CN101304383B - Method for switching message of switching network and switching system - Google Patents
Method for switching message of switching network and switching system Download PDFInfo
- Publication number
- CN101304383B CN101304383B CN2008101162258A CN200810116225A CN101304383B CN 101304383 B CN101304383 B CN 101304383B CN 2008101162258 A CN2008101162258 A CN 2008101162258A CN 200810116225 A CN200810116225 A CN 200810116225A CN 101304383 B CN101304383 B CN 101304383B
- Authority
- CN
- China
- Prior art keywords
- plane
- message
- route
- ply
- yarn drill
- 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.)
- Active
Links
Images
Abstract
The invention discloses a message exchange method for a switching network, which comprises that according to the type of an accessible line card of an exchange device, each line card is provided with an exchange plane with an independent plane cache in the exchange device; the exchange device receives the message with the type being different from a source line card of the line card and a target line card and then exchanges the message to a plane conversion device; the plane conversion device sends the received message to an exchange device and the exchange plane corresponding to the type of the target line card, and then the message is sent to the corresponding target line card through the exchange plane. The invention discloses an exchange system, a routing line card and an aether line card. The message exchange method of the invention can be used for realizing a plurality of exchange planes in one exchange device.
Description
Technical field
The present invention relates to the switching network technology, be specifically related to a kind of message switching method and a kind of switching system that is applied to switching network.
Background technology
For expansion port density, the switching equipment such as router and switch adopts modular system architecture usually at present.Modular system architecture comprises switching network backboard and a plurality of ply-yarn drills that are plugged on this switching network backboard.A plurality of ply-yarn drills carry out message switching by the switching network that the switching network backboard provides.As seen, switching network is the core of finishing message switching between ply-yarn drill.
The switching network of modular ethernet switches is called the ether exchange plane.The ether exchange plane is to be first exchange plane with exchange efficiency, allows packet loss, thereby avoids congested and the follow-up flow of influence.Fig. 1 is the structural representation of modular ethernet switches.As shown in Figure 1, this Ethernet switch comprises Ethernet packet switching chip 11 and a plurality of ether ply-yarn drill 12.M ether ply-yarn drill has been shown among Fig. 1, and m is the integer greater than 1.The ether ply-yarn drill 12 main hardware that adopt are realized at present, comprise physics (PHY) layer 121 and medium access control (MAC) layer 122.PHY layer 121 directly is connected with Ethernet packet switching chip 11, and MAC122 finishes in the ether ply-yarn drill Data Format Transform of switching network protocol massages in the IP message and switching network, and the message after will changing then mails to Ethernet packet switching chip 11 by PHY layer 121.Ethernet packet switching chip 11 is a switch, and the ether exchange plane is provided, and this Ethernet packet switching chip 11 comprises a plurality of ports that can be used as inbound port and/or outbound port, and each port links to each other with an ether ply-yarn drill 12; Ethernet packet switching chip 11 will be received from the output buffers of the message switching of inbound port to outbound port.When a plurality of ports during simultaneously to a port transmitted traffic, because the buffer memory of packet switching chip is limited, the subsequent packet that can't enter output buffers is dropped.
The switching network of modular router is called the route switching plane.For guarantee service quality (QoS, Quality of Service), requiring in the route switching plane can not packet loss, and packet loss can only occur in outside the switching network.Router can adopt special route switching chip to realize the not message switching of packet loss, but this route switching chip cost is higher.At present, a kind of low-cost Ethernet packet switching chip modular router of core in return that adopts has appearred.The modular router structural representation of Fig. 2 for adopting the Ethernet packet switching chip to realize.As shown in Figure 2, this router comprises Ethernet packet switching chip 21 and a plurality of route line card 22.Wherein, Ethernet packet switching chip 21 is switches of modular router, and its exchange principle is identical with Ethernet packet switching chip 11 among Fig. 1.Route line card 22 is basic identical with the function of ether ply-yarn drill 12, but adopts software to realize, specifically comprises exchange access unit (FAU, FabricAccess Unit) 221 and packet handler (PP, Packet Process) 222.Wherein, because PP222 can not directly link to each other with the port of Ethernet packet switching chip 21, therefore be responsible for the port of Ethernet packet switching chip 21 and the adaptive connection between the route line card 22 by FAU221, finish in the route line card Data Format Transform of switching network protocol massages in the IP message and switching network.In addition, do not allow packet loss in the route switching net, therefore FAU221 also comprises distributed scheduler 223, distributed scheduler 223 exchange schedule informations (schedule information as shown in phantom in Figure 2) on each FAU221, each distributed scheduler 223 control place route line card 22 transmitted traffic to Ethernet packet switching chip 21, thereby realization enters the flow of the same purpose ply-yarn drill of Ethernet packet switching chip 21 is less than or equal to the flow of Ethernet packet switching chip 21 to this purpose ply-yarn drill output, like this, when a plurality of ports during simultaneously to a port transmitted traffic, can not occur congestedly in the Ethernet packet switching chip 21 yet, guarantee that Ethernet packet switching chip 21 exchanges not packet loss.As seen, because the adding of distributed scheduler 223 makes all distributed schedulers 223 constitute the scheduling plane that is superimposed upon on the ether exchange plane, the Ethernet exchange plane behind the plane is dispatched in stack, has formed the route switching plane of no packet loss.
At present, a kind of trend of network equipment evolution is that route switching and ether exchange are integrated, this need realize ether exchange plane and route switching plane simultaneously on same equipment, make switching equipment can provide simultaneously with exchange efficiency and be the exchange of first Ethernet and be first route switching with service quality.
But aforementioned switch and router can only be realized a kind of exchange plane on a switching equipment.For example, the route switching plane is realized on the scheduling plane thereby the router shown in Fig. 2 passes through to superpose on the Ethernet exchange plane.If wish further to realize the ether exchange plane on the route switching plane, a kind of simple implementation is directly to insert the ether ply-yarn drill shown in Fig. 1 on the Ethernet packet switching chip 21 of router.Because each port can be shared output buffers in the existing Ethernet packet switching chip, so the shared switching network output buffers of ether exchange plane and route switching plane.After a large amount of Ethernet message enter switching network, may cause the switching network output buffers to overflow, output buffers overflows can influence route switching.
For example, a plurality of ether ply-yarn drills are simultaneously when an ether ply-yarn drill transmitted traffic, the output buffers of purpose ether port that ply-yarn drill is pegged graft can overflow because of receiving a large amount of Ethernet message simultaneously, Ethernet message will tie up the output buffers of other ports this moment, when the output buffers on route switching plane is taken in a large number, can't cause packet loss for route line card provides enough spaces cache way by message.
Again for example, Ethernet message as shown in Figure 3 influences the schematic diagram on route switching plane, if ether ply-yarn drill LC2 and route line card LC1 are simultaneously to another route line card LC0 transmitted traffic, suppose that LC1 and LC2 send message with 1G and the 10G flow port one by the Ethernet packet switching chip to LC0 respectively, flow from LC1 is represented with single arrow dotted line, represents with single arrow solid line from the flow of LC2.Ethernet message from LC2 is not dispatched, can take a large amount of outlet bandwidths of port one, suppose to take the 9G outlet bandwidth, the so actual only surplus 1G of outlet bandwidth that leaves the route switching plane for, and still dispatch according to the 10G outlet bandwidth on the route switching plane, allows LC1 that the output buffers of 10G flow entry port 1 is arranged.So, this port one can receive the flow of 9G+10G simultaneously, and outlet bandwidth has only 10G, and occur congested this moment, and the output buffers that causes port one is because of overflowing packet loss.At this moment, because the output buffers of port one is not enough, also might take other port output bufferses.
As seen, in the prior art,, cause two exchange planes in exchange process, can influence each other, do not realize really also that therefore a switching equipment has 2 exchange planes simultaneously because output buffers is shared.
Summary of the invention
In view of this, the invention provides a kind of switching network message switching method, can on a switching equipment, realize a plurality of exchange planes.
This method comprises: but according to the tie-in line Card Type of switch, in switch, have the exchange plane of separate planes buffer memory for the setting of every class ply-yarn drill;
Described switch will be received from the source ply-yarn drill and the different message of purpose ply-yarn drill type of ply-yarn drill, and plane conversion device is given in exchange;
Described plane conversion device sends in the switch the corresponding exchange plane of purpose ply-yarn drill type with this message with the message that receives, and is exchanged to corresponding purpose ply-yarn drill by exchange plane.
The present invention also provides a kind of switching system, can realize a plurality of exchange planes simultaneously.
This system comprises switch and the plane conversion device that is plugged on the switch; Described switch comprises at least 2 exchange planes that have the separate planes buffer memory respectively, and each exchange plane is corresponding with a class ply-yarn drill that can be plugged on self;
Described switch is used for and will be received from the source ply-yarn drill and the different message of purpose ply-yarn drill type of ply-yarn drill, and described plane conversion device is given in exchange; To be received from the message of described plane conversion device, exchange to the purpose ply-yarn drill by the exchange plane corresponding with the purpose ply-yarn drill type of this message;
Described plane conversion device, the message that is used for being received from described switch send to the corresponding exchange plane of purpose ply-yarn drill type of described switch and this message.
