CN110324265A - Traffic distribution method, method for routing, equipment and network system - Google Patents
Traffic distribution method, method for routing, equipment and network system Download PDFInfo
- Publication number
- CN110324265A CN110324265A CN201810271926.2A CN201810271926A CN110324265A CN 110324265 A CN110324265 A CN 110324265A CN 201810271926 A CN201810271926 A CN 201810271926A CN 110324265 A CN110324265 A CN 110324265A
- Authority
- CN
- China
- Prior art keywords
- network
- port
- exchange chip
- switching equipment
- exchange
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/10—Packet switching elements characterised by the switching fabric construction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/10—Packet switching elements characterised by the switching fabric construction
- H04L49/102—Packet switching elements characterised by the switching fabric construction using shared medium, e.g. bus or ring
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/25—Routing or path finding in a switch fabric
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/25—Routing or path finding in a switch fabric
- H04L49/253—Routing or path finding in a switch fabric using establishment or release of connections between ports
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/30—Peripheral units, e.g. input or output ports
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the present application provides a kind of traffic distribution method, method for routing, equipment and network system.In the embodiment of the present application, the exchange chip for being under the jurisdiction of heterogeneous networks is merged into a network switching equipment, node device is accessed to different networks by this network switching equipment, to construct a kind of more rail network systems, in this more rail network systems, the exchange chip for being under the jurisdiction of heterogeneous networks in the network switching equipment can share some modules, be conducive to save space resources, reduce the volume of the network switching equipment, and the network switching equipment of same level is shared between heterogeneous networks, the quantity of the network switching equipment is relatively fewer, the complexity of network system is relatively low, scale is relatively small.
Description
Technical field
This application involves network communication technology field more particularly to a kind of traffic distribution method, method for routing, equipment and nets
Network system.
Background technique
With the development of network communication technology, the network bandwidth of data center is promoted rapidly, the Netowrk tape of 25Gb/s
Width has been popularized, and also mature or even 400Gb/s network bandwidth has fermented the network bandwidth of 100Gb/s.But net
The development of network bandwidth by transceiver on physical link, such as SERDES manufacturing process limitation, the development of network bandwidth is remote
It is unable to catch up with the promotion of node computing capability, the bottleneck of network bandwidth shows more prominent in the data center.
In order to solve the network bandwidth concerns that data center faces, the prior art is by disposing multiple physically separate nets
Network allows node device to access multiple networks simultaneously, provides network service simultaneously by multiple networks for node device, this is certain
Solve the problems, such as that access bandwidth is limited in degree.
But existing multi-network framework is there is also some shortcomings, such as complexity is high, in large scale etc., needs thus newly
Technology realize the relatively low multi-network framework of a kind of complexity, scale.
Summary of the invention
The many aspects of the application provide a kind of traffic distribution method, method for routing, equipment and network system, to solve
The problems such as complexity existing for existing multi-network framework is high, in large scale.
The embodiment of the present application provides a kind of network switching equipment, comprising: N number of exchange chip and can be by N number of exchange
At least one shared sharing module of chip;N number of exchange chip is under the jurisdiction of N number of different network, and N number of exchange
Chip forms the M2 network port, for connecting M2 platform node device, accesses corresponding network for the M2 platform node device;Its
In, N and M2 are natural numbers, and N >=2.
The embodiment of the present application also provides a kind of network system, comprising: an at least node device and at least a network friendship
Exchange device;
Every network switching equipment include N number of exchange chip and can by N number of exchange chip share at least one
Sharing module;N number of exchange chip is under the jurisdiction of N number of different network, and N number of exchange chip forms M2 network-side
Mouthful, for connecting the M2 platform node device in an at least node device, accessed for the M2 platform node device corresponding
Network;Wherein, N and M2 is natural number, and N >=2;
Every node device includes the K network port, is used for and a network in an at least network switching equipment
K exchange chip in switching equipment is correspondingly connected with, and K is natural number, and K≤N.
The embodiment of the present application also provides a kind of traffic distribution method, and the node for being suitable for inclusion at least two network ports is set
It is standby, this method comprises:
Obtain current data packet;
For the purpose of flow equalization, target port selection ginseng is determined from the corresponding port selection parameter of traffic distribution strategy
Number;
Based on the target port selection parameter, the selection target network port from least two network port;
The current data packet is sent to the net that the target network port is accessed by the target network port
In network, wherein at least two network port is respectively connected to different networks.
The embodiment of the present application also provides a kind of method for routing, should suitable for the network switching equipment that previous embodiment provides
Method includes:
Based on routing policy, determine that data packet to be sent is corresponding from N number of exchange chip of the network switching equipment
First exchange chip;
When first exchange chip meets the network switching condition of setting, the is selected from other reachable exchange chips
Two exchange chips, it is right in addition to first exchange chip in N number of exchange chip that other reachable exchange chips refer to
The destination of the data packet to be sent routes reachable exchange chip;
Based on the internal network between first exchange chip and second exchange chip, second exchange is controlled
Chip sends the data packet to be sent.
The embodiment of the present application also provides a kind of node device, comprising: the K network port, memory and processor;
The K network port, for accessing the different networks of K, K is natural number, and K >=2;
The memory, for storing computer program;
The processor is coupled with the memory, for executing the computer program, to be used for:
Obtain current data packet;
For the purpose of flow equalization, target port selection ginseng is determined from the corresponding port selection parameter of traffic distribution strategy
Number;
Based on the target port selection parameter, the selection target network port from at least two K network ports;
The current data packet is sent to the net that the target network port is accessed by the target network port
In network.
In the embodiment of the present application, the exchange chip for being under the jurisdiction of heterogeneous networks is merged into a network switching equipment,
Node device is accessed to different networks by this network switching equipment, to construct a kind of more rails comprising heterogeneous networks
Network system.In this more rail network systems, the exchange chip that heterogeneous networks are under the jurisdiction of in the network switching equipment can share one
A little modules are conducive to save space resources, reduce the volume of the network switching equipment, and share same net between heterogeneous networks
The quantity of network switching equipment, the network switching equipment is relatively fewer, and therefore, the complexity of network system is relatively low, and scale is opposite
It is smaller.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 a is a kind of structural schematic diagram for network switching equipment that one exemplary embodiment of the application provides;
Fig. 1 b is the structural schematic diagram for another network switching equipment that the application another exemplary embodiment provides;
Fig. 1 c is the network for being suitable for the application exemplary embodiment and providing that the application another exemplary embodiment provides
A kind of structural schematic diagram of exchange chip of switching equipment;
Fig. 1 d is the structural schematic diagram for another network switching equipment that the application another exemplary embodiment provides;
Fig. 1 e is the structural schematic diagram for another network switching equipment that the application another exemplary embodiment provides;
Fig. 2 is a kind of structural schematic diagram for network system that the application another exemplary embodiment provides;
Fig. 3 a is a kind of flow diagram for traffic distribution method that the application another exemplary embodiment provides;
Fig. 3 b is a kind of flow diagram for traffic distribution method packet-by-packet that the application another exemplary embodiment provides;
Fig. 3 c is a kind of flow diagram of the traffic distribution method for valley-fill that the application another exemplary embodiment provides;
Fig. 3 d is the process signal for another traffic distribution method packet-by-packet that the application another exemplary embodiment provides
Figure;
Fig. 3 e is the process signal of the traffic distribution method for another valley-fill that the application another exemplary embodiment provides
Figure;
Fig. 4 a is a kind of flow diagram for method for routing that the application another exemplary embodiment provides;
Fig. 4 b is the flow diagram for another method for routing that the application another exemplary embodiment provides;
Fig. 5 is a kind of structural schematic diagram for node device that the application another exemplary embodiment provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one
Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall in the protection scope of this application.
The deficiencies of complexity existing for the existing multi-network framework is high, in large scale, in some exemplary realities of the application
It applies in example, the exchange chip for being under the jurisdiction of heterogeneous networks is merged into a network switching equipment, is set by this network exchange
It is standby that node device is accessed to different networks, to construct a kind of more rail network systems comprising heterogeneous networks.This more
The exchange chip for being under the jurisdiction of heterogeneous networks in rail network system, in the network switching equipment can share some modules, be conducive to save
Space resources reduces the volume of the network switching equipment, and the same network switching equipment, network are shared between heterogeneous networks
Switching equipment quantity is relatively fewer, and therefore, the complexity of network system is relatively low, and scale is relatively small.
