CN1456004A - Method for layer 3 address self learning capable of programming in network exchanger - Google Patents

Abstract

A packet identifier module is configured for determining whether a received data packet originated from a router. If the packet identifier module identifies the received data packet as being from a network node other than the router, the switch module selectively stores the layer 2 address and the associated layer 3 address of the data packet as an associated layer 2-layer 3 address pair in an address table. The selective storing of the associated layer 2-layer 3 address is also referred to as learning the internet protocol (IP)-media access control association (MAC) of the data packet. By selectively learning the IP-MAC associations of selected data packets from the non-router ports, the likelihood of overflowing the address table is reduced. Furthermore, by using the learned IP-MAC associations, the network switch may switch layer 3 data packets bypassing the router and reducing latency.

Description

Programmable third-layer address self-learning method in network switch

field of invention

The invention relates to the learning of network addresses of data packets in a non-blocking network switch configured for exchanging data packets in sub-networks and routers.

Background technique

A LAN uses network cables or other media to link stations on the network. Every LAN architecture uses media access control (MAC) to enable network interface devices to access network media at each network node.

The Ethernet protocol IEEE 802.3 has been developed, specifying the half-duplex media access mechanism and the full-duplex media access mechanism for data packet transmission. Full-duplex media access mechanisms provide bidirectional, point-to-point communication links between two network components, such as network nodes and switching hubs.

Switched LANs are facing increasing demands for higher-speed connections, more resilient switching performance, and the ability to adapt to more complex network architectures. For example, U.S. Patent No. 5,953,335, assigned to the same assignee as the present invention, discloses a network switch configured to switch Layer 2 type Ethernet (IEEE 802.3 ) data packet; the received data packet may contain a VLAN (virtual LAN) tagged frame according to the IEEE802.1q protocol, and the IEEE 802.1q protocol specifies another subnetwork (via a router) or a designated group of stations. Since the switching occurs at the Layer 2 level, routers must usually be used to transport packets between subnets.

Efforts to enhance the switching performance of network switches to include Layer 3 (eg, Internet Protocol) processing typically require CPU-based control of network address tables for Layer 3 address learning. For example, a router may perform a Layer 2-Layer 3 association based on a specified Address Resolution Protocol. Typically, routers are an access bottleneck to the LAN since current layer 2 switching is best configured to operate in a non-blocking state, where packets can be output from the network switch at the same rate at which the packet was received. Therefore, the use of switches with Layer 2-Layer 3 switching capabilities can reduce the load on routers and reduce latency. In addition, because within a network switch, Layer 3 learning can quickly overwrite address tables, traditional learning techniques that learn the Media Access Control ("MAC") address of each received packet in a Layer 2 switch fail at Layer 3 Not practical in exchange.

Summary of the invention

What is needed is a solution that enables network switches to provide automatic learning of network addresses for Layer 2 and Layer 3 switching for 100 megabit (Mbps) and gigabit (gigabit) links without packet blocking.

There is also a need for a scheme that enables a non-blocking network switch to selectively learn layer 2 addresses and associated layer 3 addresses of incoming packets at line rate without overwriting the network switch address table.

These needs and others are met by the present invention in which a network switch for switching data packets includes multiple ports for receiving and sending multiple data packets. Incoming data packets are evaluated by the packet identifier module to determine whether the received data packets were received by a router connected to the network switch. If the received packet is from a network node other than the router, the switching module selectively stores the Layer 2 source address and the associated Layer 3 source address of the received packet as the associated second layer address in the address table. Layer-3 address pair. Therefore, since the address table contains fewer entries, the possibility of an overflow situation in the address table is reduced.

It is an object of the present invention to provide a method for switching data packets at a network switch port. The method includes receiving a data packet by a port of a network switch, and judging whether the port has received a data packet from a router. The method also includes, based on a determination that the port receives a packet from a network node other than the router, selectively using the Layer 2 source address and the associated Layer 3 source address from the packet as the associated Layer-3 address pairs, stored in the address table. Therefore, the address table contains fewer entries, thereby reducing the possibility of overflow in the address table.

