KR100532629B1 - Apparatus for processing packet in RPR network - Google Patents

Abstract

The present invention relates to a packet processing apparatus in an RPR network for processing a packet of each node in an RPR network. The present invention relates to an apparatus for processing a packet in an RPR network including nodes having a dual ring connection function and capable of bidirectional packet transmission. A physical layer connection for physically matching a double ring and a node; An RPR MAC that processes an RPR packet and processes a fairness packet among the input packets and delivers it to the node processor when the packet is an IPS packet, and delivers the packet to the packet inspecting unit when the input packet is a data packet and a topology packet; A node processor for processing and generating an IPS packet delivered from an RPR MAC and a topology packet delivered from a packet inspecting unit; Packets input from the RPR MAC are classified, and topology packets are transmitted to the node processor, data packets are transmitted to the host processor in the form of an Ethernet frame, and an RPR header is added to the data packet input from the host processor. A packet inspecting unit for transmitting to the RPR MAC; And a host processor for processing and generating a data packet transmitted from the packet inspecting unit, the present invention can configure an RPR node and process a packet regardless of whether the host processor supports RPR.

Description

Apparatus for processing packet in RPR network

The present invention relates to an apparatus for processing a packet of each node in a resilient packet ring (RPR) network, and more particularly, is input to each node in an RPR network composed of nodes having a dual ring connection function and capable of bidirectional packet transmission. Packet processing device of RPR network that has a packet processing unit inside node when processing packet and output packet so that IPS packet and topology packet can be processed by node processor and data packet can be delivered to host processor for processing It is about.

FIG. 1A shows a configuration of a general RPR network composed of six nodes 101, 102, 103, 104, 105, and 106, and FIG. 1B shows the format of an RPR packet 130.

Referring to the drawings, the nodes 101, 102, 103, 104, 105, 106 are two links 111, 121, 112, 122, 113, 123, and (outer and inner ring), respectively. 114, 124, 115, 125, and 116, 126 are connected to adjacent nodes, each of which transmits a data packet and a control packet in opposite directions, respectively.

For example, if each node delivers a control packet in a clockwise direction through an out ring 111, 112, 113, 114, 115, 116 to an adjacent node, the data packet is an inner ring 121, 122, 123, 124. Through 125, 126 in the counterclockwise direction.

The packet 130 has a structure consisting of an RPR header, a destination address, a source address, a protocol type, data, and an FCS as shown in FIG. 1B, and each node has a destination of the packet 130 input to its node. If the input packet 130 is its own address, the address is checked and the packet is removed from the link, and if it is not its own address, the packet is forwarded to the next node.

Referring to FIG. 1A, when node 1 101 transmits a packet to node 3 103, node 1 101 transmits to node 2 102 over link 111 and node 2 102 transmits the packet. Since the destination is not its own node, it forwards it to node 3 103 via link 112. In this case, a packet is generated from its own node and sent to the link, and a packet sent from a higher node to a next node is defined as a transition.

2 is a block diagram showing a packet processing apparatus for processing packets in a conventional RPR network. Referring to the drawings, the packet processing apparatus is composed of the inner processing unit 201, 202, 203, the outer ring processing unit (204, 205, 206) and the node processor 207, each ring processing unit according to the function of the physical layer The connection unit 201, 204, RPR Medium Access Control (MAC) 202, 205 and the host processor (203, 206).

The RPR MACs 202 and 205 and the host processors 203 and 206 may be implemented as one chip or may be implemented as separate chips. The physical layer connection units 201 and 204 support a physical matching function such as Ethernet or Packet over SONET, and the host processor may be a network processor or a general micro-processor.

Referring to the flow of packets in the conventional node configured as described above, the packets input through the physical layer connection unit 201, 204 are delivered to the RPR MAC (202, 205), respectively, the RPR MAC (202, 205) of the input packet The header is inspected and forwarded to the node processor in case of IPS packet and to the host processor (203, 206) in case of topology packet or data packet, and the fairness packet is transmitted to RPR MAC (202, 205). Is processed.

The host processor 203, 206 of each node generates data packets, IPS packets, and topology packets, which are transmitted to the corresponding ring through the RPR MAC 202, 205 and the physical layer connections 201, 204. .

When a node is configured to process a packet as described above in an RPR network composed of nodes having a dual ring connection and capable of bidirectional packet transmission, the host processor generates an IPS packet and a topology packet, and generates an RPR related protocol. Since the processing of the host processor is complicated, the method of processing must be changed according to the type of the host processor.

In the RPR network, the host processor generates and processes only data packets in an Ethernet frame structure in the host processor of each node, and all protocol functions related to the RPR protocol, including IPS packets and topology packets, are generated by the node processor. It is an object of the present invention to provide a packet processing apparatus in an RPR network that can configure an RPR node and process a packet regardless of whether RPR is supported.

An apparatus for processing a packet in an RPR network of the present invention for achieving the above object is an apparatus for processing a packet in an RPR network composed of nodes having a dual ring connection function and capable of bidirectional packet transmission, wherein the dual ring and a node are physically A physical layer connection for matching; An RPR MAC that processes an RPR packet and processes a fairness packet among the input packets and delivers it to the node processor when the packet is an IPS packet, and delivers the packet to the packet inspecting unit when the input packet is a data packet and a topology packet; A node processor for processing and generating an IPS packet delivered from the RPR MAC and a topology packet delivered from a packet inspecting unit; Packets input from the RPR MAC are classified, and topology packets are transmitted to the node processor, data packets are transmitted to the host processor in the form of an Ethernet frame, and an RPR header is added to the data packet input from the host processor. A packet inspecting unit for transmitting to the RPR MAC; And a host processor for processing and generating the data packet transferred from the packet inspecting unit. Characterized in that it comprises a.

The packet inspecting unit may further include a host processor matching unit for exchanging packets of an Ethernet type with a host processor, a node processor matching unit for exchanging packets with a node processor, a MAC matching unit and headers for exchanging RPR MACs with RPR packets. It consists of a packet classification and processing unit for classifying and classifying packets and generating and removing headers for data packets.

According to this configuration, the present invention examines the header of the packet input from the RPR MAC and passes it to the node processor without changing the packet in the case of the topology packet, and removes the RPR header in the form of the Ethernet frame in the case of the data packet. Pass to host processor. In this case, the node processor can generate and process the IPS packet and the topology packet processed by the host processor in the conventional method, so that the host processor configures the RPR node regardless of whether RPR is supported or not. It can improve the data transmission speed by preventing the bottleneck caused by packet processing in the whole network.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

3 is a block diagram showing a packet processing apparatus of an RPR network according to the present invention. Referring to the drawings, the present invention adds packet inspection units 303 and 307 to existing packet devices. The packet inspecting units 303 and 307 may be implemented as one chip or two chips each divided into an inner packet inspecting unit and an out packet inspecting unit. The present invention includes physical layer connections 301 and 305 for physically matching the double ring and the node; Processes RPR packet and processes fairness packet among input packets and delivers it to node processor 309 in case of IPS packet, and transmits to packet inspection unit when input packet is data packet and topology packet. RPR MAC 302, 306; A node processor (309) for processing and generating an IPS packet delivered from the RPR MAC (302, 305) and a topology packet delivered from a packet inspection unit (303, 307); Packets input from the RPR MAC are classified, and the topology packets are transmitted to the node processor 309, the data packets are transmitted to the host processors 304 and 308 in the form of Ethernet frames, and the data packets input from the host processor. A packet checker (303, 307) for adding an RPR header to the RPR MAC and transmitting the same to the RPR MAC; And a host processor generating and processing the data packet transferred from the packet inspecting unit.

In the present invention configured as described above, the packets input through the inner ring and the outer ring are transferred to the physical layer connection units 301 and 305, respectively, and the packets input to the physical layer connection units 301 and 305 are transmitted to the RPR MACs 302 and 306. Delivered. If the input packet is a fairness packet, the RPR MACs 303 and 306 process it. If the input packet is an IPS packet, the packet is transmitted to the node processor 309.

The topology packet and the data packet are delivered to the packet inspecting units 303 and 307, and the packet inspecting units 303 and 307 classify the transmitted topology packets and data packets to form the topology packet. The data packet to the node processor 309 removes the RPR header and forwards the data packet to the host processors 304 and 308 in an Ethernet frame structure.

Through this process, the host processors 304 and 308 receive data packets irrelevant to the RPR protocol. If there is a packet to be transmitted by the host processor 304 or 308, the host processor 304 or 308 transmits the packet to the packet inspecting unit 303 or 307 in the form of an Ethernet frame. The RPR header is attached to the forwarded packet and transmitted to the RPR MACs 303 and 306.