The embodiment of the invention is that different exchange planes dispose independently plane buffer memory, has realized the mutual isolation between the different exchange planes, makes not tie up buffer memory between each exchange plane mutually.For the source ply-yarn drill message different with the purpose ply-yarn drill, directly in switch, do not exchange, but send to plane conversion device earlier, carry out plane conversion by plane conversion device, be sent back to the purpose exchange plane in the switch again.As seen, the embodiment of the invention can realize a plurality of exchange planes, but also can realize the intercommunication of a plurality of exchange planes, has realized that really a switching equipment has at least 2 exchange planes simultaneously.
Description of drawings
Fig. 1 is the structural representation of modular ethernet switches in the prior art.
The modular router structural representation that Fig. 2 realizes for the available technology adopting packet switching chip.
Fig. 3 is the schematic diagram that Ethernet message influences the route switching plane in the prior art.
Fig. 4 is that route line card sends message to route line card and ether ply-yarn drill in the embodiment of the invention, and the ether ply-yarn drill sends the process schematic diagram of message to route line card and ether ply-yarn drill.
Fig. 5 is the embodiment of the invention one LC0 sends a process chart from message to packet switching chip.
Fig. 6 is the process chart of packet switching chip in the embodiment of the invention.
Fig. 7 is the process chart of embodiment of the invention midplane conversion equipment.
The flow chart that Fig. 8 handles the message with service priority sign for embodiment of the invention midplane conversion equipment.
Fig. 9 is the embodiment of the invention two LC3 send a process chart from message to packet switching chip.
Figure 10 is the structural representation of switching network message switching system in the embodiment of the invention.
Figure 11 is the structural representation of packet switching chip 31 among Figure 10.
Figure 12 is the structural representation of packet switch unit 311 among Figure 11.
Figure 13 is the structural representation of Figure 10 midplane conversion equipment 34.
Figure 14 is the structural representation of route line card 32 among Figure 10.
Figure 15 is the structural representation of ether ply-yarn drill 33 among Figure 10.
Embodiment
The embodiment of the invention provides a kind of switching network message switching scheme, and its basic thought is: a plurality of exchange planes are set in switch, and the corresponding class ply-yarn drill of each exchange plane is for each exchange plane disposes independently plane buffer memory; Switch will be received from the source ply-yarn drill of the ply-yarn drill message switching different with purpose ply-yarn drill type to plane conversion device.Plane conversion device is the unit module that inserts switch and have the plane conversion function, and this plane conversion device sends to the message that receives in the switch and the corresponding exchange plane of this message purpose ply-yarn drill type, is exchanged to the purpose ply-yarn drill by exchange plane.
Preferably, switch can exchange to the purpose ply-yarn drill by the exchange plane corresponding with purpose ply-yarn drill type being received from the source ply-yarn drill and the identical message of purpose ply-yarn drill ply-yarn drill type of ply-yarn drill, thereby realizes the exchanges data in the same exchange plane.
Preferably, when comprising the exchange plane that does not allow packet loss in a plurality of exchange planes that are provided with, plane conversion device also needs according to the schedule information that does not allow the exchange plane of packet loss when sending, and control sends to the flow of the exchange plane that does not allow packet loss.
With switch access route line card and ether ply-yarn drill is example, and the switching network message switching scheme of the embodiment of the invention is provided with route switching plane and ether exchange plane in switch, and these two exchange planes have independently plane buffer memory, does not share mutually.Switch is when exchanging, and the message that source ply-yarn drill and purpose ply-yarn drill is route line card exchanges to the purpose ply-yarn drill by the route switching plane, and the message that source ply-yarn drill and purpose ply-yarn drill is the ether ply-yarn drill exchanges to the purpose ply-yarn drill by the ether exchange plane.For the source ply-yarn drill message different with the purpose ply-yarn drill, at first exchange to plane conversion device, is that the message of ether ply-yarn drill is delivered the ether exchange plane in the changing device by plane conversion device with the purpose ply-yarn drill, is that the message of route line card sends to the route switching plane in the switch with the purpose ply-yarn drill; Importantly, when sending, send to the flow on route switching plane according to the schedule information control on route switching plane.
As seen, route switching plane and ether exchange plane have separate plane buffer memory in the embodiment of the invention, so route switching plane and ether exchange plane can not tie up buffer memory mutually, have realized the route switching plane and the ether exchange plane of mutual isolation.In addition, for the source ply-yarn drill message different with the purpose ply-yarn drill, directly in switch, do not exchange, but send to plane conversion device earlier, by plane conversion device this part message from the ether ply-yarn drill to route line card is applied scheduling, make the message that enters the route switching plane from plane conversion device, promptly enter the message on route switching plane from the ether ply-yarn drill, can not influence the exchange on route switching plane, thereby avoid the packet loss on route switching plane.As seen, the embodiment of the invention not only can realize route switching plane and ether exchange plane, but also can realize the intercommunication of route switching plane and Ethernet exchange plane.
Describe the present invention below in conjunction with the accompanying drawing embodiment that develops simultaneously.Switch is many in the prior art realizes that by the Ethernet packet switching chip embodiment of the invention also can be like this, below will adopt the Ethernet packet switching chip in return to install, and abbreviate packet switching chip as.
Beginning to describe before the embodiment, having the ether exchange plane of independent buffer memory and the realization on route switching plane is described in detail in packet switching chip, being provided with earlier.The mode that the embodiment of the invention adopts buffer memory to distribute realizes having the exchange plane division of separate planes buffer memory, below provides two kinds of implementations.
Mode one: fixing buffer memory apportion design
1) packet switching chip outbound course buffer memory distributes:
A1, output buffers in the total buffer memory of packet switching chip is partly distributed to each outbound port, distribution principle can adopt mean allocation.Each outbound port uses the buffer memory of distributing to oneself, and this each port comprises port that inserts ply-yarn drill and the port that inserts plane conversion device.For the ease of understanding, below will be called inbound port from the port that ply-yarn drill receives message, will be called outbound port to the port of ply-yarn drill outgoing message, those skilled in the art all can understand that a port can be outbound port and inbound port simultaneously.
The output buffers of b1, each outbound port is divided into two parts, is respectively formation of ether output buffers and the formation of route output buffers; Preferably, the formation of route output buffers is further divided into two parts, is respectively formation of route data output buffers and the formation of routing scheduling information output buffers.More than separate between three buffer queues that mark off for each outbound port, shared buffer memory space not.
The concrete spatial cache size of three output buffers formations of a port can be divided according to actual needs.Preferably, divide the formation of route data output buffers more greatly, so that provide bigger spatial cache for route data.
By the distribution of above-mentioned outbound course buffer memory, the route output buffers formation of all route line card institute access interface constitutes the plane buffer memory on route switching plane; The ether output buffers formation of all ether ply-yarn drill institute access interface constitutes the plane buffer memory of ether exchange plane.
2) the input direction buffer memory distributes:
The input direction buffer memory is used to buffer into the message to be exchanged that port receives, because the inner input-buffer of packet switching chip can be regarded infinity as to the bandwidth between the output buffers, therefore can finish the exchange of message between from the input-buffer to the output buffers in a short period of time.Therefore, the input-buffer place can not produce usually overflows, and input-buffer can adopt the structure of having only an input-buffer formation, also can the input direction buffer memory be distributed according to the outbound course buffer memory method of salary distribution, is specially:
A2, input-buffer in the total buffer memory of packet switching chip is partly distributed to each inbound port, distribution principle can adopt mean allocation.Each inbound port uses the buffer memory of distributing to oneself.
The buffer memory of b2, each inbound port is divided into two parts, is respectively formation of ether input-buffer and the formation of route input-buffer; Preferably, the formation of route input-buffer is further divided into two parts, is respectively formation of route data input-buffer and the formation of routing scheduling information input-buffer.More than separate between three buffer queues that mark off for each inbound port, shared buffer memory space not.The big I of three input-buffer formations is divided according to actual needs.
By the distribution of above-mentioned input direction buffer memory, the route input-buffer formation of all route line card institute access interface belongs to the route switching plane; The ether input-buffer formation of all ether ply-yarn drill institute access interface belongs to the ether exchange plane.
Adopt fixedly buffer memory apportion design, each port of packet switching chip can connect ether ply-yarn drill, route line card or plane switch, but can only connect a kind of equipment simultaneously.When port inserted route netting twine card, this port only used formation of route input-buffer and the formation of route output buffers, but formation of ether output buffers and the formation of ether input-buffer still exist; When port inserted the ether ply-yarn drill, this port only used formation of ether input-buffer and the formation of ether output buffers, but formation of route output buffers and the formation of route input-buffer still exist; Insert the port of plane conversion device, two kinds of output buffers formations and two kinds of input-buffer formation persons use.
Mode two: dynamic buffering apportion design
1) packet switching chip outbound course buffer memory distributes:
A3, output buffers in the total buffer memory of packet switching chip is partly distributed to each outbound port.This is the same with the fixing principle of buffer memory apportion design.
The ply-yarn drill that the output buffers of b3, outbound port all offers access uses.For with fixing buffer memory apportion design in the title of individual queue unified, below when port inserts the ether ply-yarn drill, the output buffers of port is called the formation of ether output buffers; When port inserts route line card, the output buffers of port still is called the formation of route output buffers.
A kind of implementation that output buffers all offers the ply-yarn drill use of access is:
It is A that routing scheduling information output buffers queue size is set;
It is B1 that route data output buffers formation minimum capacity is set, and heap(ed) capacity is B3; A and B3 sum are total output buffers capacity of port;
It is C1 that ether output buffers formation minimum capacity is set, and heap(ed) capacity is C3; C3 is total output buffers capacity of port.