Another object of the present invention is to provide a network switch for switching received data packets at a port of the network switch. The network switch includes a plurality of ports for receiving and sending a plurality of data packets, wherein one port of the plurality of ports is connected to the router, the packet identifier module and the switching module. The packet identifier module is configured to determine whether the received data packet is from the router. The switching module is configured to selectively use a Layer 2 source address and an associated Layer 3 source address from a received packet as an associated Layer 2-Layer 3 source address based on the fact that the received packet is from a network node rather than a router. Address pairs, stored in the address table. As a result, the possibility of overflow in the address table is reduced.

Additional advantages and novel features of the invention will be set forth in part in the following description and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The advantages of the invention are achieved in particular by means of the measures and combinations indicated in the appended claims.

Brief description of the drawings

Referring to the drawings, in which components with like reference numbers denote like components throughout the specification, and in which:

FIG. 1 is a block diagram of a packet switching network including a plurality of network switches for switching data packets between corresponding sub-networks according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating the network switch of FIG. 1 according to an embodiment of the present invention.

FIG. 3 is a flowchart illustrating learning in the network switch of FIG. 1, according to an embodiment of the present invention.

FIG. 4 is another flow diagram illustrating learning within the network switch of FIG. 1 in accordance with an embodiment of the present invention.

FIG. 5 is yet another flowchart illustrating learning within the network switch of FIG. 1 in accordance with an embodiment of the present invention.

BEST MODE FOR CARRYING OUT THE INVENTION

FIG. 1 is a block diagram illustrating a packet switching network 10, such as an Ethernet (IEEEE 802.3) network. The packet-switched network includes an integrated (eg, single-chip) multiport switch 12 that enables communication of data packets between network stations 14 . Each network station 14, such as a client workstation, is typically configured according to the IEEE 802.3 protocol for sending and receiving data packets at 10 Mbps or 100 Mbps. Each integrated multiport switch 12 is interconnected via Gigabit Ethernet links 16, enabling data packets to be transmitted between subnetworks (or subnets) 18a, 18b, and 18c. Thus, each subnet includes a switch 12 and an associated group of network stations 14 .

Each switch 12 includes a switch port 20 , also referred to as a packet identifier module, that includes a media access control (MAC) module 22 and a port filter 24 . The MAC module 20 sends and receives data packets to the relevant network station 14 through a 10/100Mbps physical layer (PHY) transceiver (not shown in the figure) according to the IEEE 802.3u protocol. Each switch 12 also includes a switching module 25 configured to make frame forwarding decisions on received data packets. In particular, the switching module 25 is configured for Layer 2 switching decisions based on the source MAC address, destination MAC address and VLAN information within the Ethernet (IEEE 802.3) header; according to the evaluation of the IP data portion within the Ethernet packet, The switching module 25 is also configured for selective Layer 3 switching decisions.

As shown in FIG. 1, each switch 12 has an associated host CPU 26 and buffer memory 28, such as SSRAM. The host CPU 26 controls the overall operation of the corresponding switch 12 , including the programming of the switch module 25 . While the switch module 25 is processing forwarding decisions for received data packets, the corresponding switch 12 uses the buffer memory 28 to store data frames.

Each switch 12 also includes a memory 30 configured for limited storage of Internet Protocol (IP)-Media Access Control (MAC) associations of data packets, which acts as an address table.

As noted above, the switching module 25 is configured to perform layer two switching decisions and optional layer three switching decisions. The availability of Layer 3 switching decisions may be particularly effective if an end station 14 within subnetwork 18a wishes to send an e-mail message to selected network stations within subnetwork 18b, 18c. Since subnetworks 18b, 18c are on different subnetworks, hosts within subnetwork 18a cannot know the Layer 2 addresses of hosts on subnetworks 18b and/or 18c. The switching module 25 of the switch 12a is about to send the email message to the router 19, which causes additional delay. Switching module 25 uses the layer-3 switching decision, just makes switching module 25 be able to how to process packet (including further forward decision), and whether packet should be regarded as in the usage (such as video or audio frequency) that is sensitive to the packet time. Intelligent decisions are made on high priority packets. Because routers are usually the bottleneck for LANs, the switch module 25 can offload routers and improve round-trip latency.

