CN1177437C - Method of implementing quick retransmission and supporting load sharing - Google Patents
Method of implementing quick retransmission and supporting load sharing Download PDFInfo
- Publication number
- CN1177437C CN1177437C CNB021178747A CN02117874A CN1177437C CN 1177437 C CN1177437 C CN 1177437C CN B021178747 A CNB021178747 A CN B021178747A CN 02117874 A CN02117874 A CN 02117874A CN 1177437 C CN1177437 C CN 1177437C
- Authority
- CN
- China
- Prior art keywords
- output interface
- interface information
- message
- node
- pointer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a method for quick retransmission and load share support. Firstly, a circling chain table is set, a pointer pointing to the next output interface is additionally arranged in the data structure of each buffer information unit of a quick retransmission routing buffer table, buffered output interface information is composed of a plurality of output interface information nodes supporting holding load share, each output interface information node is generated dynamically, and the tail pointers of the output interface information nodes are connected to form a loop so as to form a circle chain table structure; firstly, a circling chain table is set, a pointer pointing to the next output interface is additionally arranged in the data structure of each buffer information unit of a quick retransmission routing buffer table, buffered output interface information is composed of a plurality of output interface information nodes supporting holding load share, each output interface information node is generated dynamically, and the tail pointers of the output interface information nodes are connected to form a loop so as to form a circling chain table structure; afterwards, the output interface information is stored in the circling chain table, and data messages are directly forwarded from corresponding output interfaces according to the output interface information of the circling chain table so as to support the load share.
Description
Technical field
The present invention relates to a kind of method that realizes that forwarding fast and holding load are shared, belong to the IP network communication technical field.
Background technology
Referring to Fig. 1 and Fig. 2, the position of quick forwarding module (Fast Switch) and load balancing module (LoadBalance) as shown in FIG. in the router.The characteristic of these two modules is separate.Introduce in the existing equipment basic handling flow process of forwarding module and load balancing module fast below respectively.
When router did not enable quick forwarding capability, its message that receives will directly be delivered the IP layer through link layer and transmit processing: this message was searched routing table at the IP layer, selected corresponding interface to send according to corresponding routing iinformation.This message is after the IP layer outputs to link layer, and the link-layer frame header with more renewing sends from physical layer then.After router has enabled quick forwarding capability, its message that receives will carry out following processing at link layer: at first search route-caching table (claiming again: transmit fast).If find and the list item of its coupling that this message will be delivered the IP layer and carry out above-mentioned forwarding processing.Different is, finishes when this message outputs to link layer process through the IP layer, when being about to give physical layer, will the first-class information of its link-layer frame being added in the route-caching table entry, and then sends this message from physical layer.Like this, when link layer is received the message (message identical as destination address) that same characteristic features is arranged with this message once more, will when searching the route-caching table, find list item with its coupling.At this moment, the new message IP layer of will not going further is handled, and directly according to the information of preserving in the route-caching table, changes its frame head information, sends from corresponding output interface.
After router has enabled the load balancing function, following variation will take place the routing table of router: the route project that originally arrives a certain address in the routing table has only one; And after having enabled the load balancing function, the route project that arrives a certain address in the routing table will have many (referring to Fig. 2, for ease of describing, to be example so that three route table items to be set, to distinguish called after: share route 1, share route 2, share route 3; The output interface that they are corresponding different respectively).Search route when handling through the IP layer successively when one group of IP message with identical destination address (being called for short an IP traffic), will select for use these three to share route in turn, send in turn, thereby realized sharing the function of flow from three different output interfaces
In Fig. 2, what dotted line was described is when not enabling forwarding fast and load balancing function, the general handling process of IP message; The centre of Fig. 2 and below solid line are described is that the handling process of IP message, characteristics are by simplifying the message handling process, improving the efficient that router E-Packets when enabling quick forwarding capability; What the solid line of Fig. 2 top was described is when enabling the load balancing function, and the handling process of IP message, characteristics are to share route by disposing many, make service traffics by the multilink load balancing.