When inserting the ether ply-yarn drill, packet has taken minimum capacity C1, at total output buffers capacity of port is arranged the free time, then under the situation that is no more than heap(ed) capacity C3, can hold over buffer memory, until reaching heap(ed) capacity C3.When inserting route line card, routing scheduling information can only take buffer memory capacity A, and route data packets has taken minimum capacity B1, total output buffers capacity at port has the free time, then under the situation that is no more than heap(ed) capacity B3, can hold over buffer memory, until reaching heap(ed) capacity B3.
2) the input direction buffer memory distributes:
With fixedly the buffer memory method of salary distribution is similar, the input direction buffer memory can not divided, also can input-buffer in the total buffer memory of packet switching chip partly be distributed to each inbound port according to outbound course buffer memory dynamic assignment mode, each port uses the buffer memory of distributing to self.The ply-yarn drill that the input-buffer of each port all offers access uses.
Need to prove that the above dynamic buffering method of salary distribution is at the port that inserts ply-yarn drill.And, need to adopt the fixing buffer memory method of salary distribution that the input-buffer and the output buffers of this port carried out the formation division for the port that plane conversion device inserts.
In fact, owing to enter the restriction that the message of plane conversion device does not allow packet loss, therefore, the input-buffer of plane conversion device institute access interface and output buffers can not carry out formation and divide.
The embodiment of the invention has only been enumerated two kinds of modes of dividing the plane buffer memory, in practice, can also adopt other modes to carry out buffer memory and divide, as long as make different plane buffer memorys not share mutually.
In order to cooperate the exchange of packet switching chip, can also and receive buffer memory to the transmission buffer memory on the plane conversion device and be provided with.Wherein, receive buffer memory and be used for the message that buffer memory receives from packet switching chip, send buffer memory and be used for the message that buffer memory sends to packet switching chip.Specifically, transmission buffer memory in the plane conversion device and reception buffer memory are distributed according to following rule:
1) sending buffer memory distributes:
Transmission buffer memory in the plane conversion device is divided into route sends buffer queue and ether transmission buffer queue; Ether sends buffer queue and is used for the message that buffer memory sends to packet switching chip ether exchange plane.Route sends buffer queue and is used for the message that buffer memory sends to packet switching chip route switching plane, and this message does not comprise schedule information, and schedule information is handled by the distributed scheduler that is arranged in the plane conversion device.Distributed scheduler in this plane conversion device is identical with distributed scheduler in the route line card, when carrying out the routing scheduling information interaction, other route line cards are regarded plane conversion device as a route line card, and the flow from the route message of plane conversion device is limited.Distributed scheduler in the plane conversion device can be sent to the flow of oneself by limiting other ply-yarn drills to other route line cards transmission schedule informations; Also can not limit other ply-yarn drills and be sent to the flow of oneself, because the output buffers formation of plane conversion device institute access interface does not belong to the route switching plane, the message that is sent to plane conversion device allows packet loss.
Route sends buffer queue and ether sends between the buffer queue and can dispatch according to scheduling strategy between the formation of setting, for example, and equity dispatching (RR) strategy, weight scheduling (WRR) strategy etc.Also can send the buffer queue priority principle, send the message in the route transmission buffer queue earlier, send the message in the ether transmission buffer queue again according to route.
Need to prove, the purpose that the transmission buffer memory is divided is: the distributed scheduler in the plane conversion device can be according to schedule information, the route data message that route is sent in the buffer queue applies scheduling targetedly, makes the message flow that enters the route switching plane controlled; Scheduling strategy between various formations can also implemented between formation.In practice, because plane conversion device is sent to the message of packet switching chip and allows packet loss, therefore can not carry out the division that route sends buffer queue and ether transmission buffer queue to sending buffer memory yet, so when applying flow control, as long as be that the message of route line card carries out flow control to the purpose ply-yarn drill according to the mark that carries in the message.
2) receiving buffer memory distributes:
Owing to allow packet loss in the plane conversion device, so the reception buffer memory in the plane conversion device can not divided formation.Certainly, also the reception buffer memory in the plane conversion device can be divided into route and receive buffer queue and ether reception buffer queue.
More than in two kinds of buffer memory methods of salary distribution, formation of route input-buffer and the formation of route output buffers all are divided into the formation of buffer memory route data packets and buffer memory routing scheduling information, and when scheduling, scheduling routing scheduling information guarantees the transmission of routing scheduling information priority earlier.In order to simplify description, when following examples are not distinguished these two formations, also be defaulted as priority scheduling routing scheduling information.
Begin description below to embodiment.
Embodiment one
Fig. 4 shows route line card and sends message to route line card and ether ply-yarn drill, and the ether ply-yarn drill sends the process schematic diagram of message to route line card and ether ply-yarn drill.In the present embodiment, the process of the transmission of the route line card shown in Fig. 4 message to route line card and ether ply-yarn drill specifically described.LC0 among Fig. 4 and LC1 are route line cards, and LC2 and LC3 are the ether ply-yarn drills.Single arrow fine dotted line represent LC0 sends message flow from message to LC1 to, single arrow fine line represent LC0 sends message flow from message to LC2 to.
At first the handling process to LC0 in the present embodiment is described.Fig. 5 shows route line card LC0 in the embodiment of the invention one and sends the process chart of message to packet switching chip, and as shown in Figure 5, this method may further comprise the steps:
After step 501:LC0 inserts packet switching chip, from packet switching chip obtain to transmit, the outbound port mark of route plane mark and plane conversion device.
In this step, LC0 carries out information exchange with packet switching chip after inserting packet switching chip, informs packet switching chip self ply-yarn drill type and relevant information, obtains from packet switching chip and transmits.This transmits the corresponding relation between the information such as destination address, message outbound port mark and purpose ply-yarn drill type that comprise message at least.Wherein, the type of purpose ply-yarn drill comprises ether ply-yarn drill and route line card.Route plane mark is the exchange plane mark, and the message that expression route line card LC0 sends should enter the route switching plane in packet switching chip.
In fact, LC1, LC2 and LC3 also obtain to transmit after inserting packet switching chip.
Step 502:LC0 receives message to be exchanged, and searches and transmits, and adds route plane mark and according to transmitting the outbound port mark of definite message in packet switching chip waiting to exchange in the message.
Step 503: judge whether the purpose ply-yarn drill type wait to exchange message is the route line card identical with self, if not, execution in step 504; If, execution in step 505.
Step 504: waiting to exchange the outbound port mark that adds plane conversion device in the message.Execution in step 505.
Step 505: message to be exchanged is sent to packet switching chip.
LC0 is a route line card, and when it waited to exchange message in transmission, the schedule information on the route switching plane of safeguarding according to the distributed scheduler among the LC0 was implemented scheduling operation to message to be sent.Obtain schedule information and implement to be scheduling to existing operation according to schedule information, be specifically as follows: the distributed scheduler of LC0 is according to the schedule information that is received from LC1, the flow that is defined as self distributing to LC1 transmission packet, with flow rate conversion is token number, judges whether to allow to send current message according to token number.
Need to prove, wait to exchange in the process of message,, then immediately schedule information is encapsulated and preferentially send if the distributed scheduler of LC0 has generated schedule information in transmission.Preferably, during encapsulation, in the schedule information message, add high priority route plane mark, thereby distinguish mutually with the route data message.Accordingly, the route plane in the route data message is labeled as low priority route plane mark.
So far, this flow process finishes.
Referring to Fig. 4, suppose LC0 receive from the outside wait exchange message A after, with the destination address of waiting to exchange message A is index, search swap table, find and wait to exchange message A corresponding output port mark P1 (port one), the ply-yarn drill type of purpose ply-yarn drill LC1 is the route line card identical with LC0, and the route plane is labeled as 2.This moment, LC0 was waiting to exchange adding mark P1 and mark 2 among the message A, sent to packet switching chip then after encapsulation.Mark 2 can be carried to be waited to exchange in the switching network packet header of message A.Wherein, switching network packet header is before message enters switching network, the switching network packet header that meets switching network port institute supported protocol that adds outside the IP head.
Still referring to Fig. 4, LC0 receive from the outside wait exchange message B after, search swap table, find and wait to exchange message B corresponding output port and be labeled as P2 (port 2), the ply-yarn drill type of purpose ply-yarn drill LC2 is the ether ply-yarn drill different with LC0, and the route plane is labeled as 2.This moment, LC0 was waiting to exchange the outbound port mark P5 that adds mark P2, mark 2 and plane conversion device among the message B, sent to packet switching chip then after encapsulation.Mark P5 can be carried at and wait to exchange in the switching network packet header of message B; Also can carry out the secondary encapsulation to the message after once encapsulating, the outbound port mark P5 of outer encapsulation plane conversion device, internal layer encapsulates actual outbound port mark P2, to realize carrying of two outbound port marks.
Present embodiment is to stamp respective markers for message to be exchanged in route line card.In practice, also can stamp respective markers for message to be exchanged, exchange according to the exchange principle again by packet switching chip.
Fig. 6 is the process chart of packet switching chip in the embodiment of the invention.This handling process all adopts fixedly at the input-buffer of plane conversion device institute access interface and output buffers, and the situation of the buffer memory method of salary distribution is described.As shown in Figure 6, this method may further comprise the steps:
Step 600: the inbound port of packet switching chip receives message.The message that is received comprises outbound port mark and exchange plane mark, may also comprise the outbound port mark of plane conversion device.In the present embodiment, the exchange plane of received packet is labeled as route plane mark.