Below in conjunction with attached drawing, the technical scheme provided by various embodiments of the present application will be described in detail.
Fig. 1 a is a kind of structural schematic diagram for network switching equipment that one exemplary embodiment of the application provides.Such as Fig. 1 a institute
Show, which includes: N number of exchange chip 101 and can be shared by least one that N number of exchange chip is shared
Module 102.Wherein, N is natural number, and N >=2, the i.e. network switching equipment 100 include at least two exchange chips 101.In network
In switching equipment 100, N number of exchange chip 101 is under the jurisdiction of N number of different network S1-SN。
In the present embodiment, network S1-SNIt is to be understood that being will be distributed over differently with communication line and communication equipment
More autonomous computer systems of point are interconnected, and according to common network protocol, share the resources such as hardware, software, most
The system of resource-sharing is realized eventually.According to the difference of interconnection range, network S1-SNIn any network can be local area network, metropolitan area
Net or wide area network.
In some optional embodiments, network S1-SNIn network can be homogeneous network, i.e., the equipment in network by
The either compatible equipment of same supplier's supply, they operate in the same operating system either network operating system
Under.In other optional embodiments, network S1-SNIn network be also possible to heterogeneous network, i.e., the equipment in network is not by
With the equipment of supplier's supply, operates in support different function or applications in different agreements in most cases.
In the present embodiment, the network switching equipment 100 for designing a kind of new construction wraps in the network switching equipment 100
The N number of exchange chip 101 for being under the jurisdiction of N number of heterogeneous networks is included, this, which is equivalent to, has been merged into a net for this N number of exchange chip 101
In network switching equipment 100.In addition, the network switching equipment 100 further includes at least one mould that can be shared by N number of exchange chip 101
Block, referred to as sharing module.These sharing modules are to refer to provide same services for N number of exchange chip 101 or can satisfy N
The module, such as clock module, power module etc. of a 101 same requirements of exchange chip.
For the network switching equipment 100 comprising N number of exchange chip 101 can form the M2 network port, for connecting
M2 platform node device 200 is connect, accesses corresponding network for M2 platform node device 200.Wherein, M2 is natural number.Preferably, M2 is
>=2 natural number.Wherein, for a node device 200, can connect with the part exchange chip in N number of exchange chip 101
It connects, to access network belonging to the part exchange chip that it is connected, alternatively, can also connect respectively with N number of exchange chip 101
It connects, to access N number of different network.
In the present embodiment, not limit node device 200 realization form, node device 200 can be it is any can
The equipment being connected on exchange chip 101, such as can be work station, terminal device, the network user or personal computer, may be used also
Equipment to be server, printer or other network connections.
The network switching equipment of the present embodiment includes the exchange chip for being under the jurisdiction of heterogeneous networks, node device can be accessed
Different network, to construct a kind of network system comprising heterogeneous networks.For ease of description, by the net based on the present embodiment
The network system that network switching equipment constructs is known as more rail network systems.Exchange chip due to being under the jurisdiction of heterogeneous networks is merged
Into a network switching equipment, so sharing same network exchange between heterogeneous networks in this more rail network systems
The quantity of equipment, the network switching equipment is relatively fewer, in addition, the exchange chip for being under the jurisdiction of heterogeneous networks in the network switching equipment can
Some modules are shared, is conducive to save space resources, reduces the volume of the network switching equipment, therefore this more rail network systems
Complexity is relatively low, and scale is also relatively small.
In some exemplary embodiments, to network S1-SNIn any network for, may include multiple exchange chips,
And concatenation technology can be used and cascade multiple exchange chips, with extended network scale, these cascade exchange chips it
Between formed cascade level.In the present embodiment, the cascade system between exchange chip is not limited, is not limited to cascade level yet
Definition.For example, the cascade level of three exchange chips can be defined as first by taking successively cascade three exchange chips as an example
Level, the second level and third level, or level A, level B and level C can also be defined as, etc..
Based on the cascade level between above-mentioned exchange chip, N number of exchange chip 101 in the network switching equipment 100 is subordinate to
In N number of different network S1-SN, and cascade level of the N number of exchange chip 101 in respective belonging network is identical.
In other exemplary embodiments of the application, as shown in Figure 1 b, at least one of network switching equipment 100 is altogether
Enjoying module may include: M2 externally port modules 103.M2 external port modules 103 in the network switching equipment 100 are main
For realizing the connection between the 100 N number of exchange chip 101 in inside of M2 platform node device 200 and the network switching equipment.
In application scenes, every node device 200 can be separately connected with N number of exchange chip 101, in order to connect
Enter N number of different network.Then, as shown in Figure 1 b, each externally one end of port module 103 connects with a node device 200
It connects, the other end is connect with N number of exchange chip 101 respectively.For any external port module 103, on the one hand can will from
The signal of its node device 200 connected is distributed to N number of exchange chip 101 in its belonging network switching equipment 100, another party
Face can will be sent to node device 200 connected to it after the signal gathering from N number of exchange chip 101.In these applications
In scene, the quantity for the external port module 103 that the network switching equipment 100 includes determines that the network exchange is set to a certain extent
The quantity of standby 100 node devices 200 that can connect.
In the embodiment of the present application, the connection side between external port module 103 and N number of exchange chip 101 is not limited
Formula.In some exemplary embodiments, as illustrated in figure 1 c, each exchange chip 101 in the network switching equipment 100 includes M2
Outside port 11.M2 outside port 11 in each exchange chip 101 is external with M2 in the network switching equipment 100 respectively
Port module 103 connects one to one.Using this exchange chip structure, external port module can be simply, easily realized
Connection between 103 and N number of exchange chip 101.
In the other exemplary embodiment of the application, as shown in Figure 1 d, at least one of network switching equipment 100 is altogether
Enjoying module can also include: Intranet control module 104.Intranet control module 104 is exchanged with N number of in the network switching equipment 100
Chip 101 connects, and carries out data exchange between N number of exchange chip 101 for controlling.Wherein, phase between N number of exchange chip 101
It connects, forms the internal network of the network switching equipment 100.Optionally, as illustrated in figure 1 c, every in the network switching equipment 100
A exchange chip 101 further include: M1 internal port 12.N number of exchange chip 101 in the network switching equipment 100 passes through respective
The M1 internal port 12 for including interconnects, and forms the internal network of the network switching equipment 100;Wherein, M1 is natural number, and M1≤
N.The embodiment of the present application does not pass through opening up for the internal network that the M1 internal port respectively contained is constituted to N number of exchange chip 101
Flutter and be defined, such as can be hub-and-spoke configuration, ring type structure, bus topology, tree, reticular structure etc..
In these exemplary embodiments, the inside for forming the network switching equipment 100 is interconnected between N number of exchange chip 101
Network, based on the internal network between N number of exchange chip 101, Intranet control module 104 can control network according to application demand
Data mutually switch between the different networks, and especially when some network occurs congestion or breaks down, Intranet controls mould
Data in the network can be neatly switched in the network of other underloadings by block 104, be realized between same position exchange chip
Interconnect, flow is arbitrarily switched in heterogeneous networks, the dynamic load leveling of more rail network systems is realized
And high availability, the multipath in more rail network systems is excavated, sufficiently also convenient for giving full play to the advantage of more rail network systems.
It is worth noting that Intranet control module 104 can be used as the control plane inside the network switching equipment 100,
Function can also be responsible for matching for the network switching equipment 100 other than control network data mutually switches between the different networks
Set management, such as the configuration of routing table etc..Intranet control module 104 controls what network data mutually switched between the different networks
Process and Intranet control module 104 carry out the process of adaptive routing between N number of exchange chip.Optionally, Intranet controls
Module 104, which can use but be not limited to following routing modes, carries out data exchange between N number of exchange chip 101 to control:
In this approach, Intranet control module 104 can be based on routing policy, and number to be sent is determined from N number of exchange chip
According to corresponding first exchange chip of packet;When the first exchange chip meets the network switching condition of setting, from other up to exchange
The second exchange chip is selected in chip, it is right in addition to the first exchange chip in N number of exchange chip that other reachable exchange chips refer to
The destination of data packet to be sent routes reachable exchange chip;In between the first exchange chip and the second exchange chip
Portion's network, the second exchange chip of control send data packet to be sent.