The present router product of selling on the market comprises the router that the applicant develops and the router of the most companies of industry, and it is transmitted not holding load fast and shares characteristic.For example the quick forwarding of company of industry authority Cisco (cisco) also not holding load share function.Can support high-speed data to transmit though the distinctive CEF of cisco company transmits (CiscoExpress Forwarding) characteristic, can share by holding load again; But it accelerates speed of route lookup by adopting the route hierarchical index, simultaneously, in the router of same platform, the CEF forwarding performance is faster transmitted much lower, that is to say by CEF to come holding load to share the efficient height of not sharing by quick forwarding holding load.
In the prior art, when the user enables quick forwarding module, when having disposed many route load sharings simultaneously, because the data structure of existing quick forwarding cache table is only preserved an output interface information (referring to Fig. 3), so router carries out in the following manner at the message that processing needs to transmit: when link layer receives first IP message of an IP stream, can find it earlier and not preserve quick forwarding cache information, just this message be delivered the IP process and transmit processing.Then, when being checked through article one load balancing route, to transmit according to this routing iinformation, and simultaneously, generate and add the corresponding route-caching project (comprising the frame head information, frame head length information of output interface, the output interface of this IP message correspondence etc.) of transmitting fast.Receive second IP message of this IP stream when link layer after,, can directly find corresponding forwarding route-caching information because they have identical index.So from second start of heading, the frame head information that all messages of this IP stream all will directly be sticked corresponding output interface sends from same output interface.That is to say that because quick forwarding has been arranged, these IP message flows all will be forwarded from identical output interface according to article one load balancing route, cause routing overhead to share inefficacy, data traffic can not require to obtain shunting according to the load balancing that the user is provided with.
Fig. 3 has showed the data structure of existing quick forwarding route-caching table, wherein each group cache item is corresponding with an IP stream, each group cache item is made up of an index entry and corresponding output interface information unit, is characterized in only having preserved statically in this output interface information unit the information of an output interface.
Summary of the invention
The purpose of this invention is to provide a kind of method that realizes that forwarding fast and holding load are shared, it is by revising the data structure of existing quick forwarding route-caching table simply, can realize the function of forwarding fast and load balancing simultaneously, utilize quick forwarding mechanism to improve the treatment effeciency of load balancing.
The object of the present invention is achieved like this: a kind of method that realizes that forwarding fast and holding load are shared is characterized in that:
(1) belt chained list is set: in the data structure of each the cache information unit in the quick forwarding route-caching table of forming by a plurality of buffer list entry, be cached with " output interface information " and " pointing to the pointer of next output interface " is set; The data structure of being somebody's turn to do " output interface information " is made up of several " output interface information nodes " of supporting this IP traffic load balancing, wherein each " output interface information node " all adopts dynamically and generates, and be connected to the ring and the list structure that form to go in ring, corresponding one " the output interface information node " of the node of each chained list with the tail pointer in " output interface information node ";
(2) with the output interface information stores in this belt chained list;
(3) data message is directly transmitted from corresponding output interface according to the output interface information that is stored in this belt chained list.
Each cache item in described a plurality of buffer list entry is corresponding to an IP traffic, and described each cache item comprises an index entry and corresponding cache information unit.
Cache information unit in described each cache item stores: " pointing to the pointer of next output interface " and " the output interface information " be made up of several " output interface information nodes ".
Described each " output interface information node " stores: the frame head information of this IP stream, corresponding output interface and tail pointer.
The handling process that described data message is directly transmitted from corresponding output interface according to the output interface information that is stored in this belt chained list further includes the following step:
(A) when system's preliminary examination, at first first " output interface information " in the route-caching table and " pointing to the pointer of next output interface " thereof equal preliminary examination is turned to " sky ";
(B) when link layer receives first message of certain bar IP stream, check its output interface information earlier, find that it is " sky ", then this message is delivered the IP layer and handled, select this IP stream article one load balancing route to transmit; And before this interface sends message, this output interface information is kept at first node of route-caching table, simultaneously the tail pointer of this node is pointed to self, and " pointing to the pointer of next output interface " still is " sky ";
(C) when link layer receives second of certain bar IP stream and subsequent packet, all be its output interface information of checking earlier, if be not " sky ", then inspection " pointing to the pointer of next output interface " at once finds that it is " sky "; Then this message is delivered the IP layer and handled, select a load balancing route of this IP stream to transmit at every turn according to priority; And before its corresponding interface sends message, relatively whether the interface message of having preserved in this output interface information and the route-caching table is different, if there is not identical interface message, then create a new node in the doubly linked list of output interface information, the pointer end to end by each node connects into ring with each node sequence in this output interface information more at every turn; Wherein " point to the pointer of next output interface " when the different routes of each load balancing all are not cached in this output interface information, be " sky " at every turn;
(D) in the different routes of all load balancings after all the branch flow transmission is crossed once the message of this IP stream, first message of the circulation second time of this IP stream is before the link corresponding layer interface E-Packets, preserved this output interface information before finding, just regeneration new node not; But the value of this " output interface information " tail pointer correspondence is composed to " pointing to the pointer of next output interface ";
(E) when the subsequent packet of this IP stream arrives once more, be checked through " pointer that points to next output interface " and be not " sky ", promptly direct output interface information from its node correspondence pointed, find the output interface of current message, and, directly forward from corresponding output interface according to this output interface change dump frame head; Simultaneously, " pointing to the pointer of next output interface " content modification is the value in this node tail pointer.