Step 601: received packet is cached in the corresponding input-buffer formation of the exchange plane mark that carries with received packet in the inbound port.
This step judges that whether input-buffer formation to be buffered into also has free space, if do not have, then abandons the message of reception before buffer memory; If free space is arranged, then the message outbound port mark that carries according to received packet or the outbound port mark of plane conversion device, judge whether the corresponding output buffers formation of exchange plane mark of carrying with received packet in the current outbound port also has free space, if do not have, then abandon the message of reception, if free space is arranged, then with the packet buffer that receives in the formation of route input-buffer.Wherein, when received packet is carried the outbound port mark of plane conversion device, the port of the outbound port mark indication that its current outbound port is a plane conversion device; If do not carry the outbound port mark of plane conversion device, then its current outbound port is the port of message outbound port mark indication.
Still Figure 4 shows that example, what port 0 received waits that exchanging message A carries out port label P1 and route plane mark 2, therefore, its current outbound port is a port one, then when judging that port 0 route input-buffer formation and the formation of port one route output buffers all have free space, will wait to exchange message A and be cached to port 0 route input-buffer formation.
Port 0 receives waits to exchange outbound port mark P5, outbound port mark P2 and the route plane mark 2 that message B carries plane conversion device, therefore, its current outbound port is a port 5, then when judging that port 0 route input-buffer formation and port 5 route output buffers formations all have free space, will wait to exchange message B and be cached to port 0 route input-buffer formation.
As previously mentioned, the formation of route input-buffer specifically comprises formation of route data input-buffer and the formation of routing scheduling information input-buffer.Therefore, this step will be carried the schedule information message of high priority route plane mark when packet buffer is arrived the formation of route input-buffer, be cached in the formation of routing scheduling information input-buffer; To carry the route data message of low priority route plane mark, be cached in the formation of route data input-buffer.
Need to prove that no matter the port institute docking device of packet switching chip is route line card, ether ply-yarn drill or plane conversion device, all according to handling from the flow process of step 601 beginning.
Step 602: with the message switching in each inbound port input-buffer formation to corresponding outbound port.
The exchange principle is:
If message to be exchanged carries the outbound port mark of plane conversion device, show that the purpose ply-yarn drill of waiting to exchange message is different with source ply-yarn drill type, need plane conversion device to handle, therefore, this is waited to exchange in the corresponding output buffers formation of exchange plane mark that message switching carries with message to the outbound port that plane conversion device is pegged graft, and present embodiment is the formation of route output buffers.Before the exchange outbound port mark of plane conversion device is deleted from message, or delete by plane conversion device, guarantee this message when plane conversion device exchanges back packet switching chip, packet switching chip can exchange processing according to actual message outbound port mark.
If wait to exchange the outbound port mark that message does not carry plane conversion device, show that the purpose ply-yarn drill of waiting to exchange message is identical with source ply-yarn drill type, it all is route line card, the then direct message outbound port mark that carries according to message to be exchanged, to wait to exchange in the corresponding output buffers formation of exchange plane mark that message switching carries with message to the corresponding outbound port, present embodiment be the formation of route output buffers.
Still Figure 4 shows that example, treating in the port 0 route input-buffer formation exchanges message A and carries out port label P1 and route plane mark 2, handles through exchange, is cached in the route output buffers formation of port one; Wait to exchange outbound port mark P5, outbound port mark P2 and the route plane mark 2 that message B carries plane conversion device, handle, be cached in the port 5 route output buffers formations through exchange.
Step 603: each port in the packet switching chip is exported the message in its output buffers formation: the outbound port of access route line card is exported the message in its route output buffers formation, and the outbound port of access ether ply-yarn drill is exported the message in its ether output buffers formation; The outbound port that inserts plane conversion device is according to scheduling strategy between the formation of setting, and the message in route output buffers formation and the formation of ether output buffers is exported in scheduling.
When output, for the port that inserts route line card and ether ply-yarn drill, no matter fix or the dynamic buffering distribution, have only in the output buffers formation data are arranged, only need this moment processing is exported in the formation that data are arranged.For port with the formation of route output buffers, preferentially enter plane conversion device in order to guarantee schedule information, preferably, the schedule information message in the formation of priority scheduling routing scheduling output information output buffers.
Wherein, be to allow packet loss owing to mail to the message of plane conversion device, so scheduling strategy can be RR scheduling strategy, WRR scheduling strategy or priority scheduling strategy between formation.Wherein, the RR scheduling strategy was an equity dispatching, with two formations of scheduling ratio scheduling of 1: 1; The WRR scheduling strategy is the weight scheduling, and with the scheduling ratio of setting, for example 1: 9, scheduling ether sends buffer queue and route sends buffer queue; The priority scheduling strategy is that formation of route output buffers and the formation of ether output buffers are provided with priority, the formation that priority scheduling priority is high.
So far, this flow process finishes.
Route line card LC1 and ether ply-yarn drill LC2 are same as the prior art to the reception operation from the packet switching chip message in the present embodiment.Need to prove that LC1 need consider plane conversion device as a flow transmission source when generating schedule information, be the flow restriction of plane conversion device configuration, and notify plane conversion device by sending schedule information to self transmitted traffic.
Processing to plane conversion device is described in detail below.Fig. 7 shows the handling process of present embodiment midplane conversion equipment.It is that example is described that this handling process is all divided formation at the reception buffer memory of plane conversion device with the transmission buffer memory.As shown in Figure 7, this flow process may further comprise the steps:
Step 701: plane conversion device receives the message from packet switching chip, and the exchange plane mark according to received packet is carried is cached to the reception buffer queue corresponding with entrained exchange plane mark with received packet.If it is full to receive buffer queue, then with the packet loss that receives.
Still Figure 4 shows that example, plane conversion device receives message B, and message B carries route plane mark, and plane conversion device is cached to route with it and receives in the buffer queue.
Step 702: route is received buffer queue and ether receive message in the buffer queue and carry out conversion between route plane mark and the ether plane mark, buffer memory is to the transmission buffer queue of conversion back exchange plane mark correspondence then.
Still Figure 4 shows that example, plane conversion device receives the message B that carries route plane mark, message B is cached to route receives buffer queue, when handling message B then, its route plane mark 2 that carries is converted to ether plane mark 3, this moment, message B carried out port label P1 and ether plane mark 3, was cached to route and sent in the buffer queue.
If desired the message from route line card is carried out format conversion, make it meet ether exchange requirement, then this format conversion operation is finished in this step.
Step 703:, route transmission buffer queue and ether transmission buffer queue are dispatched according to scheduling strategy between the formation of setting; When the message in the scheduling output route transmission buffer queue,, control its transmitted traffic according to the schedule information that is received from the route switching plane.
Wherein, scheduling strategy is RR scheduling strategy, WRR scheduling strategy or priority scheduling strategy between formation.In the present embodiment, the priority scheduling strategy is: the priority scheduling route sends buffer queue, when not having message in the route transmission buffer queue, dispatches ether again and sends buffer queue.
Adopt scheduling strategy between formation, when being dispatched to ether transmission buffer queue, the message to be sent that ether is sent in the buffer queue sends to packet switching chip; When being dispatched to route and sending buffer queue, the schedule information on the route switching plane of safeguarding according to distributed scheduler is implemented scheduling operation.
So far, this flow process finishes.
Plane conversion device can have influence on plane conversion device route switching plane and the shared bandwidth of ether exchange plane transmitted traffic in packet switching chip to the scheduling strategy that sends buffer queue.When ether exchange plane transmitted traffic was very little, it is reasonable inadequately that employing RR dispatches obvious allocated bandwidth to route switching plane transmitted traffic is very big when plane conversion device.When adopting the WRR scheduling strategy, its weight fixedly installs, and weight is provided with unreasonable, the unreasonable situation of allocated bandwidth also can occur.And when adopting route to send the scheduling of buffer memory priority principle, if constant to route switching plane transmitted traffic, may cause delaying to be sent out away to the flow that the ether exchange plane sends, serious packet loss appears.
Because of plane conversion device sends the many disadvantages that scheduling strategy that buffer queue and route send buffer queue rationally brings inadequately to ether, the embodiment of the invention adopts following steps to determine scheduling strategy for fear of above-mentioned:
Step 2, the detection plane conversion equipment actual flow that route switching plane and ether exchange plane send in switch.Realization can send buffer queue and ether sends the actual output flow of buffer queue for: plane conversion device statistics route; Perhaps, in purpose ply-yarn drill side, LC2 side is for example carried out traffic statistics to the message of being received from plane conversion device, and the result feeds back to plane conversion device with traffic statistics.Certainly, if packet switching chip has enough disposal abilities to bear the flow detection task, the flow monitoring task of this step 2 can be finished by detecting port flow by packet switching chip.