Wherein, routing policy can be traditional routing algorithms, such as can be distributed route selection method, centralized road
By selecting, hybrid dynamics route selection, link-state routing algorithm etc..Traditional routing is based on about Intranet control module 104
Algorithm determines the detailed process of the first exchange chip similarly to the prior art, and those skilled in the art are based on to traditional routing algorithms
Understanding can easily understand the determination process of the first exchange chip, it is not described here in detail.Wherein, the first exchange chip
It can be any exchange chip in N number of exchange chip.
Wherein, network switching condition can be depending on concrete application scene.For example, in application scenes, Intranet control
Module 104 is after determining the first exchange chip, it can be determined that for sending the outer of data packet to be sent on the first exchange chip
Portion port whether there is failure, and whether packet loss is more than packet loss threshold value and/or whether average queuing delay is more than delay threshold
Value, and there are failures for the outside port on the first exchange chip for sending data packet to be sent, packet loss is more than packet loss
When threshold value and/or average queuing delay are more than delay threshold value, determine that the first exchange chip meets network switching condition, to touch
The operation of hair the second exchange chip of selection.
In some exemplary embodiments, the M2 outside port 11 in each exchange chip 101 be in that same order
It is numbered, such as is from left to right followed successively by port D_1, D_2, D_3 ... ... D_M2, and can will have in each exchange chip 101
There is the outside port of identical port numbers to be connected to same external port module 103.Based on this, Intranet control module 104 is being selected
When the second exchange chip, it can be obtained based on the port numbers of the outside port on the first exchange chip for sending data packet to be sent
Take the flow load of the outside port on other reachable exchange chips with identical port numbers;According on other reachable exchange chips
The flow load of outside port with identical port numbers selects flow load to meet setting from other reachable exchange chips and wants
The exchange chip asked is as the second exchange chip.When can encounter failure/heavy congestion in a certain network in this way, by the network
In data packet be switched to another underloading network in transmit, can accomplish the load balancing between heterogeneous networks, and can drop
The transmission delay of low data packet guarantees the accessibility of data packet.
Above-mentioned sets requirement can be arranged according to application demand adaptability, such as can be that flow load is minimum, flow load
Be located within the scope of certain flow less than setting flow load threshold value or flow load etc..Based on this, above-mentioned second exchange chip can
To be the exchange chip for having the flow load of the outside port of identical port numbers minimum in other reachable exchange chips, alternatively,
The flow load for being also possible to the outside port in other reachable exchange chips with identical port numbers is less than setting flow load
The exchange chip of threshold value, alternatively, being also possible to the flow of the outside port in other reachable exchange chips with identical port numbers
Load is located at a certain range of exchange chip, etc..
In the other exemplary embodiment of the application, as shown in fig. le, at least one of network switching equipment 100 is altogether
Enjoying module can also include but is not limited to: clock module 105, power module 106 and/or radiating module 107.
Wherein, clock module 105 is connect with N number of exchange chip 101, for providing clock letter to N number of exchange chip 101
Number.In consolidated network switching equipment 100, clock module 105 is shared between N number of exchange chip 101, is conducive to accomplish different nets
It is synchronized to greatest extent between network.
Wherein, power module 106 is connect with N number of exchange chip 101, for powering to N number of exchange chip 101.Radiating mould
Block 107 is connect with N number of exchange chip 101, for radiating for N number of exchange chip 101.In consolidated network switching equipment 100, N
Power module 106 and radiating module 107 are shared between a exchange chip 101, can not only save the inside of the network switching equipment
Space reduces the cost of implementation of the network switching equipment, but also can greatly improve efficiency.
The network switching equipment provided based on the above embodiment can construct a kind of more rail network systems.As shown in Fig. 2,
More rail network systems 10 include: an at least node device 200 and an at least network switching equipment 100.The present embodiment is simultaneously
The quantity of node device 200 and the network switching equipment 100 is not limited, it can be depending on application demand and network size.
As shown in Figure 1a, every network switching equipment 100 includes N number of exchange chip.Wherein, N is natural number, and N >=2,
I.e. the network switching equipment 100 includes at least two exchange chips.In the network switching equipment 100, N number of exchange chip is under the jurisdiction of N
A different network S1-SN。
Optionally, to network S1-SNIn any network for, may include multiple exchange chips, and cascade skill can be used
Art cascades multiple exchange chips, and with extended network scale, cascade level is formed between these cascade exchange chips.
Based on this, N number of exchange chip in each network switching equipment 100 is under the jurisdiction of N number of different network S1-SN, and N number of exchange core
Cascade level of the piece in respective belonging network is identical.In addition, due in consolidated network switching equipment 100 N number of exchange chip exist
Cascade level in respective belonging network is identical, can also be by exchange chip 101 N number of in the network switching equipment 100 in respective institute
Belong to the cascade level that the cascade level in network is considered as the network switching equipment 100.
In addition, as shown in Figure 1a, every network switching equipment 100 further includes that at least one can be shared by N number of exchange chip
Module, referred to as sharing module.These sharing modules are to refer to provide same services for N number of exchange chip or can satisfy
The module, such as clock module, power module etc. of N number of exchange chip same requirements.It is worth noting that for convenient for clearer
Ground illustrates the framework of more rail network systems 10, only shows sharing module in subnetwork switching equipment 100 in Fig. 2,
The network switching equipment 100 of sharing module is not shown equally comprising sharing module in it.
N number of exchange chip in every network switching equipment 100 can form at least one network port, for connecting at least
Part of nodes equipment in one node device 200, so that these node devices access corresponding network.For ease of description, will
N number of exchange chip is formed by the quantity of the network port, that is, every network switching equipment in every network switching equipment 100
The quantity of 100 node devices connected is denoted as M2, and for heterogeneous networks switching equipment 100, M2 can be identical, can also be with
It is not identical.That is, the quantity for the node device 200 that heterogeneous networks switching equipment 100 is connected can be identical, it can not also be identical.
Wherein, M2 is natural number, and M2 is less than or equal to the total quantity of at least one node device 200.
Wherein, every node device 200 includes the K network port.Every node device 200 is used for and an at least network
A network switching equipment connection in switching equipment 100.One node device 200 connect master with a network switching equipment 100
Refer to that K exchange chip and K exchange chip in the network switching equipment 100 in the node device 200 are correspondingly connected with,
K is natural number, and 1≤K≤N.Wherein, a network port is correspondingly connected with an exchange chip, the network port with exchange core
Corresponding relationship between piece without limitation, can be according to flexible setting.For example, can be compiled respectively to the network port and exchange chip
Number, the identical network port will be numbered and be correspondingly connected with exchange chip, but not limited to this.
In the present embodiment, not limit node device 200 realization form, node device 200 can be it is any can
Be connected on exchange chip and the equipment with the K network port, for example, can be work station, terminal device, the network user or
Personal computer, the equipment that can also be server, printer or other network connections.
In some optional embodiments, K >=2, i.e. node device 200 have at least two network ports.Realizing shape
In state, node device 200 may include single multi-port network card, which provides multiple network ports, alternatively, node
Equipment 200 is also possible to comprising muti-piece network interface card (every piece of network interface card can have a port, can also there is multiple ports), muti-piece network interface card
Multiple network ports are provided jointly.
In an alternative embodiment, K=N, i.e. every node device 200 include N number of network port, respectively with a network
N number of exchange chip in switching equipment connects one to one.
In addition, can be carried out according to the cascading layers grade of the network switching equipment between an at least network switching equipment 100 mutual
Connection.Optionally, cascade level of N number of exchange chip 101 in respective belonging network is identical in the network switching equipment 100, and
When being considered as the cascade level of the network switching equipment 100, according to inner exchanging chip between an at least network switching equipment 100
Cascade level in belonging network is connected with each other.It is worth noting that in more rail network systems 10, for being in same level-one
The network switching equipment 100 for joining level, can dispose one, can also dispose more.