Innovation key of the present invention is that the data structure of the cache information unit of preserving in the quick forwarding cache is made amendment, increase by one " pointing to the pointer of next output interface ", the data structure of inciting somebody to action wherein " output interface information " simultaneously is modified as by several " output interface information nodes " of supporting this IP traffic load balancing to be formed, wherein each " output interface information node " all adopts dynamically and generates, and is linked in sequence into ring and forms ring-like list structure with the tail pointer that is somebody's turn to do in " output interface information node "; Static storage different from the past, and only store an output interface information; Be aided with corresponding processing method again, just can realize transmitting fast the characteristic that holding load is shared.
Advantage of the present invention is that the resource that can make full use of the network bandwidth is come the balance service traffics when guaranteeing the high speed forward efficiency.This method is revised very simple to existing quick forwarding process, realize easily; And carry out in interruption owing to transmit fast, so the present invention is also very little to the treatment effeciency influence of existing quick forwarding process, result of theoretic analysis is: than the forwarding treatment effeciency height of CEF characteristic under the holding load sharing condition of cisco.
Description of drawings
Fig. 1 is load balancing and the quick residing position view of characteristic of transmitting in existing route system.
Fig. 2 is the quick forwarding of existing IP message and the handling process schematic diagram of load balancing.
Fig. 3 is the data structure schematic diagram of existing quick forwarding route-caching table.
Fig. 4 is that realization of the present invention is transmitted fast and the flow chart of the method that holding load is shared.
Fig. 5 transmits fast in the route-caching table route-caching item number according to the schematic diagram of structure among the present invention.
Embodiment
Referring to Fig. 4, the present invention is a kind of method that realizes that forwarding fast and holding load are shared, and it comprises the following steps: that (1) is provided with belt chained list;
(2) with the output interface information stores in this belt chained list;
(3) data message is directly transmitted from corresponding output interface according to the output interface information that is stored in this belt chained list.
The present invention is that a kind of chained list that goes in ring that adopts is realized the method that forwarding fast and holding load are shared, its data structure of transmitting the route-caching table fast is made up of a plurality of buffer list entry, wherein each group cache item is corresponding to an IP traffic, and each group cache item comprises an index entry and corresponding cache information unit; Innovative point of the present invention is to transmit in the data structure of each the cache information unit in the route-caching table fast at this, except " the output interface information " of buffer memory, one " pointing to the pointer of next output interface " is set also; And the data structure of being somebody's turn to do " output interface information " in the route-caching table also is revised as by several " output interface information nodes " of supporting this IP traffic load balancing to be formed, wherein each " output interface information node " all adopts dynamically and generates, and be connected to the ring and the list structure that form to go in ring, corresponding one " the output interface information node " of the node of each chained list with the tail pointer in " output interface information node "; Being aided with corresponding message handling process again realizes.
The handling process that data message of the present invention is directly transmitted from corresponding output interface according to the output interface information that is stored in this belt chained list further includes the following step:
(A) when system's preliminary examination, at first first " output interface information " in the route-caching table and " pointing to the pointer of next output interface " thereof equal preliminary examination is turned to " sky ";
(B) when link layer receives first message of certain bar IP stream, check its output interface information earlier, find that it is " sky ", then this message is delivered the IP layer and handled, select this IP stream article one load balancing route to transmit; And before this interface sends message, this output interface information is kept at first node of route-caching table, simultaneously the tail pointer of this node is pointed to self, and " pointing to the pointer of next output interface " still is " sky ";
(C) when link layer receives second of certain bar IP stream and subsequent packet, all be its output interface information of checking earlier, if be not " sky ", then inspection " pointing to the pointer of next output interface " at once finds that it is " sky "; Then this message is delivered the IP layer and handled, select a load balancing route of this IP stream to transmit at every turn according to priority; And before its corresponding interface sends message, relatively whether the interface message of having preserved in this output interface information and the route-caching table is different, if there is not identical interface message, then create a new node in the doubly linked list of output interface information, the pointer end to end by each node connects into ring with each node sequence in this output interface information more at every turn; Wherein " point to the pointer of next output interface " when the different routes of each load balancing all are not cached in this output interface information, be " sky " at every turn;
(D) in the different routes of all load balancings after all the branch flow transmission is crossed once the message of this IP stream, first message of the circulation second time of this IP stream is before the link corresponding layer interface E-Packets, preserved this output interface information before finding, just regeneration new node not; But the value of this " output interface information " tail pointer correspondence is composed to " pointing to the pointer of next output interface ", that is to say, " pointing to the pointer of next output interface " at this moment is " sky " no longer, but has pointed to second " output interface information " of load balancing;
(E) when the subsequent packet of this IP stream arrives once more, be checked through " pointer that points to next output interface " and be not " sky ", promptly direct output interface information from its node correspondence pointed, find the output interface of current message, and, directly forward from corresponding output interface according to this output interface change dump frame head; Simultaneously, " pointing to the pointer of next output interface " content modification is the value in this node tail pointer; Like this, follow-up message just can forward from each load balancing interface high speed that disposes on rotation ground, has realized the function that quick forwarding holding load is shared.
Introduce one embodiment of the present of invention (it has disposed three load balancing routes) below in conjunction with Fig. 5:
At first start router, system's preliminary examinationization, first " output interface information " in the route-caching table and " pointing to the pointer of next output interface " thereof all preliminary examination turn to " sky "; " the output interface information node 2 " of " the output interface information node 1 " of " output interface information " lining and back thereof and " output interface information node 3 " all also do not generate at this moment.
When link layer received first message of certain bar IP stream, being checked through " output interface information " was " sky ", then this message was delivered the IP process and was transmitted, and can find article one load balancing route when carrying out route querying, i.e. " output interface information node 1 "; Before link layer is about to send this message, the information data of its corresponding output interface (the frame head information, output interface and the tail pointer that comprise this IP stream) is kept in the node of " output interface information node 1 " of route-caching table manipulative indexing item, simultaneously the tail pointer of this node is pointed to self; At this moment, it " points to the pointer of next output interface " still for " sky ";
When second message of this IP stream arrives, check that earlier " output interface information " is not " sky ", check at once then " pointing to the pointer of next output interface ", find that it still is " sky "; Then this message is delivered the IP layer and handled, the load balancing software function module that has enabled this moment will forward the output interface of this message from second load balancing route; Before link layer sends this message, during the information of relatively having preserved in this output interface information and the route-caching, find the not preservation of current output interface information, then create a new node " output interface information node 2 ", simultaneously, the tail pointer of revising " output interface information node 1 " points to " output interface information node 2 ", and the tail pointer of " output interface information node 2 " points to " output interface information node 1 ", and two nodes connect into ring; And " pointing to the pointer of next output interface " still is " sky ".
When the 3rd message of this IP stream arrives, handle as second message.Promptly submit the IP layer and handle, find the 3rd load balancing route.Before the respective links layer interface sends message, do not preserve this output interface information before finding, so in route-caching, add new node " output interface information node 3 ".Simultaneously, the tail pointer of revising " output interface information node 2 " points to " output interface information node 3 ", and the tail pointer of " output interface information node 3 " points to " output interface information node 1 ", connects into ring." point to the pointer of next output interface " and still be " sky ".
When the 4th message of this stream arrives, also be to handle equally as second message.Promptly submit the IP layer and handle, the load balancing software function module will be delivered article one with this message and shared routing forwarding this moment.Before the respective links layer interface sends message, had preserved this interface message before finding, so regeneration new node not; But the value of " output interface information node 1 " tail pointer correspondence is composed to " pointing to the pointer of next output interface ", that is to say and no longer for " sky ", pointed to " output interface information node 2 " this moment " pointing to the pointer of next output interface ".
When the subsequent packet of this stream arrives, be checked through " pointer that points to next output interface " and be not " sky ", promptly, change frame head according to the output interface information of its node correspondence pointed, directly forward from corresponding output interface.Simultaneously, " pointing to the pointer of next output interface " content modification is the value in this node tail pointer.For example in the present embodiment, the 5th message just no longer submitted the IP layer and handled, but directly finds output interface 2 from " pointing to the pointer of next output interface ", according to directly transmitting behind " output interface information node 2 " replacing frame head.Revise " pointing to the pointer of next output interface " sensing " output interface information node 3 " simultaneously
Like this, follow-up IP message just can forward from the load balancing interface high speed that disposes in turn, thereby has realized the function that quick forwarding holding load is shared.
Claims (5)
1, a kind of method that realizes that forwarding fast and holding load are shared is characterized in that:
(1) belt chained list is set: in the data structure of each the cache information unit in the quick forwarding route-caching table of forming by a plurality of buffer list entry, be cached with " output interface information " and " pointing to the pointer of next output interface " is set; The data structure of being somebody's turn to do " output interface information " is made up of several " output interface information nodes " of supporting this IP traffic load balancing, wherein each " output interface information node " all adopts dynamically and generates, and be connected to the ring and the list structure that form to go in ring, corresponding one " the output interface information node " of the node of each chained list with the tail pointer in " output interface information node ";
(2) with the output interface information stores in this belt chained list;
(3) data message is directly transmitted from corresponding output interface according to the output interface information that is stored in this belt chained list.
2, a kind of method that realizes that forwarding fast and holding load are shared according to claim 1, it is characterized in that: each cache item in described a plurality of buffer list entries is corresponding to an IP traffic, and described each cache item comprises an index entry and corresponding cache information unit.
3, a kind of method that realizes that forwarding fast and holding load are shared according to claim 2, it is characterized in that: the cache information unit in described each cache item stores: " pointing to the pointer of next output interface " and " the output interface information " be made up of several " output interface information nodes ".
4, a kind of method that realizes that forwarding fast and holding load are shared according to claim 3, it is characterized in that: described each " output interface information node " stores: the frame head information of this IP stream, corresponding output interface and tail pointer.
5, a kind of method that realizes that forwarding fast and holding load are shared according to claim 1, it is characterized in that: the handling process that described data message is directly transmitted from corresponding output interface according to the output interface information that is stored in this belt chained list further includes the following step:
(A) when system's preliminary examination, at first first " output interface information " in the route-caching table and " pointing to the pointer of next output interface " thereof equal preliminary examination is turned to " sky ";
(B) when link layer receives first message of certain bar IP stream, check its output interface information earlier, find that it is " sky ", then this message is delivered the IP layer and handled, select this IP stream article one load balancing route to transmit; And before this interface sends message, this output interface information is kept at first node of route-caching table, simultaneously the tail pointer of this node is pointed to self, and " pointing to the pointer of next output interface " still is " sky ";
(C) when link layer receives second of certain bar IP stream and subsequent packet, all be its output interface information of checking earlier, if be not " sky ", then inspection " pointing to the pointer of next output interface " at once finds that it is " sky "; Then this message is delivered the IP layer and handled, select a load balancing route of this IP stream to transmit at every turn according to priority; And before its corresponding interface sends message, relatively whether the interface message of having preserved in this output interface information and the route-caching table is different, if there is not identical interface message, then create a new node in the doubly linked list of output interface information, the pointer end to end by each node connects into ring with each node sequence in this output interface information more at every turn; Wherein " point to the pointer of next output interface " when the different routes of each load balancing all are not cached in this output interface information, be " sky " at every turn;