Step 3, plane conversion device are adjusted the current scheduling ratio according to the actual flow of route switching plane in switch and the transmission of ether exchange plane.For example, the flow that is sent to the route switching plane is called the route message flow, the flow that is sent to the ether exchange plane is called the Ethernet message flow, then when judging the Ethernet message flow less than preset first threshold value, during for example less than 1G, reduce the current scheduling ratio, make the route switching plane can utilize idle bandwidth; During less than second threshold value, during for example less than 5G, improve the current scheduling ratio at the route message flow, make the ether exchange plane can utilize idle bandwidth.Again for example, at the ratio of judging Ethernet message flow and the total flow of Ethernet message flow and route message flow during less than the 3rd threshold value, for example 1/20, show that the Ethernet message flow is very little, reduce the current scheduling ratio; At the ratio of judging route message flow and total flow during less than the 4th threshold value, for example 5/10, show that the route message flow is very little, improve the current scheduling ratio.Again for example, at the ratio of judging Ethernet message flow and route message flow less than the 5th threshold value, for example 0.1/8, and the route message flow has reached the maximum of permission, show the bigger flow of route message needs, and Ethernet message only needs very low discharge, at this moment, reduce the current scheduling ratio; At the ratio of judging Ethernet message flow and route message flow greater than the 6th threshold value, for example 2/3, and the Ethernet message flow reached the maximum of permission, at this moment, scheduling ratio before improving.
Step 4, after adjusting the scheduling ratio, can after surpassing Preset Time, the scheduling ratio be reverted to the initial schedule ratio.
In practice, can also carry the service priority sign in the message.The schematic flow that Fig. 8 handles the message with service priority sign for plane conversion device.As shown in Figure 8, each buffer queue in the plane conversion device is by the subqueue of the corresponding different business priority of difference.Plane conversion device has following 6 treatment steps after receiving message, and is specific as follows:
1. receive message, carry out message classification according to exchange plane mark in the message and service priority sign, so which priority the message that identification receives belongs to from which exchange plane from packet switching chip.
2. according to the message classification result, message is deposited in the reception buffer queue corresponding with the exchange plane mark and the corresponding reception buffer memory subqueue of service priority sign.Fig. 8 midplane conversion equipment the first half empty lattice represents that ether receives buffer queue, and the lattice that has oblique line to fill represents that route receives buffer queue; Plane conversion device the latter half empty lattice represents that ether sends buffer queue, and the lattice that has oblique line to fill represents that route sends buffer queue.
3. route reception buffer queue and ether receive buffer queue a scheduler are set respectively, according to the subqueue in the corresponding reception buffer queue of scheduling by service priority.
4. the message that is scheduled out is carried out the message conversion: route plane mark and ether plane mark are exchanged; Former service priority can also be converted to professional priority in the purpose exchange plane under the message.
5. according to new exchange plane mark and service priority sign, will identify corresponding subqueue with service priority in the transmission buffer queue that the message after the conversion deposits in newly the exchange plane mark is corresponding.
6. route sends buffer queue and ether transmission buffer queue employing second-level dispatching.Transmit port is provided with a Port Scheduling device, and each exchange plane is provided with respectively one and goes out group scheduler.The Port Scheduling device is according to the transmission buffer queue of scheduling strategy scheduling Different Plane between formation; Scheduling strategy between by default formation, when being dispatched to route and sending buffer queue, the route switching plane go out group scheduler according to service priority, route is sent the dispatching message of each subqueue goes out team in the buffer queue; When being dispatched to ether and sending buffer queue, the ether exchange plane go out group scheduler according to service priority, ether is sent the dispatching message of each subqueue goes out team in the buffer queue.Port physical layer sends to packet switching chip with the data of dispatching out team.
More than 6 steps are processes that plane conversion device is handled the message with service priority sign.Correspondingly, various buffer queues in the packet switching chip also can further be divided into subqueue according to service priority, after which kind of formation the mark that packet switching chip carries according to message determines, further determine to get final product into which subqueue according to the sign of the service priority in the message.When going out group, also need further to go out team according to priority scheduling.
Embodiment two
Present embodiment is described the process of the transmission of the ether ply-yarn drill shown in Fig. 4 message to route line card and ether ply-yarn drill.Identical with embodiment one, the purpose ply-yarn drill still is route line card LC1 and ether ply-yarn drill LC2 in the present embodiment, and difference is that the source ply-yarn drill of present embodiment is ether ply-yarn drill LC3.Single arrow thick dashed line represent LC3 sends message flow from message to LC2 to, single arrow heavy line represent LC3 sends message flow from message to LC1 to.
Present embodiment is identical with embodiment one to the distribution of packet switching chip inner buffer.
Fig. 9 shows ether ply-yarn drill LC3 in the embodiment of the invention two sends process chart from message to packet switching chip.As shown in Figure 9, this method may further comprise the steps:
After step 901:LC3 inserts packet switching chip, from packet switching chip obtain to transmit, the outbound port mark of ether plane mark and plane conversion device.
Step 902:LC3 receives message to be exchanged, and searches and transmits, and adds ether plane mark and according to transmitting the outbound port mark of definite message in packet switching chip waiting to exchange in the message.The message that carries this ether plane mark will be identified as at the ether exchange plane by packet switching chip and exchange.
Step 903: judge whether the purpose ply-yarn drill type wait to exchange message is the ether ply-yarn drill identical with self, if not, execution in step 904; If, execution in step 905.
Step 904: waiting to exchange the outbound port mark that adds plane conversion device in the message.Execution in step 905.
Step 905: message to be exchanged is sent to packet switching chip.
So far, this flow process finishes.
The handling process of packet switching chip is identical with Fig. 6 in the present embodiment two.The handling process of plane conversion device is identical with Fig. 7.
Referring to Fig. 4, suppose that the LC3 transmission purpose ply-yarn drill of present embodiment two is the message C of LC2 and the message D that the purpose ply-yarn drill is LC1.Present embodiment LC3 and packet switching chip are as follows to the handling process of message C and D:
For message C, suppose LC3 receive wait to exchange message C after, search and transmit, finding the corresponding outbound port of message C is P2 (port 2), the ply-yarn drill type of purpose ply-yarn drill LC2 is the ether ply-yarn drill identical with LC3, and the ether plane is labeled as 3.At this moment.LC3 adds mark P2 and mark 3 in message C, and sends to packet switching chip after the encapsulation.After the port 3 of packet switching chip receives message C, according to its mark that carries 3 it is buffered into port 3 ether input-buffer formations and wait for the exchange processing, when message C is exchanged when handling, mark P2 that packet switching chip carries according to message C and mark 3 exchange to it and wait in port 2 ether output bufferses and output to LC2.
For message D, suppose LC3 receive wait to exchange message D after, search and transmit, finding the corresponding outbound port of message D is P1 (port one), the ply-yarn drill type of purpose ply-yarn drill LC1 is the route line card different with LC3, and the ether plane is labeled as 3.At this moment, LC3 adds the outbound port mark P5 of mark P1, mark 3 and plane conversion device in message D, sends to packet switching chip after encapsulation.After the port 3 of packet switching chip receives message D, according to its mark that carries 3 it is buffered into port 3 ether input-buffer formations and wait for the exchange processing, when message D is exchanged when handling, mark P5 that packet switching chip carries according to message C and mark 3 exchange to it and wait in port 5 ether output bufferses and output to plane conversion device.After message D enters plane conversion device, plane conversion device is according to its mark that carries 3, it is cached to ether receives the medium processing to be converted of buffer queue, when conversion, the mark 3 of message D is replaced with 2 and with the P5 deletion, and buffer memory to route sends buffer queue etc. and is sent to packet switching chip then.The port 5 of packet switching chip receives message D once more, and the mark 2 that carries according to message D is buffered in port 5 route input-buffer formations wait exchange processing with it.In when exchange, mark P1 that carries according to message D and mark 2 exchange to the formation of port one route output buffers with it and wait for and output to LC1.
More than among two embodiment the input-buffer formation can be the real cache formation, also can be the virtual cache formation.When being the virtual cache formation, the virtual cache formation is made up of counter and controller, utilizes counter and controller can manage the cache size of input-buffer formation.For example, when the buffer memory that takies when the incoming message of inbound port had surpassed the maximum count value that counter allowed, controller abandoned subsequent packet.The virtual cache formation be embodied as the known technology means, be not described in detail here.
Can sum up by above two embodiment and to obtain, packet switching chip according to the exchange principle that message carries mark is: at first ply-yarn drill should add and the corresponding exchange plane mark of self type for message, if the message source ply-yarn drill is different with the purpose ply-yarn drill, then add the plane conversion device sign again; Outbound port mark and message outbound port sign as for plane conversion device can be determined by packet switching chip.Packet switching chip is given plane conversion device with the message switching of carrying the plane conversion device sign after receiving message then; Give the output buffers formation corresponding in the corresponding outbound port with the message switching of not carrying the plane conversion device sign with the exchange plane type; The plane conversion device that receives message will the plane conversion device corresponding with the message source ply-yarn drill be converted to corresponding to the purpose ply-yarn drill, returns then to packet switching chip, and when sending, control enters the flow on route switching plane.And the message switching that packet switching chip is sent plane conversion device gets final product for the corresponding output buffers formation of exchange plane type of carrying with message in the corresponding outbound port.
Adopt the switching method of the embodiment of the invention, can solve the prior art problem shown in Fig. 3, still referring to Fig. 3, LC1 and LC2 send the message of 10G and 9G respectively to LC0,10G route message enters the route switching plane and is cached to the port one route and sends buffer queue, the 9G Ethernet message enters the ether exchange plane also by after the conversion of plane conversion device, enters the route switching plane of packet switching chip and be cached to the port one route to send in the buffer queue under the Flow Control of plane conversion device.Because LC0 all has control to the flow that LC1 and plane conversion device send to self, therefore, the source ply-yarn drill is that the message of LC2 can uncontrolled entry port 1 route not send buffer queue, thereby ties up the spatial cache that should be used to handle from the LC1 message.If the message from LC2 is too much, these messages can be abandoned by plane conversion device before mouth 1 route in inlet side sends buffer queue, therefore can not influence route switching plane exchange process.As seen, the message switching method of the embodiment of the invention can solve the problems of the prior art.