Wherein, the chain interconnected between the network switching equipment 100 and between node device 200 and the network switching equipment 100
Road is mainly used for the transmission of several parallel signals of bearing link both ends transmission, can be wire link, is also possible to nothing
Wired link.Wherein, the medium of wire link can be multi-core optical fiber, multicore high speed cable or the optical fiber for carrying WDM signal.Nothing
Wired link refers to the path spacing for propagating electromagnetic wave.Between the network switching equipment 100 and node device 200 is set with network exchange
More rail network systems can be formed after being connected with each other between standby 100.
In the present embodiment, the network switching equipment includes the exchange chip for being under the jurisdiction of heterogeneous networks, can be by node device
Different networks is accessed, to construct a kind of network system comprising heterogeneous networks.For ease of description, this implementation can will be based on
The network system that the network switching equipment of example constructs is known as more rail network systems.Due to being under the jurisdiction of the exchange chip of heterogeneous networks
It is integrated into a network switching equipment, so sharing same net between heterogeneous networks in this more rail network systems
The quantity of network switching equipment, the network switching equipment is relatively fewer, in addition, being under the jurisdiction of the exchange of heterogeneous networks in the network switching equipment
Chip can share some modules, be conducive to save space resources, reduce the volume of the network switching equipment, therefore this more rail networks
The complexity of system is relatively low, and scale is also relatively small.
In some exemplary embodiments of the application, every node device 200 further includes convergence/distribution (mux/de-mux)
Module.Convergence/the distribution module is connect with the K network port in its affiliated node device 200, for that will come from its affiliated section
The network switching equipment connecting with its affiliated node device 200 is sent in point device 200 after the signal gathering of the K network port
In K exchange chip, or by the K exchange chip in the network switching equipment being connect with its affiliated node device 200
Signal be distributed to the K network port in its affiliated node device 200.
In some exemplary embodiments of the application, as shown in Figure 1 b, at least one of every network switching equipment is altogether
Enjoying module includes: M2 external port modules.One end of each externally port module is connect with a node device, the other end and
N number of exchange chip connection in its belonging network switching equipment, for that will distribute from the signal of node device connected to it
N number of exchange chip into its belonging network switching equipment, or will be from N number of exchange chip in its belonging network switching equipment
Signal gathering after be sent to node device connected to it.
Optionally, as illustrated in figure 1 c, each exchange chip includes M2 outside port.M2 of each exchange chip is external
Externally port module connects one to one with M2 in its belonging network switching equipment for port.
In some exemplary embodiments of the application, as shown in Figure 1 d, at least one of every network switching equipment is altogether
Enjoy module further include: Intranet control module.Intranet control module in every network switching equipment is exchanged with its belonging network to be set
N number of exchange chip connection in standby, carries out data friendship for controlling between N number of exchange chip in its belonging network switching equipment
It changes;Wherein, it is connected with each other between N number of exchange chip in Intranet control module belonging network switching equipment.
Optionally, as illustrated in figure 1 c, each exchange chip also includes M1 internal port.In every network switching equipment
N number of exchange chip passes through the M1 internal port respectively contained and interconnects;M1 is natural number, and M1≤N.
In some exemplary embodiments of the application, as shown in fig. le, at least one of every network switching equipment is altogether
It enjoys module to further include but be not limited to: clock module, power module and/or radiating module.
Wherein, clock module is connect with N number of exchange chip, for providing clock signal to N number of exchange chip.In same net
In network switching equipment, clock module is shared between N number of exchange chip, is conducive to accomplish to synchronize to greatest extent between heterogeneous networks.
Wherein, power module is connect with N number of exchange chip, for powering to N number of exchange chip.Radiating module and N number of friendship
Chip connection is changed, for radiating for N number of exchange chip.In consolidated network switching equipment, power supply is shared between N number of exchange chip
Module and radiating module can not only save the inner space of the network switching equipment, reduce the cost of implementation of the network switching equipment,
But also efficiency can be greatly improved.
For the node device in above-mentioned more rail network systems, the embodiment of the present application provides a kind of traffic distribution method.Such as
Shown in Fig. 3 a, this method comprises:
301, current data packet is obtained.
302, for the purpose of flow equalization, target port choosing is determined from the corresponding port selection parameter of traffic distribution strategy
Select parameter.
303, target port selection parameter, the selection target network-side from least two network ports of node device are based on
Mouthful, wherein at least two network ports of node device are respectively connected to different networks.
304, current data packet is sent in the network that target network port is accessed by target network port.
In the present embodiment, node device has at least two network ports, and at least two network ports are respectively connected to
Different networks.For node device, packet delivery can be transmitted into corresponding network to be sent to destination.
In order to which as far as possible equably by packet delivery into heterogeneous networks, node device can carry out data packet for the purpose of flow equalization
Distribution.
Wherein, node device distribute each data packet mode it is same or like, in the present embodiment, with current data packet
For be described in detail packet delivery process.
In the present embodiment, the traffic distribution strategy of node device use can be preset, and is traffic distribution strategy
Corresponding port selection parameter is configured, which is used to select to pass under corresponding traffic distribution strategy for node device
Network used in transmission of data packet.
In application scenes, node device is data source, oneself can generate data packet and need to send out data packet
It send to destination, then when node device generates new data packet, available newly generated data packet is worked as to be distributed
Preceding data packet.In other application scenarios, node device can be used as transfer or trunking, then is set downstream when receiving
It, can be using the data packet received as current data packet to be sent when the standby data packet for reporting or sending.
Either any application scenarios, after obtaining current data packet, node device can for the purpose of flow equalization,
Target port selection parameter is determined from the corresponding port selection parameter of traffic distribution strategy, is based on target port selection parameter,
The selection target network port from least two network ports of node device, and then pass through target network port for current data
Packet is sent in the network accessed with target network port.
In the present embodiment, node device combination traffic distribution strategy and the corresponding port selection ginseng of traffic distribution strategy
Number selects transmission network for the purpose of flow equalization for data packet, it is ensured that flow equalization in each network, convenient for playing more nets
The advantage of network.
Wherein, traffic distribution strategy is different, corresponding to port selection parameter also can all differences.It is given below two kinds
Example:
In a kind of example, traffic distribution strategy is traffic distribution strategy packet-by-packet, in this scenario, traffic distribution strategy
Corresponding port selection parameter is the serial number of at least two network ports.In this example, as shown in Figure 3b, a kind of traffic distribution
Process includes:
311, current data packet is obtained.
312, according at least two network ports of the serial number for the network port for sending last data packet and node device
Port sum, determines initial sequence number value, as target port selection parameter from the serial number of at least two network ports, wherein
At least two network ports are respectively connected to different networks.
313, successively subsequent each since at least two network ports the network port of serial number initial sequence number value
Whether the currently available flow of the network port is greater than or equal to the length of current data packet, and is greater than or waits from currently available flow
The selection target network port in the network port of the length of current data packet.
314, current data packet is sent in the network that target network port is accessed by target network port.
Optionally, a kind of implementation of step 312 includes: to determine initial sequence number according to formula s=(t+1) mod K
Value.Wherein, s indicates initial sequence number value, and t indicates to send the serial number of the network port of last data packet, and K indicates at least two networks
The port sum of port, K is natural number.
In the present embodiment, using traffic distribution strategy packet-by-packet, based on the network port for sending last data packet,
And the currently available flow selection target network port of each network port is combined, be conducive to data packet evenly spreading to difference
In network, flow equalization is realized.
Further, in conjunction with the network switching equipment provided by the above embodiment, believed between heterogeneous networks using identical clock
Number, it can accomplish that maximum clock is synchronous between heterogeneous networks, the different data packet in same data flow is in heterogeneous networks
Middle transmission will not cause random ordering.
In another example, traffic distribution strategy is the traffic distribution strategy of valley-fill, in this scenario, traffic distribution plan
Slightly corresponding port selection parameter is at least two hash functions.It in this example, can when determining target port selection parameter
With according to the diversity factor of the currently available flow of at least two network ports, the selection target Hash from least two hash functions
Function, as target port selection parameter.As shown in Figure 3c, a kind of traffic distribution process includes:
321, current data packet is obtained.
