Adapting device for resilient packet ring and method

Abstract

This invention discloses an adapting device used for adapting elastic sectionalizing ring to Ethernet, which comprises a chain circuit layer interface unit; an Ethernet MAC process unit; a reformat unit to convert the Ethernet message from Ether MAC process unit into RPR message; an upper-lower ring dispatching unit to lower-ring dispatch the RPR message converted by reformat unit; a table looking-up unit to Actualize inquiry of aim node address when adapting upper ring, and according the address inquire ID and existence time; a RPR message detaching unit to convert the RPR message into Ethernet message when adapting lower ring; a multiple-queue memory interface to memorize Ethernet message waiting to adapt and RPR message.

Description

technical field [0001] The invention relates to network data transmission and data technology, in particular to a device and method for realizing the adaptation between elastic packet ring and Ethernet. Background technique [0002] In order to transmit large-capacity data services in the MAN, a new MAC layer (MediaAccess Control, Ethernet data link layer sublayer) protocol: RPR (Resilient Packet Ring, Resilient Packet Ring) has emerged. RPR has a variety of technical advantages and can fully meet the requirements of the next-generation metropolitan area network. [0003] The early RPR technology has multiple technical standards, and each standard is incompatible with each other. In November 2000, the IETF established the IEEE 802.17 working group and began to formulate a unified RPR standard. After launching several standard drafts, the standard gradually matured, and the official standard of IEEE 802.17 is about to be released now. [0004] see figure 1 , RPR adopts a b...

AI Technical Summary

Problems solved by technology

[0009] First of all, this technology is limited by the NP microcode space. If NP is used to complete the adaptation of RPR frames to Ethernet, it needs to consume a certain amount of microcode space, and the microcode space of NP is already a bottleneck for protocol processing above the L3 layer. , NP cannot perform more processing on L3 and above-layer services; the packet format conversion process needs to add or strip headers, and also consumes more microcode space of NP;

[0010] Secondly, for the RPR layer, the ring selection mechanism of the RPR message requires additional CAM resources. For example, 256 nodes need 256*48 (48BIT refers to the MAC address) CAM, and RAM is also required to realize the storage of the ring ID. The NP solution needs to use RAM and CAM resources to complete the adaptation, which is also limited by the CAM / RAM resources inside the NP, and these resources also affect the actual processing capabilities of the NP and the resource requirements of other business layers;

[0011] Secondly, in this technology, the QOS mapping process from Ethernet to RPR also directly limits the microcode space of protocols above the L3 layer of NP.

[0012] In addition, since the NP cannot realize the service processing from one interface to two interfaces (corresponding to the inner and outer sub-rings of the RPR), it cannot complete the replication function of broadcast and multicast packets.

Therefore, this technology uses NP to implement adaptation, and needs to use an additional switching chip from one interface to two interfaces, or use multiple NPs to complete the processing, making management more complicated and costly

Images