In order to realize switching network message switching method of the present invention, the embodiment of the invention also provides a kind of switching network message switching system.Figure 10 is the structural representation of switching network system in the embodiment of the invention.As shown in figure 10, this system comprises switch 31 and plane conversion device 34.
If switch 31 adopts application-specific integrated circuit (ASIC) to realize, just be called packet switching chip 31.The port of packet switching chip 31 can insert ply-yarn drill peace face conversion equipment 34.Packet switching chip 31 comprises at least 2 exchange planes that have the separate planes buffer memory respectively, and each exchange plane is corresponding with a class ply-yarn drill that can be plugged on self.
Packet switching chip 31 is used for the source ply-yarn drill message identical with the ply-yarn drill type of purpose ply-yarn drill that will be received from ply-yarn drill, exchanges to the purpose ply-yarn drill by the exchange plane corresponding with purpose ply-yarn drill type; With source ply-yarn drill and the different message of purpose ply-yarn drill type, plane conversion device 34 is given in exchange; To be received from the message of plane conversion device 34, exchange to the purpose ply-yarn drill by the exchange plane corresponding with the purpose ply-yarn drill type of this message.
Plane conversion device 34, the message that is used for being received from packet switching chip 31 send to the corresponding exchange plane of purpose ply-yarn drill type of packet switching chip 31 and this message; When sending, according to the schedule information of the exchange plane that does not allow packet loss, control sends to the flow of the exchange plane that does not allow packet loss.
With packet switching chip 31 access route line cards 32 and ether ply-yarn drill 33 is example, referring to Figure 10, and interaction data bag and schedule information between route line card 32 and the packet switching chip 31; Interaction data bag between ether ply-yarn drill 33 and the packet switching chip 31; Interaction data bag and schedule information between plane conversion device 34 and the packet switching chip 31.Single arrow solid line among Figure 10 between packet switching chip 31 and the ply-yarn drill is represented packet, and the double-head arrow dotted line is represented schedule information.
Packet switching chip 31 provides route switching plane and ether exchange plane, and each exchange plane has independently plane buffer memory; With source ply-yarn drill and purpose ply-yarn drill is that the message of route line card exchanges to the purpose route line card by the route switching plane, is that the message of ether ply-yarn drill exchanges to purpose ether ply-yarn drill at the ether exchange plane with source ply-yarn drill and purpose ply-yarn drill; For from the source ply-yarn drill of the ply-yarn drill message different with the purpose ply-yarn drill, exchange to plane conversion device 34, will be received from the message of plane conversion device 34 again, exchange to the purpose ply-yarn drill by the exchange plane corresponding with purpose ply-yarn drill type.
And will being received from the message on route switching plane, plane conversion device 34 sends in the ether exchange plane; The message that is received from the ether exchange plane is sent in the route switching plane, and when sending, according to routing scheduling information Control transmitted traffic.Wherein, routing scheduling information is by obtaining alternately with route line card 32, and its reciprocal process is same as the prior art, can realize as long as distributed scheduler is set in plane conversion device 34.
Figure 11 shows the structural representation of packet switching chip 31 among Figure 10, and as shown in figure 11, this packet switching chip 31 comprises packet switch unit 311, inserts a plurality of port units 312 of ply-yarn drill and the port unit 313 of access plane conversion device 34; Wherein,
For the port unit 312 that inserts ply-yarn drill, be configured to when inserting route line card, to have independently route output buffers formation 1102, when inserting the ether ply-yarn drill, have independently ether output buffers formation 1101.A port unit 312 can be done inbound port and/or outbound port.The route switching plane has been formed in each route output buffers formation 1102 that inserts in the port unit 312 of route line card, and each inserts in the port unit 312 of ether ply-yarn drill ether output buffers formation 1101 and has formed the ether exchange plane.As described in method embodiment, can be by the fixedly configuration of buffer memory apportion design or dynamic buffering apportion design realization output buffers formation.When adopting fixedly the buffer memory apportion design, at first dispose independently output buffers for each port unit 312, then output buffers is divided into separate ether output buffers formation 1101 and route output buffers formation 1102, and the capacity of two output buffers formations is fixed.When adopting the dynamic buffering apportion design, the output buffers of each port unit 312 is divided into separate ether output buffers formation 1101 and route output buffers formation 1102, when inserting route line card, ether output buffers formation 1101 capacity are 0, when inserting the ether ply-yarn drill, route output buffers formation 1102 capacity are 0.Only show a port unit 312 among Figure 11, the concrete structure when promptly port unit 2 is as outbound port, its structure was all identical when those skilled in the art can understand each port unit 312 as outbound port.
This port unit 312 also comprises output scheduling module 1103, is used for exporting according to the order of sequence the message of the output buffers formation that has data.For the preferential schedule information that sends in the route output buffers formation 1102, preferably, route output buffers formation 1102 further is divided into the route data output buffers formation that is used for the buffer memory route data packets and is used for the schedule information output buffers formation (divisions of Figure 11 not shown this two formations) of cache way by schedule information.In this case, output scheduling module 1103 further dispatching priorities send the routing scheduling information in the formation of routing iinformation output buffers, and to be sent finishing sends the route data packets in the formation of route data output buffers again.
When port unit 312 during as inbound port, its input-buffer can not divided formation or divide formation.A port unit 312 has been shown among Figure 11, the concrete structure when promptly port unit 1 is as inbound port and division formation, its structure was all identical when those skilled in the art can understand each port unit 312 as inbound port.As shown in figure 11, port unit 312 comprises message field sub-module 1113 and input-buffer, and input-buffer is divided into separate ether input-buffer formation 1111 and route input-buffer formation 1112; As described in method embodiment, the division of input-buffer can be according to fixedly buffer memory apportion design or dynamic buffering apportion design.The input-buffer of different port unit 312 is separate.
Message field sub-module 1113 is used for according to the type that inserts ply-yarn drill, and it is to be sent that the message that route line card is sent sends to route input-buffer formation 1112 etc., and it is to be sent that the message that the ether ply-yarn drill is sent sends to ether input-buffer formation 1111 etc.There is the formation of message to send message in self formation.
Ether input-buffer formation 1111 and route input-buffer formation 1112 can also can be the virtual cache formation for real buffer queue, if adopt the virtual cache formation, then each virtual cache formation comprises counter and controller, when the buffer memory that takies when the incoming message of inbound port had surpassed the maximum count value that counter allowed, controller abandoned subsequent packet.In embodiments of the present invention, the maximum count value of counter can be set to the fixed size that buffer queue distributes.
For the port unit 313 that inserts plane conversion device 34, this port unit 313 has independently input-buffer and output buffers, input-buffer and output buffers can not divided formation, perhaps input-buffer is divided into separate ether input-buffer formation 1107 and route input-buffer formation 1108, output buffers is divided into separate ether output buffers formation 1104 and route output buffers formation 1105 according to fixing buffer memory apportion design.The embodiment of the invention adopts the formation mode of dividing.
This port unit 313 also comprises message field sub-module 1109 and scheduler module 1106;
Message field sub-module 1109 is used for the exchange plane mark that carries according to the message that is received from plane conversion device 34, with the packet buffer that receives in corresponding input-buffer formation.
Scheduler module 1106, be used for by scheduling strategy between formation, ministry dispatching sends the message in ether output buffers formation 1104 and the route output buffers formation 1105 outside packet switching chip 31, the message in 311 scheduling transmission ether input-buffer formations 1107 of packet switch unit and route input-buffer formation 1108.The scheduling strategy of input and output can be identical or different.
Below packet switch unit 311 is described in detail.
Figure 12 is the structural representation of packet switch unit 311, and packet switch unit 311 comprises that ply-yarn drill inserts detection module 1201, mark maintenance module 1202 and exchange processing module 1203; Wherein,
Ply-yarn drill inserts detection module 1201, be used for when detecting ply-yarn drill access interface unit 312, return to the ether ply-yarn drill 33 that inserts transmit, the outbound port mark of ether plane mark and plane conversion device, return to the route line card 32 that inserts transmit, the outbound port mark of route plane mark and plane conversion device;
Below plane conversion device 34 is described in detail.
Figure 13 is the structural representation of Figure 10 midplane conversion equipment 34.As shown in figure 13, this plane conversion device comprises taxon 1300, receives buffer memory 1301, sends buffer memory 1302, converting unit 1303, distributed scheduler 1304 and transmitting element 1305, wherein,
Receive buffer memory 1301, be used for the message that buffer memory is received from packet switching chip 31.This reception buffer memory 1301 is divided into route and receives buffer queue and ether reception buffer queue.
Send buffer memory 1302, be used for the buffer memory message of giving packet switching chip 31 to be sent.This transmission buffer memory 1302 is divided into route and sends buffer queue and ether transmission buffer queue.