The node processor 309 receives the IPS packet from the RPR MACs 302 and 306 and receives the topology packet from the packet inspecting units 303 and 307, respectively, and processes the IPS packet and the topology packet. By initiating a system that delivers the packet to the packet inspecting units 303 and 307, the RPR node can be configured regardless of whether the host processor supports the RPR protocol.

4 is a detailed block diagram illustrating a packet inspecting unit of the present invention. The packet inspecting units 303 and 307 are composed of a node processor matching unit 401, a MAC matching unit 402, a packet classification and processing unit 403, and a host processor matching unit 404, respectively.

The node processor matching unit 401, the MAC matching unit 402, and the host processor matching unit 404 generate data and control signals for each interface, and the packet classification and processing unit 403 is configured to generate an input packet. The header is inspected and transmitted to the node processor matching unit 401 in the case of the topology packet, and the RPR header is removed in the case of the data packet and then transferred to the host processor matching unit 404 in the case of the data packet.

The data packet transmitted from the host processor matching unit 404 is added to the MAC matching unit 402 after the RPR header is added by the packet classification and processing unit 403. The IPS packet and the topology packet transmitted from the node processor matching unit 401 are delivered to the MAC matching unit 402 through the packet classification and processing unit 403. In this case, the packet classification and processing unit 403 gives priority to the data packet, the IPS, and the topology packet, and simultaneously inputs the IPS packet and the topology packet to the MAC matching unit 402.

As described above, the present invention processes a RPR packet in the form of an Ethernet protocol by adding a packet inspecting unit to a conventional packet processing apparatus, thereby applying a reliable data transmission function to an RPR network regardless of the structure of the network, the type and function of the host processor. Can be done.

As described above, the present invention provides a packet processing unit inside a node when processing an input packet and an output packet to each node so that a node processor processes an IPS packet and a topology packet, and processes and generates a data packet. By allowing the host processor to process RPR control packets independently of the host processor, the reliability of data transmission can be improved.

What has been described above is only one embodiment for explaining a packet processing apparatus in an RLP network according to the present invention, and the present invention is not limited to the above-described embodiment, but the present invention is claimed in the following claims. Without departing from the gist of the present invention, those skilled in the art to which the present invention pertains to the spirit of the present invention to the extent that various changes can be made.

Figure 1a is a view showing the configuration example of a general RPR network, Figure 1b is a view showing the format of an RPR packet,

2 is a block diagram showing a conventional packet processing apparatus;

3 is a block diagram of a packet processing apparatus according to the present invention;

4 is a block diagram showing the configuration of a packet inspection unit according to the present invention.

Explanation of symbols on main parts of drawing

301, 305; physical layer connection 302, 306; RRP MAC

303, 307; Packet inspection unit 304, 308; Host processor

309; node processor

Claims (4)

An apparatus for processing a packet in an RPR network consisting of nodes having a dual ring connection function and capable of bidirectional packet transmission, A physical layer connection for physically matching the double ring and the node; An RPR MAC that processes an RPR packet and processes a fairness packet among the input packets and delivers it to the node processor when the packet is an IPS packet, and delivers the packet to the packet inspecting unit when the input packet is a data packet and a topology packet; A node processor for generating and processing an IPS packet and a topology packet delivered from the RPR MAC; Packets input from the RPR MAC are classified, and topology packets are transmitted to the node processor, data packets are transmitted to the host processor in the form of an Ethernet frame, and an RPR header is added to the data packet input from the host processor. A packet inspecting unit for transmitting to the RPR MAC; And A host processor for processing and generating a data packet transferred from the packet inspecting unit; Packet processing apparatus in the RPR network, comprising a. The method of claim 1, wherein the packet inspection unit, Packet classification is performed by examining the host processor matching unit for exchanging packets with the host processor and the Ethernet processor, the node processor matching unit for exchanging packets with the node processor, the MAC matching unit and headers for exchanging RPR MACs with RPR packets, and headers. The packet processing apparatus of the RPR network, characterized in that the data packet comprises a packet classification and processing unit for generating and removing the header. The RPR network according to claim 2, wherein the packet classification and processing unit prioritizes an IPS and a topology packet when priority is input between data packets, an IPS and a topology packet, and is simultaneously input. Packet processing device in. The apparatus according to any one of claims 1 to 3, wherein the packet inspecting unit comprises one or two chips.