According to the described embodiment, network switch 12 is configured to learn the IP-MAC association of selected packets. The packet identifier module 24 of each network switch is configured to determine whether the received data packet is received by the router 19 . If the packet identifier module 24 confirms that the received data packet is from a network node other than the router 19, the switch module 25 of the network switch 12 selectively stores the second layer source address of the data packet and the related third layer source address as the address Associated Layer 2-Layer 3 address pairs in the table. Selective storage of associated Layer 2-Layer 3 addresses is also known as learning of IP-MAC associations for data packets. Reduces the possibility of address table overflow by selectively learning IP-MAC associations for selected packets from non-router ports. In addition, by using the learned IP-MAC association, the network switch can perform Layer 3 switching, which can bypass routers between connected subnets, thereby reducing latency.

FIG. 2 shows a more detailed block diagram of port filter 24 shown in FIG. 1 . Port filter 24 includes receive first in first out (FIFO) buffer 51, MAC queuing logic 52, memory 53, MAC dequeue (dequeing) logic 54, transmit FIFO 55 and processor interface module 57.

The receive FIFO 51 is a buffer configured to temporarily store an incoming data packet in response to receiving an incoming data packet from the receiving portion of the port 20.

MAC queuing logic 52 provides various functions for port filter 24 . MAC queuing logic 52 provides received data packets to SSRAM 28 for writing from receive FIFO 51 via data bus 59 to external memory interface 26. MAC queuing logic 52 also provides a number of status signals 58 to switch module 25 in response to MAC queuing logic 25 processing received packets. Status signal 58 provides an indication to switch module 25 that the received data packet was transferred to external memory interface 26 without error, or that the transfer of the received data packet was complete. Status signals 58 also include a subnet routing selection signal (RNET_ENABLE) and a learning signal (L3IRC_LEARN).

When the MAC queuing logic 52 asserts the RNETS_ENABLE signal, the switch module 25 is notified that the received packet is part of the intra-subnet traffic between subnets directly connected to the network switch 12a.

When the MAC queuing logic 52 sets L3RC_LEARN, the switching module 25 will learn the IP-MAC address association for the received data packets.

Memory 53 provides MAC queuing logic 52 with register space 53a for parameters to implement learning and subnet routing functions. Register space 53a provides at least SUBNET_ID and SUBNET_MASK registers for the MAC queuing logic. The CPU 26 programs the registers via the processor interface (pi_mod) 57 . The SUBNET_ID register provides for storing the IP address to which a single port belongs. Each port on switch 12a has a SUBNET_ID register. SUBNET_MASK provides the 32-bit IP address mask of a single port for storage. Each port on switch 12a has a SUBNET_MASK mask register.

The MAC dequeue logic 54 is responsive to processing by the switch module 25, providing for retrieval of received packets from the SSRAM 28 and delivery of the packets to the appropriate ports.

Transmit FIFO 55 provides a buffer for packets to be transmitted prior to port 20 transmission.

Incoming data packets are received at port 20 and buffered in receive FIFO 51. The MAC queuing logic 52 transmits the data packet to the external memory interface 56, and stores it in the SSRAM 28 via the data bus 59.

MAC queuing logic 52 searches for layer 3 information within received packets, such as IP packets, by examining the packet's header and frame data. Using the layer 3 information, MAC queuing logic 52 can determine whether the received packet is part of internal subnet traffic by comparing the destination IP address in the IP header with values stored in registers in memory 53 .

In particular, if the received packet is received from a router port, the IP destination address will be masked to the SUBNET_MASK of all other ports. The result of the mask operation is then compared against the SUBNET_ID registers of all remaining ports. If the result of the comparison operation is successful, the MAC queuing logic 52 sets the RNETS_ENABLE signal to the switching module 25 .

The MAC queuing logic 52 can also be configured to determine whether the switch module 25 needs to learn the identity of the received packet for the IP address within the subnet when the subnet is directly connected to the network switch 12a when the packet arrives through a non-router port. Source IP-MAC address association. Host CPU 26 is responsible for programming registers in memory 53 so that switch module 25 knows which port is connected to the router.

The MAC queuing logic 52 may also be configured such that an IP-MAC association is learned when the MAC queuing logic 52 of the port identifier module 24 determines that a received packet is intended for the router and may be part of internal subnet traffic.

Additionally, the MAC queuing logic 52 of the packet identifier module 24 may be configured to learn the IP-MAC association of received packets when they are intended for the router and are part of internal subnet traffic.

In particular, MAC queuing logic 52 compares the MAC destination address of the received packet with the router's MAC address stored in memory 53 . If the result of the comparison operation is successful, the received data packet is intended to be sent to the router. Next, the MAC queuing logic 52 of the port identifier module 24 notifies the switch module 25 to optionally store the Layer 2 source address and the associated Layer 3 source address as an associated Layer 2-Layer 3 address pair in address table 30.

The MAC queuing logic 52 of the packet identifier module 42 may be configured to perform any of these functions, depending on the traffic flow in the network switch 12 or user preferences. The MAC queuing logic 52 can also be configured to implement various other functions that the user deems necessary.

FIG. 3 is a flowchart illustrating the learning performed by the PID module 24 shown in FIG. 2 . In step 310 , a data packet is received at port identifier module 24 from one port 20 .

The port identifier module 24 is configured to determine in step 320 whether the received data packet is received from a non-router port. The host CPU is responsible for which port is programmed to connect to the router.

If in step 320, the received packet is being received by a non-router port, then the MAC queuing logic 52 of the port identifier module 24 asserts the learning signal to notify the switch module 25 that in step 330, the layer 2 address or MAC addresses and associated layer 3 addresses or IP addresses are stored in an address table within memory 30 as associated layer 2-layer 3 addresses.

If at step 320 the received packet is being received by a router port, then the MAC queuing logic 52 of the port identifier module 24 does not assert the learning signal to the switch module 25 at step 340 . The IP-MAC associations of received data packets are thus not learned.

FIG. 4 is another flowchart illustrating the learning performed by the PID module 24 shown in FIG. 2 . In this rule, when a packet is received from a non-router port and has the router's destination MAC address, the IP-MAC association will be learned. In step 410 , a data packet is received at the port identifier module 24 via one of the ports 20 .

The port identifier module 24 is configured to determine at step 410 whether the received data packet is being received by a non-router port and the MAC destination address of the data packet is the router. The host CPU 26 is responsible for programming which port is connected to the router, and the MAC queuing logic 52 compares the destination MAC address of the received packet with the router's MAC address.

At step 420 , if the comparison is successful, the received packet is being received on a non-router port and the destination MAC address is a router at step 430 . In step 430, the MAC queuing logic 52 of the port identifier module 24 notifies the switching module 25 that the Layer 2 source address or MAC source address and the associated Layer 3 source address or IP source address are used as the associated Layer 2- The third layer address is stored in the address table of the memory 30 .

If in step 420 , one or both of the comparisons fail, then in step 440 , the port identifier module 24 does not notify the switching module 25 . The IP-MAC association of received packets will not be learned.

FIG. 5 is another flowchart illustrating the learning performed by the PID module 24 shown in FIG. 4 . In this rule, when a packet is received that is part of internal subnet traffic and the destination MAC address is a router, then the IP-MAC association will be learned. In step 510 , a data packet is received at port identifier module 24 from one of ports 20 .

The port identifier module 24 is configured to determine in step 510 whether the received data packet is being received by a non-router port. The MAC queuing logic 52 masks the destination IP address for all other SUBNET MASKs within the switch module 25. The masked result is then compared to all other SUBNET ID addresses within switch 12a. Furthermore, the MAC queuing logic 52 also compares the destination MAC address of the received data packet with the MAC address of the router.

If all comparisons by step 520 are successful, the MAC queuing logic 52 of the port identifier module 24 promptly notifies the switching module 25: in step 530, the second layer source address or the MAC source address and the relevant third layer source address Or the IP source address is stored in the address table in the memory 30 as the associated layer 2-layer 3 address.

If all comparisons by step 520 fail, in step 540 the MAC queuing logic 52 of the port identifier module 24 will not notify the switch module 25 . IP-MAC associations for received packets will not be learned.

According to the described embodiment, the packet identifier module is configured to determine whether the received data packet originates from a router. If the packet identifier module identifies that the received packet is from a network node rather than a router, the switch module selectively uses the layer 2 address and the associated layer 3 address of the packet as the associated layer 2-3 address Yes, stored in the address table. By storing the IP-MAC association for selected packets, network switches can reduce lookup times in the address table when the switch references the address table during switching. Thus, the PID module enables network switches to provide Layer 3 and Layer 2 switching capabilities of 100Mbps or Gigabit links without packet blocking.

Although the present invention has been described in terms of what is presently considered to be the most practicable preferred embodiment, it should be understood that the present invention is not limited to the described embodiments, but rather, the invention should be considered to cover the spirit of the invention contained in the appended claims. and changes within the scope and equivalent schemes.

Claims (14)

1. A method for a network switch, the method comprising: receiving a data packet by a port of the network switch; determining whether the one port has received a data packet from the router; and Based on a determination that said one port received a packet from a network node other than said router, selectively using the layer 2 address and associated layer 3 address from the packet as an associated layer 2-layer Layer 3 address pairs are stored in the address table. 2. The method of claim 1, wherein: The selectively storing step includes determining that the one port receives a packet from a network node other than the router, and specifying the router based on a determination that the destination layer 2 address within the packet specifies the router. The Layer 2 address and the associated Layer 3 address from the packet are stored as an associated Layer 2-Layer 3 address pair. 3. The method of claim 2, wherein: The layer-2 address and the associated layer-3 address are source addresses. 4. The method of claim 1, wherein: The network switch includes a second port and a third port respectively connected to the first subnet and the second subnet. 5. The method of claim 4, further comprising: judging whether the data packet includes address information specifying the transmission of the data packet between the second port and the third port. 6. The method of claim 5, wherein: The selective storing step includes determining that the one port receives a data packet from a network node other than the router, and assigning the data packet between the second port and the third port based on the data packet includes and store the layer-2 address and the associated layer-3 address as the associated layer-2-layer-3 address pair. 7. The method of claim 2, further comprising: Judging by the port identifier module in the network switch that the data packet contains address information specifying the transmission of the data packet between the second port and the third port, wherein the second port and the third port are respectively connected to the first port The first and second subnetworks. 8. A network switch, comprising: A plurality of ports are used to receive and transmit a plurality of data packets, wherein one of the plurality of ports is connected to a router; A packet identifier module configured to determine whether the received data packet is from a router; and a switching module configured to selectively use the Layer 2 address and the associated Layer 3 address from the received packet as an associated Layer 2-3 Layer address pairs are stored in the address table. 9. The network switch of claim 8, wherein: The packet identifier module is configured to determine whether the received data packet contains address information specifying a destination address of the router; and The switch module is configured to selectively use the layer 2 address and the associated layer 3 address from the received packet as a result of the address information including the destination address of the router based on the received packet being from a network node The associated layer 2-layer 3 address pairs are stored. 10. The network switch of claim 8, wherein: The layer-2 address and the associated layer-3 address are source addresses. 11. The network switch of claim 8, wherein: a second port of the plurality of ports is connected to the first sub-network; a third port of the plurality of ports is connected to a second sub-network; and The packet identifier module is configured to determine whether the received data packet contains address information specifying the transmission of the received data packet between the first subnetwork and the second subnetwork. 12. The network switch of claim 11, wherein: The switching module selectively selects the received data packet based on the fact that the received data packet is from a network node and according to the address information specifying that the received data packet is transmitted between the first subnetwork and the second subnetwork The layer 2 address and the associated layer 3 address of , are stored as the associated layer 2-layer 3 address pair. 13. The network switch of claim 8, further comprising: a second port of the plurality of ports is connected to the first sub-network; a third port of the plurality of ports is connected to a second subnetwork; and The packet identifier module is configured to determine whether the received data packet contains address information specifying the transmission of the received data packet between the first subnetwork and the second subnetwork. 14. The network switch of claim 8, wherein: Each port of the plurality of ports is configured to include the packet identifier module.

Images