Converting unit 1303 is used for the message that receives buffer memory 1303 is carried out conversion between ether plane mark and the route plane mark, can also carry out the message format conversion if desired, with the packet buffer after the conversion to sending in the buffer memory 1302.Specifically, the packet buffer that carries route plane mark is sent buffer queue to route, the packet buffer that carries ether plane mark is sent buffer queue to ether.
Transmitting element 1305 is used for exporting the message that sends buffer memory 1302.Specifically, transmitting element is according to scheduling strategy between default formation, and the scheduling route sends buffer queue and ether transmission buffer queue sends message.
Distributed scheduler 1304 is used to safeguard the schedule information that is received from packet switching chip 31; When transmitting element 1305 sends routes and sends message in the buffer queues, when promptly the message of route plane mark is carried in scheduling, control its transmitted traffic according to schedule information.
Improvement to route line card 32 in the embodiment of the invention shown in Figure 10 and ether ply-yarn drill 31 is described respectively below.
Figure 14 is the structural representation of route line card 32 among Figure 10.As shown in figure 14, route line card 32 comprises packet labeling module 1401, sends buffer queue 1410 and distributed scheduler 1402, wherein,
Send buffer queue 1410, be used for the message to be exchanged that buffer memory sends to packet switching chip 31.
Distributed scheduler 1402 is used to safeguard the schedule information that is received from packet switching chip 31; Control sends the transmitted traffic of message in the buffer queue 1410 according to schedule information.
When the embodiment of the invention is applied in the route line card 22 of Fig. 2 to the improvement of route ply-yarn drill, as shown in figure 14, realize among the PP of packet labeling module 1401 in route line card in the embodiment of the invention, in practice, also can in FAU, realize.Sending buffer queue 1410 realizes in FAU.
Figure 15 is the structural representation of ether ply-yarn drill 33 among Figure 10.As shown in figure 15, this ether ply-yarn drill 33 comprises packet labeling module 1501 and transmission buffer queue 1502,
Send buffer queue 1402, be used for buffer memory and wait for the message to be exchanged that sends to packet switching chip 31.
When the embodiment of the invention is applied in the ether ply-yarn drill 12 of Fig. 1 to the improvement of ether ply-yarn drill, the packet labeling module 1501 in the embodiment of the invention and send buffer queue 1502 and all realize at the MAC layer.
In sum, more than be preferred embodiment of the present invention only, be not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (21)
1. switching network message switching method is characterized in that this method comprises:
But, in switch, has the exchange plane of separate planes buffer memory for the setting of every class ply-yarn drill according to the tie-in line Card Type of switch;
Described switch will be received from the source ply-yarn drill and the different message of purpose ply-yarn drill type of ply-yarn drill, and plane conversion device is given in exchange;
Described plane conversion device sends in the switch the corresponding exchange plane of purpose ply-yarn drill type with this message with the message that receives, and is exchanged to corresponding purpose ply-yarn drill by exchange plane.
2. the method for claim 1 is characterized in that, but described tie-in line Card Type according to switch, and the exchange plane that has a separate planes buffer memory for the setting of every class ply-yarn drill in switch is:
In switch, be each port arrangement output buffers independently; The employed output buffers of port that inserts the same type ply-yarn drill constitutes the plane buffer memory of the type ply-yarn drill place exchange plane.
3. method as claimed in claim 2 is characterized in that, but described tie-in line Card Type comprises ether ply-yarn drill and route line card, and the exchange plane of described route line card correspondence is not for allowing the exchange plane of packet loss;
Described for each port arrangement independently output buffers comprise:
In switch, for each inserts the independently ether output buffers formation of port arrangement of ether ply-yarn drill, for each inserts the independently route output buffers formation of port arrangement of route line card, for inserting the independently output buffers formation of port arrangement of plane conversion device.
4. method as claimed in claim 2 is characterized in that, but described tie-in line Card Type comprises ether ply-yarn drill and route line card;
Described for each port arrangement independently output buffers comprise:
In switch, be each port arrangement output buffers independently; The output buffers that inserts each port of ply-yarn drill is divided into formation of ether output buffers and the formation of route output buffers separate and that capacity is fixing; The port that inserts the ether ply-yarn drill uses the formation of ether output buffers; The port that inserts route line card uses the formation of route output buffers.
5. as claim 3 or 4 described methods, it is characterized in that this method further comprises:
Switch is when detecting the ply-yarn drill access interface, return to the ether ply-yarn drill that inserts transmit, the outbound port mark of ether plane mark and plane conversion device, return to the route line card that inserts transmit, the outbound port mark of route plane mark and plane conversion device;
Described switch will be received from the source ply-yarn drill and the different message of purpose ply-yarn drill type of ply-yarn drill, and exchange to plane conversion device is:
As the ether ply-yarn drill of source ply-yarn drill, the purpose ply-yarn drill that sends to switch be add ether plane mark in the message of route line card, according to transmitting the definite message outbound port mark and the outbound port mark of plane conversion device;
As the route line card of source ply-yarn drill, the purpose ply-yarn drill that sends to switch be add route plane mark in the message of ether ply-yarn drill, according to transmitting the definite message outbound port mark and the outbound port mark of plane conversion device;
Switch judges when the message be received from ply-yarn drill has the outbound port mark of plane conversion device, with the output buffers formation of the message switching that received to plane conversion device institute access interface, and output;
Described plane conversion device sends in the switch the corresponding exchange plane of purpose ply-yarn drill type with this message with the message that receives, and is exchanged to corresponding purpose ply-yarn drill by exchange plane to comprise:
Plane conversion device is carried out conversion between ether plane mark and the route plane mark to the message that receives AMSU auto-manual switching unit, and the message after the conversion is sent to switch; When sending, according to the schedule information on route switching plane, the message flow that carries route plane mark that control sends to switch;
Switch is according to the message outbound port mark in the message that is received from plane conversion device, to the route output buffers formation of corresponding outbound port, the message switching that will carry ether plane mark is to the ether output buffers formation of corresponding outbound port with the message switching of carrying route plane mark.
6. method as claimed in claim 5 is characterized in that, the output buffers formation of described plane conversion device institute access interface further is divided into formation of ether output buffers and the formation of route output buffers;
Described with the output buffers formation of the message switching that received to plane conversion device institute access interface, and be output as:
Switch will carry the route output buffers formation of the message switching of route plane mark to plane conversion device institute access interface, will carry the ether output buffers formation of the message switching of ether plane mark to plane conversion device institute access interface; According to scheduling strategy between default formation, the route output buffers formation of scheduling output plane conversion equipment institute access interface and the message in the formation of ether output buffers.
7. method as claimed in claim 5 is characterized in that, this method further comprises: in advance the transmission buffer memory in the plane conversion device is divided into separate ether and sends buffer queue and route transmission buffer queue;
Message after described will the conversion sends to switch; When sending, according to the schedule information on route switching plane, control comprises to the message flow that carries route plane mark that switch sends:
The packet buffer that will carry route plane mark after plane conversion device will be changed sends buffer queue to route; The packet buffer that carries ether plane mark after the conversion is sent buffer queue to ether;
By scheduling strategy between default formation, scheduling transmission route transmission buffer queue and ether send the message in the buffer queue;
When dispatching the message that sends in the route transmission buffer queue, according to the schedule information that is received from route switching plane in the switch, the control route sends the message transmitted traffic of buffer queue.
8. method as claimed in claim 7, it is characterized in that, describedly transmission buffer memory in the plane conversion device is divided into separate ether sends the operation that buffer queue and route send buffer queue and further comprise:, respectively described ether is sent buffer queue and route and send buffer queue and be divided into subqueue corresponding to each service priority according to service priority;
The packet buffer that will carry route plane mark after described will the conversion sends buffer queue to route; The packet buffer that carries ether plane mark after the conversion is sent buffer queue to ether is: the service priority sign that plane conversion device is carried according to message receives the packet buffer that carries route plane mark in the subqueue corresponding with the service priority sign in the buffer queue to route; The packet buffer that carries ether plane mark is received in the subqueue corresponding with the service priority sign in the buffer queue to ether.
9. method as claimed in claim 7 is characterized in that, and is described by scheduling strategy between default formation, and the message that scheduling transmission route transmission buffer queue and ether send in the buffer queue is:
Pre-configured scheduling ether sends the initial schedule ratio of buffer queue and route transmission buffer queue in plane conversion device;
Plane conversion device is adjusted described initial schedule ratio according to the detected actual flow that is sent to ether exchange plane and route switching plane in the switch.
10. the method for claim 1 is characterized in that, the exchange plane that is provided with in switch comprises the exchange plane that does not allow packet loss;
Described transmit operation further comprises: according to the schedule information of the exchange plane that does not allow packet loss, control sends to the flow of the exchange plane that does not allow packet loss.
11. the method for claim 1 is characterized in that, described setting has after the exchange plane of separate planes buffer memory, and this method further comprises:
Described switch will be received from the source ply-yarn drill and the identical message of purpose ply-yarn drill type of ply-yarn drill, exchange to the purpose ply-yarn drill by the exchange plane corresponding with purpose ply-yarn drill type.
12. a switching system is characterized in that, this system comprises switch and the plane conversion device that is plugged on the switch; Described switch comprises at least 2 exchange planes that have the separate planes buffer memory respectively, and each exchange plane is corresponding with a class ply-yarn drill that can be plugged on self;
Described switch is used for and will be received from the source ply-yarn drill and the different message of purpose ply-yarn drill type of ply-yarn drill, and described plane conversion device is given in exchange; To be received from the message of described plane conversion device, exchange to the purpose ply-yarn drill by the exchange plane corresponding with the purpose ply-yarn drill type of this message;
Described plane conversion device, the message that is used for being received from described switch send to the corresponding exchange plane of purpose ply-yarn drill type of described switch and this message.
13. switching system as claimed in claim 12 is characterized in that, described switch comprises packet switch unit and a plurality of port unit;
Described port unit comprises independently output buffers formation; The employed output buffers formation of port unit that inserts the same type ply-yarn drill constitutes the plane buffer memory of the type ply-yarn drill place exchange plane;
Described packet switch unit is used for and will be received from the source ply-yarn drill and the different message of purpose ply-yarn drill type of ply-yarn drill, and the output buffers formation in the described plane conversion device institute access interface unit is given in exchange; To be received from the message of described plane conversion device, exchange to the output buffers formation in this message purpose ply-yarn drill institute access interface unit.
14. switching system as claimed in claim 13 is characterized in that, the ply-yarn drill that can be plugged on the switch is route line card and ether ply-yarn drill; Output buffers formation in the port unit of access ether ply-yarn drill is the formation of ether output buffers; Output buffers formation in the port unit of access route line card is the formation of route output buffers.
15. switching system as claimed in claim 14 is characterized in that, each port unit that inserts ply-yarn drill comprises formation of ether output buffers and the formation of route output buffers separate and that capacity is fixing;
The port unit of described access ether ply-yarn drill uses the formation of ether output buffers;
The port unit of described access route line card uses the formation of route output buffers.
16., it is characterized in that the clamping of described packet switch unit pack vinculum is gone into detection module, mark maintenance module and exchange processing module as claim 14 or 15 described switching systems;
Described ply-yarn drill inserts detection module, be used for when detecting the ply-yarn drill access interface, return to the ether ply-yarn drill that inserts transmit, the outbound port mark of ether plane mark and plane conversion device, return to the route line card that inserts transmit, the outbound port mark of route plane mark and plane conversion device;
Described mark maintenance module is used to preserve default described ether plane mark and route plane mark, and offers described ply-yarn drill access detection module and exchange processing module;
Described exchange processing module is used for receiving message from port unit, with carry plane conversion device the outbound port mark wait to exchange the output buffers of message switching to the described plane conversion device institute access interface unit.
17. switching system as claimed in claim 16, it is characterized in that the output buffers formation that inserts the port unit of plane conversion device further is divided into and is used for buffer memory and waits for and issue the route output buffers formation of carrying route plane mark of plane conversion device and be used for the ether output buffers formation of carrying ether plane mark that plane conversion device is issued in the buffer memory wait.
18. switching system as claimed in claim 16 is characterized in that, described plane conversion device comprises the reception buffer memory, sends buffer memory, converting unit, transmitting element and distributed scheduler;
Described reception buffer memory is used for the message that buffer memory is received from switch;
Described transmission buffer memory is used for buffer memory and waits for the message that sends to switch;
Described converting unit is used for the message that receives buffer memory is carried out conversion between ether plane mark and the route plane mark, with the packet buffer after the conversion in described transmission buffer memory;
Described transmitting element is used for exporting the message of described transmission buffer memory;
Described distributed scheduler is used for safeguarding the schedule information that is received from switch route switching plane; According to the schedule information of safeguarding, control the transmitted traffic that described transmitting element sends the message that carries route plane mark.
19. switching system as claimed in claim 18 is characterized in that, described transmission buffer memory comprises that independently route sends buffer queue and ether transmission buffer queue;
Described converting unit is further used for, and the packet buffer that carries ether plane mark after the conversion is sent buffer queue to ether, and the packet buffer that carries route plane mark after the conversion is sent buffer queue to route;
Described transmitting element is further used for according to scheduling strategy between default formation, and described route transmission buffer queue of scheduling transmission and ether send the message in the buffer queue.
20. switching system as claimed in claim 12 is characterized in that, described switch is further used for, and with being received from the source ply-yarn drill and the identical message of purpose ply-yarn drill type of ply-yarn drill, exchanges to the purpose ply-yarn drill by the exchange plane corresponding with purpose ply-yarn drill type.
21. switching system as claimed in claim 12 is characterized in that, described plane conversion device is further used for, and when sending, according to the schedule information of the exchange plane that does not allow packet loss, control sends to the flow of the exchange plane that does not allow packet loss.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101162258A CN101304383B (en) | 2008-07-07 | 2008-07-07 | Method for switching message of switching network and switching system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101162258A CN101304383B (en) | 2008-07-07 | 2008-07-07 | Method for switching message of switching network and switching system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101304383A CN101304383A (en) | 2008-11-12 |
| CN101304383B true CN101304383B (en) | 2010-10-27 |
Family
ID=40114096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101162258A Active CN101304383B (en) | 2008-07-07 | 2008-07-07 | Method for switching message of switching network and switching system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101304383B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101895398B (en) | 2010-07-15 | 2012-07-25 | 华为技术有限公司 | Method and device for data communication |
| CN102006226B (en) * | 2010-11-19 | 2013-01-02 | 福建星网锐捷网络有限公司 | Message cache management method and device as well as network equipment |
| CN102065009A (en) * | 2010-12-16 | 2011-05-18 | 中兴通讯股份有限公司 | Method for allocating distributed queue and distributed system |
| CN102739536B (en) * | 2012-06-26 | 2016-03-02 | 华为技术有限公司 | A kind of message caching method and router |
| CN102769538B (en) * | 2012-07-02 | 2015-06-17 | 福建星网锐捷网络有限公司 | Port changing method, related device and system |
| CN103178996B (en) * | 2013-03-15 | 2015-07-01 | 烽火通信科技股份有限公司 | Distributed packet-switching chip model verification system and method |
| CN105812290B (en) * | 2016-03-10 | 2018-12-07 | 中国电子科技集团公司第五十四研究所 | A kind of method for interchanging data based on more exchange planes |
| CN107689923B (en) | 2016-08-04 | 2021-02-12 | 华为技术有限公司 | Message processing method and router |
| CN111343087B (en) * | 2020-02-26 | 2022-02-01 | 深圳市风云实业有限公司 | Reusable multipurpose distributed router system and implementation method |
| CN115567435A (en) * | 2021-07-02 | 2023-01-03 | 中国船舶重工集团公司第七二四研究所 | FPGA-based routing information parallel and rapid searching and managing method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1612562A (en) * | 2003-10-28 | 2005-05-04 | 武汉烽火网络有限责任公司 | Method and device for realizing different IP data packet repeating using strategic flow |
| CN101150527A (en) * | 2007-11-09 | 2008-03-26 | 杭州华三通信技术有限公司 | A PCIE data transmission method, system and device |
-
2008
- 2008-07-07 CN CN2008101162258A patent/CN101304383B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1612562A (en) * | 2003-10-28 | 2005-05-04 | 武汉烽火网络有限责任公司 | Method and device for realizing different IP data packet repeating using strategic flow |
| CN101150527A (en) * | 2007-11-09 | 2008-03-26 | 杭州华三通信技术有限公司 | A PCIE data transmission method, system and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101304383A (en) | 2008-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101304383B (en) | Method for switching message of switching network and switching system | |
| CN101299721B (en) | Method for switching message of switching network, and switching device | |
| CN100349443C (en) | Method and apparatus for scheduling message processing | |
| US9967638B2 (en) | Optical switching | |
| CN1745549B (en) | System for content based message processing | |
| US7042883B2 (en) | Pipeline scheduler with fairness and minimum bandwidth guarantee | |
| CN104104616B (en) | The method, apparatus and system of data dispatch and exchange | |
| JP3618390B2 (en) | Apparatus and method for processing bandwidth requests in ATM switches | |
| US6934250B1 (en) | Method and apparatus for an output packet organizer | |
| JP2981082B2 (en) | Packet cell scheduling device | |
| US7855960B2 (en) | Traffic shaping method and device | |
| EP0717532A1 (en) | Dynamic fair queuing to support best effort traffic in an ATM network | |
| CN101478483A (en) | Method for implementing packet scheduling in switch equipment and switch equipment | |
| JPH10178451A (en) | Hybrid exchange, exchange, and method for re-assignment of stm data in the exchanges. | |
| US8086770B2 (en) | Communication apparatus with data discard functions and control method therefor | |
| CN1359241A (en) | Distribution type dispatcher for group exchanger and passive optical network | |
| CN101212467A (en) | MPLS network service scheduling method | |
| CN1980189A (en) | Communication business buffer isolation and control | |
| Bouabdallah et al. | Resolving the fairness issues in bus-based optical access networks | |
| Zeghid et al. | Modified optical burst switching (OBS) based edge node architecture using real-time scheduling techniques | |
| CN101212412A (en) | MPLS network service scheduling system | |
| RU2294601C1 (en) | Method for performing statistical multiplexing during transfer of information | |
| CN101729373B (en) | Method and device for dispatching multipriority traffic | |
| CN101771603A (en) | Data exchange method and data exchange equipment | |
| CN101542985A (en) | A packet switching system and a method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No. Patentee after: Xinhua three Technology Co., Ltd. Address before: 310053 Hangzhou hi tech Industrial Development Zone, Zhejiang province science and Technology Industrial Park, No. 310 and No. six road, HUAWEI, Hangzhou production base Patentee before: Huasan Communication Technology Co., Ltd. |