Abstract:
PURPOSE: A routing table data structure using a class division method, and a method and an apparatus for retrieving a routing path by a routing table are provided to retrieve the routing table composed using the class division method. CONSTITUTION: Prefix information composed as a network address and valid length information is received(710). A class is judged using a length of the prefix to be retrieved(720). It moves to an index tree corresponding to the class, and the first data corresponding to the class are received(730). The received first data are compared with one byte of the network address(740). It moves to a next node, a key value is read, and a next one byte value of the prefix is read(750). It is judged whether a data node of a last destination address path is searched(760). If the data node of the last destination address path is searched, an external interface path value corresponding to the data node is outputted(770).
Abstract:
PURPOSE: A method for exchanging and managing routing information between nodes in a heterogeneous network interworking communication system is provided to construct a routing table through routing information obtained from an operation and management system when a node is initialized, to modify routing tables through a periodic exchange of routing information between nodes, and to revise routing tables using modified routing table information provided from operation and management systems. CONSTITUTION: When a node is initialized, the node registers its own node information, such as a system identifier, a network identifier, provider service identifiers, etc(401). The node configures the maximum available channel value for each provider service identifier and sets the number of allocated channels to '0'(402). If the configuration of node information is completed, the node transmits a routing table request message to an operation and management system, receives a routing table response message, and constructs a routing table(403,404). The node creates a routing information transmitting process in order to broadcast a routing information response message periodically(405). Then the node creates a routing information receiving process to receive and process a routing information response message received from each node(406). Finally the node creates a routing table management process that receives a message from the operation and management system and manages the routing table(407).
Abstract:
PURPOSE: A high speed router system capable of transmitting an IPC(Inter Processor Communication) message having high reliability is provided to enable 2 RP(Router Processor) boards mounted in a high speed router to simultaneously receive messages from many LP(Linecard Processor) boards, thereby minimizing duplexing data to be transmitted between the RP boards. CONSTITUTION: An IPC board(13) switches a transmission path of an IPC message. A switch board(14) switches a transmission path of user data. Active RP boards(10,11) manage a high speed router system, process a routing protocol, and receive or transmit the IPC message through the IPC board(13). Many LP boards(12a-12d) decide whether data received from an external network is the IPC message or the user data, transmit the received data to the IPC board(13) or the switch board(14) according to a decided result, and receive the IPC message transmitted from the active RP boards(10,11) through the IPC board(13).
Abstract:
PURPOSE: A skew alignment operator between multi-lanes for the division transmission of mass data, a method therefor, and a recording medium are provided to perform a parallel process with respect to giga-level multi-lanes and transmit high-speed data with 10 giga bits or more by compensating skew phenomenon between the lanes, which is generated when transmitting the mass data to the multi-lanes. CONSTITUTION: An alignment character column comparing unit(502) detects alignment characters according to lanes from character columns received from multi-lanes, and outputs a control signal for indicating the receiving state of each lane-classified alignment character. A count control unit(509) receives the control signal of the alignment character column comparing unit(502), and outputs a control signal for informing whether an alignment character column is detected to all lanes within a skew alignment setup interval after initially detecting an alignment character from one lane. A parallel storing device control unit(503) receives the control signals of the alignment character column comparing unit(502) and the count control unit(509) and a receiving data column, and successively stores data received after the alignment character column of each lane in a setup storing device. If the storage of one interval data is completed, the parallel storing device control unit(503) successively outputs the stored data, and aligns the data column. An output selection function unit(514) outputs a receiving data column before the skew alignment of the parallel storing device is completed, and outputs the aligned data column outputted from the parallel storing device control unit(503) after the skew alignment is completed.
Abstract:
PURPOSE: A packet processor for supporting a CSIX-L1(Common Switch Interface Specification-Level 1) interface is provided so as to be compatible with a switch fabric manufactured by another company. CONSTITUTION: A packet processor(300k) for supporting a CSIX-L1 interface consists of a lookup processor(320) and a traffic processor(340). The lookup processor(320) executes the discrimination of packet types and the determination of output ports. The traffic processor(340) provides an interface function to a switch fabric, and manages traffics according to output switch ports. The lookup processor(320) is divided into a data transmitting part and a data receiving part according to packet transfer directions. The data receiving part, composed of an input interface circuit(321), an input parsing part(322), an input search part(323), and an input modification part(324), receives packets from a physical link interface part and transfers them to the traffic processor(340). The data transmitting part, composed of an output interface part, transfers packets, transferred from the traffic processor(340), to the physical link interface part.