(D) in the different routes of all load balancings after all the branch flow transmission is crossed once the message of this IP stream, first message of the circulation second time of this IP stream is before the link corresponding layer interface E-Packets, preserved this output interface information before finding, just regeneration new node not; But the value of this " output interface information " tail pointer correspondence is composed to " pointing to the pointer of next output interface ";
(E) when the subsequent packet of this IP stream arrives once more, be checked through " pointer that points to next output interface " and be not " sky ", promptly direct output interface information from its node correspondence pointed, find the output interface of current message, and, directly forward from corresponding output interface according to this output interface change dump frame head; Simultaneously, " pointing to the pointer of next output interface " content modification is the value in this node tail pointer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021178747A CN1177437C (en) | 2002-05-24 | 2002-05-24 | Method of implementing quick retransmission and supporting load sharing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021178747A CN1177437C (en) | 2002-05-24 | 2002-05-24 | Method of implementing quick retransmission and supporting load sharing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1414736A CN1414736A (en) | 2003-04-30 |
| CN1177437C true CN1177437C (en) | 2004-11-24 |
Family
ID=4744555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB021178747A Expired - Fee Related CN1177437C (en) | 2002-05-24 | 2002-05-24 | Method of implementing quick retransmission and supporting load sharing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1177437C (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7761569B2 (en) * | 2004-01-23 | 2010-07-20 | Tiversa, Inc. | Method for monitoring and providing information over a peer to peer network |
| CN100364279C (en) * | 2005-12-07 | 2008-01-23 | 中国科学院计算技术研究所 | Distributed load equalizing method for structured P2P system |
| CN101273590B (en) * | 2005-12-29 | 2012-01-18 | 中兴通讯股份有限公司 | Packet fast forwarding method and system |
| CN100450038C (en) * | 2006-09-29 | 2009-01-07 | 杭州华三通信技术有限公司 | Link load sharing method and relay device |
| CN102316012B (en) * | 2010-06-30 | 2014-05-14 | 杭州华三通信技术有限公司 | Method for realizing Internet protocol (IP) express forwarding and three-layer forwarding equipment |
| CN106656811A (en) * | 2016-12-06 | 2017-05-10 | 上海斐讯数据通信技术有限公司 | Method and system for processing remote call based on AllJoyn frame |
| CN108924050A (en) * | 2018-06-29 | 2018-11-30 | 优刻得科技股份有限公司 | Data forwarding method and its device, storage medium and network card equipment |
-
2002
- 2002-05-24 CN CNB021178747A patent/CN1177437C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1414736A (en) | 2003-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7180895B2 (en) | System and method for classifying network packets with packet content | |
| US7436830B2 (en) | Method and apparatus for wire-speed application layer classification of upstream and downstream data packets | |
| US7418505B2 (en) | IP address lookup using either a hashing table or multiple hash functions | |
| US6091725A (en) | Method for traffic management, traffic prioritization, access control, and packet forwarding in a datagram computer network | |
| US6819670B1 (en) | Data packet routing for mobile networks | |
| US5095480A (en) | Message routing system for shared communication media networks | |
| CN101631086B (en) | Routing list partitioning and placing method searched by parallel IP route | |
| CN1255975C (en) | Method for selecting route for grouping in route apparatus | |
| CN108848032B (en) | Named object network implementation method supporting multi-interest type processing | |
| SE520752C2 (en) | Process and system for recovery of overhead bandwidth in packet switched networks | |
| CN108710629B (en) | Top-k query method and system based on named data network | |
| CN1801775A (en) | Flow assignment | |
| CN1177437C (en) | Method of implementing quick retransmission and supporting load sharing | |
| CN101035059A (en) | Method for improving the classification searching speed of the three-folded content addressable memory message | |
| US7248586B1 (en) | Packet forwarding throughput with partial packet ordering | |
| EP2747336A1 (en) | Content processing method, device and system | |
| CN1777148A (en) | Routing table next-hop IP address to MAC address analytic method | |
| US20080130671A1 (en) | Packet distributing apparatus and packet distributing method | |
| CN107749826B (en) | Data packet forwarding method and system | |
| CN1885820A (en) | Engine apparatus for route forwarding table address searching | |
| CN110096458B (en) | Named data network content storage pool data retrieval method based on neural network | |
| CN1719805A (en) | Multi-path routing method for message order-preserving in 3D Torus exchange network | |
| CN102594650B (en) | Method for processing multicast message and exchange access apparatus | |
| CN1897562A (en) | Method for storing routing hop and next skip list by routing hop | |
| CN1161931C (en) | Radio Three layer tunnel spare and burden sharing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20041124 Termination date: 20110524 |