322, from the currently available flow of at least two network ports of node device, determine maximum useful discharge and
Minimum utilizable flow, wherein at least two network ports are respectively connected to different networks.
323, judge whether maximum useful discharge and the difference of minimum utilizable flow are greater than given threshold;If judging result is
It is that the i.e. difference of maximum useful discharge and minimum utilizable flow is greater than given threshold, then executes 324.Conversely, executing step 325.
324, other Hash of the selection in addition to the hash function that last data packet uses from least two hash functions
Function executes step 326 as target hash function.
325, the hash function for using last data packet is as target hash function, and executes step 326.
326, it is based on target hash function, the selection target network port from least two network ports, and executes step
327。
It is alternatively possible to use target hash function based on the corresponding stream information X of current data packet (such as 5 tuples)
Hashd(X) Hash is carried out, using the result of the Hash as the number of target network port, selects the network-side with the number
Mouth is used as target network port.
327, current data packet is sent in the network accessed with target network port by target network port.
Optionally, a kind of embodiment of step 324 includes: to determine target Hash letter using formula d=(c+1) mod m
Number.Wherein, d indicates the label of target hash function;C indicates the label of the first hash function;M indicates at least two hash functions
Sum.
In the present embodiment, using a variety of hash functions, and combine the flow difference in network, when the difference in flow compared with
It is when big, i.e., new by being selected for current data packet when the difference of maximum useful discharge and minimum utilizable flow is greater than given threshold
Hash function can be by current data thus in the case where current data packet and last data packet belong to same data flow
Packet is distributed in new network, achievees the purpose that flow equalization.
During the traffic distribution shown in above-mentioned Fig. 3 b and Fig. 3 c, it can also be configured for each network port of node device
Flowmeter counter, is respectively intended to count the flow information of each network port, such as currently available flow or has currently used flow
Deng, and the counter update cycle is set for these flowmeter counters, to be the prset counter update cycle in current time
When integral multiple, the flowmeter counter of each network port is reset into initial value, realizes the update to flowmeter counter.It is optional
Ground, counter update cycle can be and the time-consuming relevant numerical value of the transmission of data packet in a network.For example, can be rule of thumb
It counts each network transmission and completes the time span of a data packet, and set the counter update cycle for the time span.
In conjunction with above-mentioned flowmeter counter and counter update cycle, as shown in Figure 3d, a kind of traffic distribution side packet-by-packet
Method includes:
331, node device is initialized, flowmeter counter FC, flowmeter is respectively set for the K network port of node device
Number device FC is for recording the data volume size that corresponding network port can also export, and the initial value that flowmeter counter FC is arranged is
Q, the period for restoring initial value is Ta。
332, after the completion of initialization operation, the network port that random selection number is orig is as first transmission data packet
The network port, go to step 333.
333, when receiving the data packet that a length is L1, which is sent from the network port numbered as orig
The network accessed to the number by the network port of orig, and update the flowmeter counter for the network port that number is orig
FCorig=FCorig- L1, the currently scheduled number s=orig of record, skips to step 334;
334, judge whether current time is the cycle T for restoring initial valueaIntegral multiple, if so, going to step
338;If it is not, then going to step 335;
335, the data packet that length is Lx is received, since number s, first is found and meets FC >=Lx network port
T gos to step 337 if can find;If it is not, then going to step 336.
336, the value for updating the flowmeter counter of all-network port is FC=Q+FC, gos to step 334.
337, the data packet that length is Lx is sent to the network port institute that the number is t from the network port numbered as t
The network of access, and update the flowmeter counter FC for the network port that number is tt=FCt- Lx, the currently scheduled number of record
S=(t+1) mod K, gos to step 334.
338, the value of the flowmeter counter FC of all-network port is re-set as initial value Q, and randomly chooses number
For orig the network port as the network port for sending next data packet, go to step 334.
It is that current data packet selects network-side based on the currently scheduled network port in above-mentioned packet-by-packet distribution method
Mouthful, and combine and periodically randomly choose the network port again, flow equalization may be implemented to a certain extent, flow is avoided to be stacked into certain
Congestion is caused in a network.
In conjunction with above-mentioned flowmeter counter and counter update cycle, as shown in Figure 3 e, a kind of traffic distribution side of valley-fill
Method includes:
341, node device is initialized, flowmeter counter FCN, flow is respectively set for the K network port of node device
Counter FCN is arranged flowmeter counter FCN and restores initial for recording the data volume size that corresponding network port has exported
The period of value is Tb, and m hash function of setting be labeled as Hash_1~Hash_m, gos to step 342.Wherein, m is certainly
So number, and m >=2.
342, judge whether current time is the cycle T for restoring initial valuebIntegral multiple, if so, going to step
349;If it is not, then going to step 343.
343, the data packet that length is Ly is received, gos to step 344.
344, hash function Hash_c (X) is used based on the corresponding stream information X of data packet (such as 5 tuples) that length is Ly
Hash is carried out, using the result of the Hash as number tar, gos to step 345.
Wherein, Hash_c (X) indicates currently scheduled hash function, that is, the Hash used when the upper data packet of transmission
Function, c indicate the number of currently scheduled hash function, and c is natural number, and 1≤c≤m.
345, the maximum value and minimum value in all flowmeter counter FC are obtained.
346, judge whether the difference between maxima and minima is greater than given threshold;If maxima and minima it
Between difference be greater than given threshold, illustrate to load not perfectly flat weighing apparatus between heterogeneous networks, go to step 347;If maximum value
Difference between minimum value is less than or equal to given threshold, illustrates the load relative equilibrium between heterogeneous networks, gos to step
348。
347, hash function is replaced using formula Hash_c=Hash_ [(c+1) mod m], gos to step 344.
348, the data packet that length is Ly is sent to the network port numbered as tar from the network port numbered as tar
The network accessed, and update FCNtar=FCNtar+ Ly, gos to step 342.
Wherein, FCNtarIndicate the flowmeter counter for the network port that number is tar.
349, the value of the flowmeter counter FC of all-network port is re-set as 0, gos to step 343.
In above-mentioned valley-fill distribution method, multiple hash functions are based on, can be in the unbalanced situation of network load, it will
Flow equalization may be implemented into heterogeneous networks in packet delivery in same data flow to a certain extent, and flow is avoided to accumulate
Congestion is caused into some network.
For the network switching equipment in above-described embodiment, the embodiment of the present application provides a kind of method for routing.Such as Fig. 4 a institute
Show, this method comprises:
401, it is based on routing policy, determines that data packet to be sent is corresponding from N number of exchange chip of the network switching equipment
First exchange chip, N are natural numbers, and N >=2.
402, when the first exchange chip meets the network switching condition of setting, the is selected from other reachable exchange chips
Two exchange chips, other reachable exchange chips refer in N number of exchange chip in addition to the first exchange chip to data packet to be sent
Destination route reachable exchange chip.
403, based on the internal network between the first exchange chip and the second exchange chip, controlling the second exchange chip will be to
Data packet is sent to send.
It include N number of exchange chip in the network switching equipment of the present embodiment, N number of exchange chip is under the jurisdiction of different nets
Network, interconnection forms the internal network of the network switching equipment between N number of exchange chip, and N is natural number, and N >=2.For entering net
Data packet in network switching equipment, the internal network based on the network switching equipment can when meeting network switching condition by it from
The network that the network switching that one exchange chip is accessed is accessed to the second exchange chip realizes between same position exchange chip
Interconnect, flow arbitrarily switched in heterogeneous networks, realize more rail network systems dynamic load leveling and
High availability has also sufficiently excavated the multipath in more rail network systems, convenient for giving full play to the advantage of more rail network systems.
In above method embodiment, network switching condition can be depending on concrete application scene.In application scenes,
Whether network switching condition can be current exchange chip and breaks down and/or congestion.Based on this, as shown in Figure 4 b, Yi Zhonglu
Include: by method
411, it is based on routing policy, determines that data packet to be sent is corresponding from N number of exchange chip of the network switching equipment
First exchange chip, N are natural numbers, and N >=2.
412, judge on the first exchange chip for sending the outside port of data packet to be sent with the presence or absence of failure.If
Judging result be it is yes, illustrate that the first exchange chip meets network switching condition, execute step 413-415;If it is judged that being
It is no, execute step 416.
