NETWORK PACKET STORAGE METHOD AND NETWORK PACKET TRANSMITTING APPARATUS USING THE SAME

• US 20070086428A1
• Filed: 01/03/2006
• Published: 04/19/2007
• Est. Priority Date: 10/19/2005
• Status: Active Grant

• Current Assignee: A10 Networks Incorporated
• Original Assignee: Industrial Technology Research Institute
• Inventors: Lai, Yu-Jen, Chiang, Hsin-Chieh
• Granted Patent: US 7,627,672 B2
• Time in Patent Office: Days
• US Class Current: 370/351
• CPC Class Codes: H04L 49/103 using a shared central buffer; using a shared memory
• CPC Class Codes: H04L 49/251 Cut-through or wormhole routing
• CPC Class Codes: H04L 49/3027 Output queuing
• CPC Class Codes: H04L 49/3045 Virtual queuing

Abstract

A network packet storage method and network packet transmitting apparatus using the same is provided. A packet content and a path control information are obtained from a received packet, and a packet storage address of the packet content is determined according to a predetermined storage rule. The packet content is stored in a data block and packet storage addresses and path control information of packets output via the same output port are stored in a control block. The data block is output when its size reaches a predetermined data block size such that a dynamic random access memory receives the data block by using burst mode. Further, the control block is output when its size reaches a predetermined control block size.

Claims

1. A network packet transmitting apparatus, comprising:

a packet receiving module for receiving a packet and obtaining a packet content and a path control information from the packet;

an input packet control module coupled to the packet receiving module, for storing the packet content and determining a packet storage address of the packet content according to a predetermined storage rule, and for outputting a data block accumulated by all the stored packet contents when the size reaches a predetermined data block size, and further for aggregating the packet storage addresses and path control information of the packet contents from the same output port into a control block according to the path control information, and outputting the control block when thes size reaches a predetermined control block size, and providing an output trigger signal;

a control information queue unit for coupling the input packet control module with the packet output module, and storing the output trigger signals and sending them one by one;

a dynamic random access memory coupled to the input packet control module, for storing the data block and the control block; and

a packet output module coupled to the dynamic random access memory and the input packet control module, for obtaining the corresponding data for outputting from the dynamic random access memory according to the output trigger signal.

2. The network packet transmitting apparatus as claimed in claim 1, wherein the input packet control module comprises:

a buffer area management unit, for initially obtaining a control area storage state which indicates the use state of the storage area of the dynamic random access memory for storing the control block, then providing an available row address according to the control area storage state when the packet control module needs an available row, and adjusting the control area storage state dynamically according to the available row address and the data output state of the packet output module; and

a control unit coupled to the buffer area management unit and the control information queue unit, for determining an address of the data block according to a data area storage state indicating the use state of the storage area of the dynamic random access memory for storing the data block, determining an address of the control block according to the available row address, aggregating and outputting the data block and the control block, and outputting the output trigger signal when outputting the control block.

3. The network packet transmitting apparatus as claimed in claim 2, wherein the control unit comprises:

a transmitting interface coupled to the dynamic random access memory, for transmitting data;

a data block agent coupled to the transmitting interface, for providing the data block to the transmitting interface;

a control block agent coupled to the transmitting interface, for providing the control block to the transmitting interface;

a link manager coupled to the data block agent, for obtaining the data area storage state through the transmitting interface;

a packet manager coupled to the link manager and the packet receiving module, for receiving the packet content and the path control information and determining the packet storage address according to the data area storage state;

a row allocator coupled to the buffer area management unit and the packet manager, the packet manager having a plurality of logic rows for storing the packet storage address and the path control information, and determining the corresponding addresses of the logic rows in the dynamic random access memory according to the available row address; and

a built-in memory coupled to the packet manager and the data block agent, for storing the packet content to generate the data block, and providing the data block to the data block agent.

4. A network packet storage method, comprising:

obtaining a packet content and a path control information from a received packet;

determining a packet storage address of the packet content;

storing the packet content in a data block;

storing packet storage addresses and path control information of packets output via the same output port in a control block;

outputting the data block when the size reaches a predetermined data block size, such that a dynamic random access memory receives the data block in burst mode; and

outputting the control block when the size reaches a predetermined control block size.

5. The network packet storage method as claimed in claim 4, wherein the step of storing packet storage addresses and path control information of packets output via the same output port in a control block comprises:

obtaining a control area storage state indicating the use state of the storage area of the dynamic random access memory for storing the control block; and

providing a plurality of logic rows for storing the packet storage addresses and the path control information, wherein the corresponding addresses of the logic rows in the dynamic random access memory are determined by the control area storage state.

6. The network packet storage method as claimed in claim 5, wherein the step of providing a plurality of logic rows for storing the packet storage addresses and the path control information comprises:

providing a plurality of private storage spaces and a common storage space to each of the logic rows, wherein the private storage spaces correspond to a plurality of existing output ports one by one, the output ports each having a priority storage sequence for re-arranging a priority of the logic rows arranged according to an original priority initially, and the respective priority storage sequences of the output ports being completely the same initially; and

for each packet, the following steps are performed from the logic row with the highest priority in the priority storage sequence of the output port corresponding to the packet;

a. determining whether there is enough space in the private storage space corresponding to the output port via which the packet is output to store the packet storage address and the path control information, and if so, storing the packet storage address and the path control information in the private storage space;

b. if there is not enough space in the private storage space, determining whether there is enough space in the common storage space to store the packet storage address and the path control information, and if so, storing the packet storage address and the path control information in the common storage space; and

c. if there is not enough space in the common storage space, adjusting the priority of the logic row with the highest original priority in the priority storage sequence to be the lowest, while adjusting the priorities of other logic rows upwards for one level, and repeating the operations of a˜

c.

7. The network packet storage method as claimed in claim 6, wherein when the logic rows with the highest priority in the priority storage sequences of all of the output ports are no longer the logic rows with the highest original priority, the data of the logic row with the highest original priority is transferred to the dynamic random access memory, and the original priority of the logic row is adjusted to be the lowest, while the original priorities of other logic rows are adjusted upwards for one level.

8. The network packet storage method as claimed in claim 6, wherein if the logic row with the highest priority in the priority storage sequence of any of the output ports is the same as the logic row with the lowest priority in the priority storage sequence of another output port, then:

the priority storage sequence corresponding to an output port is adjusted such that the priority of the logic row with the highest priority is adjusted to be the lowest, while the priorities of other logic rows are adjusted upwards for one level, wherein the adjusted output port is the one the logic row with the highest original priority is used as the logic row with the highest priority in the corresponding priority storage sequence;

the data of the logic row with the highest original priority is transferred to the dynamic random access memory, and the original priority of the logic row is adjusted to be the lowest; and

the original priorities of other logic rows are adjusted upwards for one level.