Abstract:
PURPOSE: A routing table data structure using a class division method, and a method and an apparatus for retrieving a routing path by a routing table are provided to retrieve the routing table composed using the class division method. CONSTITUTION: Prefix information composed as a network address and valid length information is received(710). A class is judged using a length of the prefix to be retrieved(720). It moves to an index tree corresponding to the class, and the first data corresponding to the class are received(730). The received first data are compared with one byte of the network address(740). It moves to a next node, a key value is read, and a next one byte value of the prefix is read(750). It is judged whether a data node of a last destination address path is searched(760). If the data node of the last destination address path is searched, an external interface path value corresponding to the data node is outputted(770).
Abstract:
PURPOSE: A packet connection method between input/output ports for a high speed router is provided to attain a high yield by using a packet priority order determination method for allocating ports according to every packets and prevent the poverty without regard to the data type. CONSTITUTION: A plurality of input/output function modules(110,130) performs a connection function of a physical layer and MAC(Media Access Control) layer and also performs a lookup and buffering function for a packet forwarding. A switch function module(120) has a plurality of packet connection priority order determination function module(121), performs a relay function for determining the switching order when a connection to an output port is set and manages the switch fabric connection for switching by a distribution control as each input port. According to the packet connection priority determination method, one packet among the waited packets is selected and connected to the output port.
Abstract:
PURPOSE: A network processor having a packet generator for a test inside and a method for testing a packet path using the same are provided to generates packets for the test without complex software procedures, thereby executing a path test of the packets with only network processor. CONSTITUTION: A packet generator(407) for a test is installed between a physical layer interface(401) and a switch interface(403). The packet generator(407) generates packets for a test under the control of a processor for packet processing, so that the packet generator(407) provides the generated packets to the physical layer interface(401) or the switch interface(403). The packet generator(407) generates the packets appropriate for the physical layer interface(401) or the switch interface(403) according to kinds of interfaces to be tested. The processor for packet processing sets up a loop back path for executing the test.
Abstract:
PURPOSE: A device and a method for managing a lookup table of the LPM(Longest Prefix Matching)-based CAM(Content Addressable Memory) and a recording medium thereof are provided to arrange an upper pointer and a lower pointer in each band of data with the same prefix in length, and to prepare a space for added data by moving data only indicated by the upper pointers and the lower pointers of each data band, so as to simply update the lookup table. CONSTITUTION: A CAM(460) includes a lookup table that stores data with a long prefix starting from the lowest address. A pointer storage(420) includes a lower pointer and an upper pointer. The lower pointer memorizes an address of data with the lowest address among data with the same prefix length every band of the data with the same prefix length, and the upper pointer memorizes the next address of an address of data with the highest address among data with the same prefix length. And a table manager(440) records data of an address memorized by the lower pointer of each band in the address memorized by the upper pointer of the each band, for the each band of data with a prefix shorter than the prefix of added data if the data are added in the lookup table of the CAM(460). Also, the table manager(440) records the added data in the address memorized by the upper pointer for the band of data with the same prefix length as the added data, and updates the addresses memorized by the upper pointer and the lower pointer included in the pointer storage(420).
Abstract:
PURPOSE: A multi-layer IP(Internet Protocol) for an end-to-end service of a private Internet and a method for transmitting and receiving a multi-layer IP packet are provided to use 1 A class, 16 B classes, and 255 C classes in all access networks and offer a directional Internet service. CONSTITUTION: A public Internet address of a source is stored in a source address field. A public Internet address of a designation is stored in a destination address field. An option field has an option class field for storing data which indicate that information about a private Internet address is stored, an option length field for storing data about a length of Internet address information, a source sub-address field for storing private Internet address information of the source, and a destination sub-address field for storing private Internet address information of the destination.