XML(eXtensible Markup Language) result of an XML document retrieval system are provided to shorten a response time when a user accesses the result by offering the partial result of a proper size according to user's needs. CONSTITUTION: A result managing module(11) manages a retrieval result until the connection to a server is cancelled by completing the result transmission needed to a client. A result access module(12) provides the access to a predetermined partial result document from an entire result document set of the server in order to support the partial result tree transmission. A DOM(Document Object Model) interface module(13) transparently displays the partial retrieval result document selected from the result access module to the client in a DOM tree form of the W3C(World Wide Web Consortium), transmits the partial result tree of the partial result document between the server and the client, and reconfigures an original result tree.
Abstract:
PURPOSE: A distributed/shared RAID(Redundant Array of Inexpensive Disks) system having a function for recovering an error of a multi-RAID controller and an error recovery method thereof are provided to make the RAID controller having the mirror data continuously perform the function of the RAID controller occurring the error by mirroring the write cache data to the rear RAID controller. CONSTITUTION: All RAID controllers(101-104) are capable of communicating with each other through the network such as the Ethernet, an optical channel switch, and the SCI(Scalable Coherence Interface). Forming the RAID controllers in a logical circular linked list, each RAID controller prevents the data loss when the error is generated by mirroring the write cache data of the RAID controller of a front side in the list to a mirroring cache of the RAID controller of a rear side. Thus, the operation of the RAID controller is continuously performed by making a role of the RAID controller generating the error succeeded to the RAID controller having the mirroring cache for the RAID controller generating the error.
Abstract:
PURPOSE: A prefix prefetching system and a method for multimedia streaming service on the Internet are provided to continuously offer the streaming service in response to an optional position playback request of a user by dividing an area matched with a size of the original data into sectors and caching the data to a recording area of each sector. CONSTITUTION: A wide area server(300) stores the original multimedia streaming data. A local server(100) decides a ratio of the original data to be stored in a local area, and divides each sector(110) into a recording and a non-recording sector(110a,110b) after dividing the area for storing the original data into the sectors by the ratio. The local server stores the data to each recording sector by extracting the data matched with each recording sector from the original data. When the data is not present in the playback position according to the optional position playback request, the local server searches the recording sector close to a playback position, and transmits the data to the user by receiving the rest part of the original data from the wide area server after transmitting the data of the searched recording sector.