413, the port numbers based on the outside port on the first exchange chip for sending data packet to be sent, obtain other
Up to the flow load of the outside port on exchange chip with identical port numbers.
414, according on other reachable exchange chips with identical port numbers outside port flow load, from it is other can
Up to selecting flow load to meet the exchange chip of sets requirement as the second exchange chip in exchange chip.
Optionally, the second exchange chip can be the outside port in other reachable exchange chips with identical port numbers
The minimum exchange chip of flow load, alternatively, being also possible to the outer end in other reachable exchange chips with identical port numbers
The flow load of mouth is less than the exchange chip of setting flow load threshold value, alternatively, being also possible to have in other reachable exchange chips
There is the flow load of the outside port of identical port numbers to be located at a certain range of exchange chip, etc..
415, based on the internal network between the first exchange chip and the second exchange chip, controlling the second exchange chip will be to
Data packet is sent to send.
416, the first exchange chip of control sends data packet to be sent.
Optionally, in step 412, can also judge on the first exchange chip for sending the outside of data packet to be sent
Whether the packet loss of port is more than packet loss threshold value.If it is judged that be it is yes, illustrate that the first exchange chip meets network switching
Condition thens follow the steps 413-415;If it is judged that being no, execution step 416.
Optionally, in step 412, can also judge on the first exchange chip for sending the outside of data packet to be sent
Whether the average queuing delay of port is more than delay threshold value etc..If it is judged that be it is yes, then follow the steps 413-415;If
Judging result is no, execution step 416.
Optionally, in step 412, the condition that can be combined with two or more carries out comprehensive descision.
When the present embodiment can encounter failure/heavy congestion in a certain network, the data packet in the network is switched to
It is transmitted in another underloading network, can accomplish the load balancing between heterogeneous networks, and the transmission that can reduce data packet is prolonged
When, guarantee the accessibility of data packet.
It should be noted that containing in some processes of the description in above-described embodiment and attached drawing according to specific suitable
Multiple operations that sequence occurs, but it should be clearly understood that these operations can not be held according to its sequence what appears in this article
Capable or parallel execution, serial number of operation such as 301,302 etc. are only used for distinguishing each different operation, and serial number itself is no
It represents and any executes sequence.In addition, these processes may include more or fewer operations, and these operations can be by suitable
Sequence executes or parallel execution.It should be noted that the description such as " first " herein, " second ", is for distinguishing different disappear
Breath, equipment, module etc. do not represent sequencing, also do not limit " first " and " second " and are different type.
In addition to the above-mentioned network switching equipment, network system and correlation method embodiment, some embodiments of the application are also
A kind of node device is provided, as shown in figure 5, the node device includes: K network port 51, memory 52 and processor 53.
Wherein, the K network port 51, for accessing the different networks of K, K is natural number, and K >=2.
Memory 52 for storing computer program, and can be configured to store various other data to support in node
Operation in equipment.The example of these data includes the finger of any application or method for operating on node device
It enables, contact data, telephone book data, message, picture, video etc..
Memory 52 can be by any kind of volatibility or non-volatile memory device or their combination realization, such as
Static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable is read-only to be deposited
Reservoir (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or
CD.
Processor 53 is coupled with memory 52, for executing the computer program in memory 52, to be used for: acquisition is worked as
Preceding data packet;For the purpose of flow equalization, target port selection is determined from the corresponding port selection parameter of traffic distribution strategy
Parameter;Based on target port selection parameter, the selection target network port from the K network port;It will by target network port
Current data packet is sent in the network that target network port is accessed.
In an optional embodiment, as shown in figure 5, the node device further include: convergence/distribution module 54.
Convergence/distribution module 54 is connect with the K network port 51, for will be from the signal gathering of the K network port 51
The K exchange chip or automatic network switching equipment in future being sent in the network switching equipment being connect with the node device afterwards
The signal of middle K exchange chip is distributed to K network port 51.Wherein, the network switching equipment includes N number of exchange chip, and N is certainly
So number, N >=2 and K≤N.
In the present embodiment, the realization form of node device is not limited, node device can be any be connectable to
On exchange chip and the equipment with the K network port, such as it can be work station, terminal device, the network user or personal count
Calculation machine, the equipment that can also be server, printer or other network connections.
In conjunction with the realization form of node device, in some optional embodiments, which can also include: communication
Other components such as component, display, power supply module, audio component.
Wherein, wired or wireless way between equipment and other equipment where communication component is configured to facilitate communication component
Communication.Equipment where communication component can access the wireless network based on communication standard, such as WiFi, 2G or 3G or they
Combination.In one exemplary embodiment, communication component receives the broadcast from external broadcasting management system via broadcast channel
Signal or broadcast related information.In one exemplary embodiment, the communication component further includes near-field communication (NFC) module,
To promote short range communication.For example, NFC module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology,
Ultra wide band (UWB) technology, bluetooth (BT) technology and other technologies are realized.
Wherein, display includes screen, and screen may include liquid crystal display (LCD) and touch panel (TP).If
Screen includes touch panel, and screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes
One or more touch sensors are to sense the gesture on touch, slide, and touch panel.The touch sensor can be not only
The boundary of a touch or slide action is sensed, but also detects duration and pressure associated with the touch or slide operation.
Wherein, power supply module, the various assemblies of equipment provide electric power where power supply module.Power supply module may include electricity
Management system, one or more power supplys and other are associated with electric power is generated, managed, and distributed for equipment where power supply module
Component.
Wherein, audio component can be configured to output and/or input audio signal.For example, audio component includes a wheat
Gram wind (MIC), the equipment where audio component is in operation mode, when such as call mode, recording mode, and voice recognition mode,
Microphone is configured as receiving external audio signal.The received audio signal can be further stored in memory or via
Communication component is sent.In some embodiments, audio component further includes a loudspeaker, is used for output audio signal.
Correspondingly, the embodiment of the present application also provides a kind of computer readable storage medium for being stored with computer program, should
Computer program, which is performed, can be realized each step or operation that can be executed by node device in above method embodiment.
Correspondingly, the embodiment of the present application also provides another computer readable storage medium for being stored with computer program,
The computer program is performed that can be realized in above method embodiment can be by each step that the network switching equipment executes or behaviour
Make.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (31)
1. a kind of network switching equipment characterized by comprising N number of exchange chip and can be shared by N number of exchange chip
At least one sharing module;N number of exchange chip is under the jurisdiction of N number of different network, and N number of exchange chip is formed
The M2 network port accesses corresponding network for the M2 platform node device for connecting M2 platform node device;Wherein, N and M2
It is natural number, and N >=2.
2. the network switching equipment according to claim 1, which is characterized in that N number of exchange chip is in respective affiliated net
Network cascade level is identical.
3. the network switching equipment according to claim 1, which is characterized in that at least one described sharing module includes: M2
A external port module;
One end of each externally port module is connect with a node device, and the other end is connect with N number of exchange chip, is used for
Signal from node device connected to it is distributed to N number of exchange chip, or N number of exchange chip will be come from
Signal gathering after be sent to node device connected to it.
4. the network switching equipment according to claim 3, which is characterized in that each exchange chip includes M2 outer end
Mouthful, the M2 outside port connects one to one with the M2 external port modules.
5. the network switching equipment according to claim 1, which is characterized in that at least one described sharing module includes: interior
Net control module;
The Intranet control module is connect with N number of exchange chip, carries out data between N number of exchange chip for controlling
Exchange, wherein be connected with each other between N number of exchange chip.
6. the network switching equipment according to claim 5, which is characterized in that each exchange chip includes M1 inside end
Mouthful;N number of exchange chip passes through the M1 internal port respectively contained and interconnects;M1 is natural number, and M1≤N.
7. the network switching equipment according to claim 5, which is characterized in that the Intranet control module is specifically used for:
Based on routing policy, corresponding first exchange chip of data packet to be sent is determined from N number of exchange chip;
When first exchange chip meets the network switching condition of setting, second is selected to hand over from other reachable exchange chips
Change chip, other reachable exchange chips refer in N number of exchange chip in addition to first exchange chip to described
The destination of data packet to be sent routes reachable exchange chip;
Based on the internal network between first exchange chip and second exchange chip, second exchange chip is controlled
The data packet to be sent is sent.
8. the network switching equipment according to claim 7, which is characterized in that the Intranet control module is specifically used for:
When, for sending the outside port of the data packet to be sent there are failure, packet loss is more than on first exchange chip
When packet loss threshold value and/or average queuing delay are more than delay threshold value, determine that first exchange chip meets network switching item
Part.
9. the network switching equipment according to claim 7, which is characterized in that the Intranet control module is specifically used for:
Based on the port numbers of the outside port on first exchange chip for sending the data packet to be sent, described in acquisition
The flow load of outside port on other reachable exchange chips with identical port numbers;
According on other reachable exchange chips with identical port numbers outside port flow load, from it is described it is other can
Up to selecting flow load to meet the exchange chip of sets requirement as second exchange chip in exchange chip.
10. -9 described in any item network switching equipment according to claim 1, which is characterized in that at least one described shared mould
Block includes:
The clock module being connect with N number of exchange chip, for providing clock signal to N number of exchange chip;And/or
The power module being connect with N number of exchange chip, for powering to N number of exchange chip;And/or
The radiating module being connect with N number of exchange chip, for radiating for N number of exchange chip.
11. a kind of network system characterized by comprising an at least node device and at least a network switching equipment;
Every network switching equipment includes N number of exchange chip and can be shared by least one that N number of exchange chip is shared
Module;N number of exchange chip is under the jurisdiction of N number of different network, and N number of exchange chip forms the M2 network port,
For connecting the M2 platform node device in an at least node device, corresponding net is accessed for the M2 platform node device
Network;Wherein, N and M2 is natural number, and N >=2;
Every node device includes the K network port, is connected for corresponding with K exchange chip in a network switching equipment
It connects, K is natural number, and K≤N;
Wherein, it is connected with each other between an at least network switching equipment according to the cascade level of the network switching equipment.
12. network system according to claim 11, which is characterized in that N number of exchange core in every network switching equipment
Piece is identical in respective belonging network cascade level;Exist between an at least network switching equipment according to inner exchanging chip
Cascade level in belonging network is connected with each other.
13. network system according to claim 11, which is characterized in that every node device further includes convergence/distribution mould
Block;
Convergence/the distribution module is connect with the K network port in its affiliated node device, for that will come from its affiliated node
K be sent to after the signal gathering of the K network port in equipment in the network switching equipment being connect with its affiliated node device
Exchange chip, or the signal of the K exchange chip in the network switching equipment connecting with its affiliated node device is distributed
The K network port into its affiliated node device.
14. network system according to claim 11, which is characterized in that at least one of every network switching equipment is altogether
Enjoying module includes: M2 external port modules;
One end of each externally port module is connect with a node device, the N in the other end and its belonging network switching equipment
A exchange chip connection, for the signal from node device connected to it to be distributed in its belonging network switching equipment
N number of exchange chip, or connect being sent to after the signal gathering of N number of exchange chip in its belonging network switching equipment with it
The node device connect.
15. network system according to claim 14, which is characterized in that each exchange chip includes M2 outside port,
Externally port module connects one to one the M2 outside port with M2 in its belonging network switching equipment.
16. network system according to claim 11, which is characterized in that at least one of every network switching equipment is altogether
Enjoying module includes: Intranet control module;
The Intranet control module is connect with N number of exchange chip in its belonging network switching equipment, for controlling its affiliated net
Data exchange is carried out in network switching equipment between N number of exchange chip;Wherein, the Intranet control module belonging network switching equipment
In N number of exchange chip between be connected with each other.
17. network system according to claim 16, which is characterized in that each exchange chip includes M1 internal port;
N number of exchange chip in every network switching equipment passes through the M1 internal port respectively contained and interconnects;M1 is natural number, and M1
≤N。
18. the described in any item network systems of 1-17 according to claim 1, which is characterized in that in every network switching equipment
At least one sharing module includes:
The clock module being connect with its internal N number of exchange chip, for providing clock signal to its internal N number of exchange chip;With/
Or
The power module being connect with its internal N number of exchange chip, for powering to its internal N number of exchange chip;And/or
The radiating module being connect with its internal N number of exchange chip, for radiating for its internal N number of exchange chip.
19. a kind of traffic distribution method, is suitable for inclusion in the node device of at least two network ports, which is characterized in that described
Method includes:
Obtain current data packet;
For the purpose of flow equalization, target port selection parameter is determined from the corresponding port selection parameter of traffic distribution strategy;
Based on the target port selection parameter, the selection target network port from least two network port;
The current data packet is sent in the network that the target network port is accessed by the target network port,
Wherein, at least two network port is respectively connected to different networks.
20. according to the method for claim 19, which is characterized in that the corresponding port selection parameter of the traffic distribution strategy
For the serial number of at least two network port;
It is described for the purpose of flow equalization, from the corresponding port selection parameter of traffic distribution strategy determine target port selection ginseng
Number, comprising:
According to the port of the serial number for the network port for sending last data packet and at least two network port sum, from described
Initial sequence number value is determined in the serial number of at least two network ports, as the target port selection parameter.
21. according to the method for claim 20, which is characterized in that it is described to be based on the target port selection parameter, from institute
State the selection target network port at least two network ports, comprising:
It is successively subsequent each since at least two network port the network port of initial sequence number value described in serial number
Whether the currently available flow of the network port is greater than or equal to the length of the current data packet, and is greater than from currently available flow
Or the target network port is selected in the network port equal to the length of the current data packet.
22. according to the method for claim 20, which is characterized in that the network port according to transmission last data packet
The port of serial number and at least two network port sum, determines initial sequence from the serial number of at least two network port
Number value, comprising:
According to formula s=(t+1) mod K, the initial sequence number value is determined;
Wherein, s indicates the initial sequence number value, and t indicates to send the serial number of the network port of last data packet, and K expression is described extremely
The port sum of few two network ports.
23. according to the method for claim 19, which is characterized in that the corresponding port selection parameter of the traffic distribution strategy
For at least two hash functions;
It is described for the purpose of flow equalization, from the corresponding port selection parameter of traffic distribution strategy determine target port selection ginseng
Number, comprising:
According to the diversity factor of the currently available flow of at least two network port, selected from least two hash function
Target hash function is selected, as the target port selection parameter.
24. according to the method for claim 23, which is characterized in that described according to the current of at least two network port
The diversity factor of utilizable flow, the selection target hash function from least two hash function, comprising:
From the currently available flow of at least two network port, maximum useful discharge and minimum utilizable flow are determined;
If the difference of the maximum useful discharge and the minimum utilizable flow is greater than given threshold, from least two Hash
Select other hash functions in addition to the hash function that last data packet uses as the target hash function in function;
If the difference of the maximum useful discharge and the minimum utilizable flow is less than or equal to given threshold, by last data packet
The hash function used is as the target hash function.
25. according to the method for claim 24, which is characterized in that described to select to remove from least two hash function
Other hash functions except the hash function that last data packet uses are as the target hash function, comprising:
Using formula d=(c+1) mod m, the target hash function is determined;
D indicates the label of the target hash function;
C indicates the label of first hash function;
M indicates the sum of at least two hash function.
26. the described in any item methods of 9-25 according to claim 1, which is characterized in that further include:
When current time is the integral multiple of prset counter update cycle, the flow of at least two network port is counted
Think highly of and be set to initial value, the flowmeter counter is used to record the flow information of corresponding network port.
27. a kind of method for routing, which is characterized in that be suitable for the described in any item network switching equipment of claim 1-10, institute
The method of stating includes:
Based on routing policy, data packet corresponding first to be sent is determined from N number of exchange chip of the network switching equipment
Exchange chip;
When first exchange chip meets the network switching condition of setting, second is selected to hand over from other reachable exchange chips
Change chip, other reachable exchange chips refer in N number of exchange chip in addition to first exchange chip to described
The destination of data packet to be sent routes reachable exchange chip;
Based on the internal network between first exchange chip and second exchange chip, second exchange chip is controlled
The data packet to be sent is sent.
28. according to the method for claim 27, which is characterized in that in the second exchange of selection from other reachable exchange chips
Before chip, the method also includes:
When, for sending the outside port of the data packet to be sent there are failure, packet loss is more than on first exchange chip
When packet loss threshold value and/or average queuing delay are more than delay threshold value, determine that first exchange chip meets network switching item
Part.
29. the method according to claim 27 or 28, which is characterized in that described to select from other reachable exchange chips
Two exchange chips, comprising:
Based on the port numbers of the outside port on first exchange chip for sending the data packet to be sent, described in acquisition
The flow load of outside port on other reachable exchange chips with identical port numbers;
According on other reachable exchange chips with identical port numbers outside port flow load, from it is described it is other can
Up to selecting flow load to meet the exchange chip of sets requirement as second exchange chip in exchange chip.
30. a kind of node device characterized by comprising the K network port, memory and processor;
The K network port, for accessing the different networks of K, K is natural number, and K >=2;
The memory, for storing computer program;
The processor is coupled with the memory, for executing the computer program, to be used for:
Obtain current data packet;
For the purpose of flow equalization, target port selection parameter is determined from the corresponding port selection parameter of traffic distribution strategy;
Based on the target port selection parameter, the selection target network port from the K network port;
The current data packet is sent in the network that the target network port is accessed by the target network port.
31. node device according to claim 30, which is characterized in that further include: convergence/distribution module,
Convergence/the distribution module is connect with the K network port, for that will converge from the signal of the K network port
K exchange chip being sent to after poly- in the network switching equipment being connect with the node device, or the network will be come from
The signal of K exchange chip is distributed to the K network port in switching equipment;The network switching equipment includes N number of exchange
Chip, N are natural number, N >=2 and K≤N.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810271926.2A CN110324265B (en) | 2018-03-29 | 2018-03-29 | Traffic distribution method, routing method, equipment and network system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810271926.2A CN110324265B (en) | 2018-03-29 | 2018-03-29 | Traffic distribution method, routing method, equipment and network system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110324265A true CN110324265A (en) | 2019-10-11 |
CN110324265B CN110324265B (en) | 2021-09-07 |
Family
ID=68110886
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810271926.2A Active CN110324265B (en) | 2018-03-29 | 2018-03-29 | Traffic distribution method, routing method, equipment and network system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110324265B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114079627A (en) * | 2020-08-14 | 2022-02-22 | 华为技术有限公司 | Data transmission device and method |
CN114697275A (en) * | 2020-12-30 | 2022-07-01 | 深圳云天励飞技术股份有限公司 | Data processing method and device |
CN116192696A (en) * | 2023-01-13 | 2023-05-30 | 北京信而泰科技股份有限公司 | Method, device and storage medium for traffic congestion prevention configuration |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101404616A (en) * | 2008-11-04 | 2009-04-08 | 北京大学深圳研究生院 | Load balance grouping and switching structure and its construction method |
CN101729424A (en) * | 2009-12-16 | 2010-06-09 | 杭州华三通信技术有限公司 | Flow forwarding method, devices and system |
CN104243357A (en) * | 2014-09-02 | 2014-12-24 | 深圳市腾讯计算机系统有限公司 | Switch, switching system, switching network chip assemblies and forwarding chip assemblies |
US20150036682A1 (en) * | 2005-10-25 | 2015-02-05 | Brocade Communications Systems, Inc. | Interface Switch for use with Fibre Channel Fabrics in Storage Area Networks |
US20160170916A1 (en) * | 2014-12-12 | 2016-06-16 | Freescale Semiconductor, Inc. | Coherent memory interleaving with uniform latency |
CN106302252A (en) * | 2015-05-15 | 2017-01-04 | 华为技术有限公司 | Data exchange system framework, the method sending data traffic and switch |
CN107493245A (en) * | 2017-09-22 | 2017-12-19 | 锐捷网络股份有限公司 | The board and stream compression forwarding method of interchanger |
-
2018
- 2018-03-29 CN CN201810271926.2A patent/CN110324265B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150036682A1 (en) * | 2005-10-25 | 2015-02-05 | Brocade Communications Systems, Inc. | Interface Switch for use with Fibre Channel Fabrics in Storage Area Networks |
CN101404616A (en) * | 2008-11-04 | 2009-04-08 | 北京大学深圳研究生院 | Load balance grouping and switching structure and its construction method |
CN101729424A (en) * | 2009-12-16 | 2010-06-09 | 杭州华三通信技术有限公司 | Flow forwarding method, devices and system |
CN104243357A (en) * | 2014-09-02 | 2014-12-24 | 深圳市腾讯计算机系统有限公司 | Switch, switching system, switching network chip assemblies and forwarding chip assemblies |
US20160170916A1 (en) * | 2014-12-12 | 2016-06-16 | Freescale Semiconductor, Inc. | Coherent memory interleaving with uniform latency |
CN106302252A (en) * | 2015-05-15 | 2017-01-04 | 华为技术有限公司 | Data exchange system framework, the method sending data traffic and switch |
CN107493245A (en) * | 2017-09-22 | 2017-12-19 | 锐捷网络股份有限公司 | The board and stream compression forwarding method of interchanger |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114079627A (en) * | 2020-08-14 | 2022-02-22 | 华为技术有限公司 | Data transmission device and method |
CN114697275A (en) * | 2020-12-30 | 2022-07-01 | 深圳云天励飞技术股份有限公司 | Data processing method and device |
CN114697275B (en) * | 2020-12-30 | 2023-05-12 | 深圳云天励飞技术股份有限公司 | Data processing method and device |
CN116192696A (en) * | 2023-01-13 | 2023-05-30 | 北京信而泰科技股份有限公司 | Method, device and storage medium for traffic congestion prevention configuration |
CN116192696B (en) * | 2023-01-13 | 2024-06-07 | 北京信而泰科技股份有限公司 | Method, device and storage medium for traffic congestion prevention configuration |
Also Published As
Publication number | Publication date |
---|---|
CN110324265B (en) | 2021-09-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10700965B2 (en) | System and method for a control plane reference model framework | |
Wu et al. | Towards distributed SDN: Mobility management and flow scheduling in software defined urban IoT | |
US8982731B2 (en) | Bluetooth network configuration | |
CN103222287B (en) | For the probe message that direct link connects | |
CN103404094B (en) | Data processing method, system and switch | |
WO2023039965A1 (en) | Cloud-edge computing network computational resource balancing and scheduling method for traffic grooming, and system | |
CN108353314A (en) | Switching is for by providing both backhaul and forward pass(XHAUL)The data-signal of at least two types of the transmission of internuncial transmission network | |
US20130235774A1 (en) | Energy-saving mobile node control method using wireless multi-interfaces | |
CN110324265A (en) | Traffic distribution method, method for routing, equipment and network system | |
CN103024001A (en) | Business scheduling method and device, and fusion equipment | |
CN105684530B (en) | For constructing the distributed method of the hierarchical structure synchronous for device-to-device with safeguarding | |
CN109510780A (en) | Flow control method, exchange chip and the network equipment | |
US20180007091A1 (en) | Synchronous media rendering over a spatially extended wireless network | |
CN107113188A (en) | Access network based on cloud | |
CN105391651B (en) | Virtual optical network multi-layer resource convergence method and system | |
CN102904837B (en) | Differentiated service survivability method based on virtual service plane | |
US20150043911A1 (en) | Network Depth Limited Network Followed by Compute Load Balancing Procedure for Embedding Cloud Services in Software-Defined Flexible-Grid Optical Transport Networks | |
WO2017070254A1 (en) | System for media distribution and rendering on spatially extended wireless networks | |
CN113315717A (en) | Service balancing method and device, electronic equipment and storage medium | |
US11272423B2 (en) | Live video analytics over high frequency wireless networks | |
CN103813480A (en) | Method for construction of mobile communication network | |
Wang et al. | Novel optical access network virtualization and dynamic resource allocation algorithms for the internet of things | |
Xu et al. | A real plug-and-play fog: Implementation of service placement in wireless multimedia networks | |
CN105873155B (en) | A kind of information processing method and electronic equipment | |
WO2019080744A1 (en) | Control channel implementation method, apparatus, device, storage medium, and processing method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |