CN107846364A - A kind for the treatment of method and apparatus of message - Google Patents
A kind for the treatment of method and apparatus of message Download PDFInfo
- Publication number
- CN107846364A CN107846364A CN201610833366.6A CN201610833366A CN107846364A CN 107846364 A CN107846364 A CN 107846364A CN 201610833366 A CN201610833366 A CN 201610833366A CN 107846364 A CN107846364 A CN 107846364A
- Authority
- CN
- China
- Prior art keywords
- message
- processor
- address
- sent
- hashed value
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1023—Server selection for load balancing based on a hash applied to IP addresses or costs
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
Abstract
The application provides a kind for the treatment of method and apparatus of message, and applied on the load-balancing device including multiple processors, methods described includes：When receiving message, judge the message destination address whether be the load-balancing device address；If it is not, then determining that the message is the message that user end to server is sent, and calculate hashed value corresponding to the source address of the message；Determine the processor corresponding to the hashed value；The message is sent to the processor, so that the processor is handled the message.Pass through the technical scheme of the application, only multiple messages of same session are handled by a processor, rather than multiple processors compete and this multiple message are handled simultaneously, avoid the competition between multiple processors, so as to make full use of substantial amounts of processor, the resource of processor is saved, improves the utilization rate of processor, the performance and disposal ability of load balancing are lifted, lifts the performance of business.
Description
Technical field
The application is related to Internet technical field, more particularly to a kind for the treatment of method and apparatus of message.
Background technology
As shown in figure 1, it is SLB (Server Load Balancing, server load balancing) or LVS (Linux
Virtual Server, Linux virtual server) networking schematic diagram, multiple servers in server cluster provide jointly
Same application, such as WEB application.The corresponding virtual ip address of multiple servers in server cluster, and each server
There is corresponding real IP address.
When client needs to access server, it is the virtual ip address that can send purpose IP address to load-balancing device
Message.Load-balancing device can be that the message selects a server, and utilize the server after the message is received
Real IP address, the message is sent to the server.Wherein, load-balancing device can be based on load-balancing algorithm as report
Selected works select server, in this manner it is ensured that substantial amounts of message is uniformly distributed to each server, so as to improve multiple services
The disposed of in its entirety performance of device.
In the prior art, multiple processors can be generally included on load-balancing device, and this multiple processor can be simultaneously
Message of the row processing for different clients.For example, processor 1 is used to handle the message from client 1, and processing server
1 is sent to the message of client 1, and processor 2 is used to handle the message from client 2, and processing server 2 is sent to client
The message at end 2.
But above-mentioned technical proposal can cause business processing mistake under some scenes.For example, processor 1 is used to handle
Message 1 and message 2 from client 1, and message 1 and message 2 are sent to server 1, server 1 utilizes message 1 and report
The business of literary 2 pairs of clients 1 is handled.But after load-balancing device receives message 2, message 2 processing is not supplied to
Device 1, but message 2 is supplied to processor 2, now, processor 2 can reselect server, may such as select server 2,
So, message 1 and message 2 are sent to different servers, therefore, it is impossible to by same server by utilizing message 1 and report
The business of literary 2 pairs of clients 1 is handled, so as to cause business processing mistake.
The content of the invention
The application provides a kind of processing method of message, applied on the load-balancing device including multiple processors, institute
The method of stating comprises the following steps：When receiving message, whether the destination address for judging the message is that the load balancing is set
Standby address；If it is not, then determine that the message is the message that user end to server is sent, and with calculating the source of the message
Hashed value corresponding to location；Determine the processor corresponding to the hashed value；The message is sent to the processor, so that institute
Processor is stated to handle the message.
The application provides a kind of processing method of message, applied on the load-balancing device including multiple processors, institute
The method of stating comprises the following steps：When receiving the message that server is sent to client, pass through the destination address of the message
The mapping relations of the address being pre-configured with and processor are inquired about, determine processor corresponding to the destination address of the message；By institute
State message and be sent to the processor, so that the processor is handled the message.
The application provides a kind of processing unit of message, applied on the load-balancing device including multiple processors, institute
Stating device includes：Judging unit, for when receiving message, judging whether the destination address of the message is that the load is equal
The address for the equipment that weighs；Computing unit, for when the judged result of the judging unit is no, it to be client to determine the message
The message sent to server, and calculate hashed value corresponding to the source address of the message；Determining unit, for determining described dissipate
Processor corresponding to train value；Transmitting element, for the message to be sent into the processor, so that the processor is to institute
Message is stated to be handled.
The application provides a kind of processing unit of message, applied on the load-balancing device including multiple processors, institute
Stating device includes：Determining unit, for when receiving the message that server is sent to client, passing through the purpose of the message
The address and the mapping relations of processor that address lookup is pre-configured with, determine processor corresponding to the destination address of the message；
Transmitting element, for the message to be sent into the processor, so that the processor is handled the message.
Based on above-mentioned technical proposal, in the embodiment of the present application, multiple messages for same session can be sent to together
One processor, multiple messages are handled by the processor, so as to avoid business processing mistake.For example, for same
The message 1 and message 2 of session, load-balancing device are supplied to processor 1 after message 1 is received, by message 1, received
After message 2, message 2 is supplied to processor 1, message 1 and message 2 are sent to server 1 by processor 1, so as to avoid industry
Business processing mistake.Moreover, on load-balancing device, only multiple messages of same session can be carried out by a processor
Processing, rather than multiple processors compete and this multiple message are handled simultaneously, avoid the competition between multiple processors,
So as to make full use of substantial amounts of processor, the resource of processor is saved, improves the utilization rate of processor, lifts load balancing
Performance and disposal ability, lift the performance of business.
Brief description of the drawings
, below will be to the application in order to clearly illustrate the embodiment of the present application or technical scheme of the prior art
The required accompanying drawing used is briefly described in embodiment or description of the prior art, it should be apparent that, in describing below
Accompanying drawing is only some embodiments described in the application, for those of ordinary skill in the art, can also be according to these
Accompanying drawing obtains other accompanying drawings.
Fig. 1 and Fig. 2 is SLB or LVS networking schematic diagram；
Fig. 3 is the flow chart of the processing method of the message in a kind of embodiment of the application；
Fig. 4 is the flow chart of the processing method of the message in the application another embodiment；
Fig. 5 is the hardware structure diagram of the load-balancing device in a kind of embodiment of the application；
Fig. 6 is the structure chart of the processing unit of the message in a kind of embodiment of the application；
Fig. 7 is the structure chart of the processing unit of the message in the application another embodiment.
Embodiment
Purpose in term used in this application merely for the sake of description specific embodiment, and unrestricted the application.This Shen
Please " one kind " with the singulative used in claims, " described " and "the" are also intended to including most forms, unless
Context clearly shows that other implications.It is also understood that term "and/or" used herein refers to comprising one or more
Associated list items purpose any or all may combine.
It will be appreciated that though various information, but this may be described using term first, second, third, etc. in the application
A little information should not necessarily be limited by these terms.These terms are only used for same type of information being distinguished from each other out.For example, do not departing from
In the case of the application scope, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as
One information.Depending on linguistic context, in addition, used word " if " can be construed to " ... when " or " when ... "
Or " in response to determining ".
In one example, in SLB or LVS networking, when load-balancing device includes multiple processors, for
Client be sent to server TCP (Transmission Control Protocol, transmission control protocol) messages or
UDP (User Datagram Protocol, UDP) message, the forwarding chip (i.e. network interface card) of load-balancing device
After TCP message or UDP messages is received, five-tuple information (such as source can be parsed from TCP message or UDP messages
Address, source port number, destination address, destination slogan, protocol number etc.), and choose corresponding handle using the five-tuple information
Device, and TCP message or UDP messages are sent to the processor.ICMP (the Internet of server are sent to for client
Control Message Protocol, Internet Control Message agreement) message, the forwarding chip of load-balancing device receiving
To after icmp packet, source address, destination address, protocol number can be parsed from icmp packet, and with utilizing the source parsed
Location, destination address, processor corresponding to protocol number selection, and icmp packet is sent to the processor.Certainly, load balancing is set
The standby process for receiving message, it is not limited to receive message using forwarding chip, can also be configuration in load-balancing device
Functional unit (being realized by program) receives message, as long as load-balancing device can receive message.Retouch for convenience
State, subsequently illustrated so that the forwarding chip (i.e. network interface card) of load-balancing device receives each message as an example.
Clearly as TCP message/UDP messages are different from the selection mode of the processor of icmp packet, therefore, can cause
It is different from the processor of icmp packet to belong to TCP message/UDP messages of same session, so as to cause business processing mistake.Example
Such as, the processor for belonging to TCP message/UDP messages of same session is processor 1, belongs to the processing of the icmp packet of this session
Device is processor 2.So, TCP message/UDP messages are handled by processor 1, TCP message/UDP messages are sent to server 1,
Icmp packet is handled by processor 2, still, because processor 2 does not have session information or only vicious session information, so
Icmp packet can be dropped or error handle, therefore, can cause business processing mistake.
In order to solve the above problems, in one example, after TCP message/UDP messages are assigned to processor 1, when
Receive after belonging to the icmp packet of same session with the TCP message/UDP messages, in order that the processor of processing icmp packet
It is identical with the processor (being assumed to be processor 1) for handling TCP message/UDP messages, then when handling icmp packet, it is necessary to place
Other processors outside reason device 1, which forbid interrupting, locked etc., to be operated, so that other processors can not handle icmp packet,
So as to avoid the competition between processor.But because other processors are locked, therefore the same time, only one processing
Device can handle message, and the resource of other processors is wasted, and cause that multiple processor parallel processing messages can not be played
Advantage, so as to realize that the height of load-balancing device is handled up, reduce process performance.
For above-mentioned discovery, a kind of processing method of message is proposed in the embodiment of the present application, this method can be applied in SLB
Or on the load-balancing device under LVS networking scenes, and the load-balancing device includes multiple processors, in the present embodiment
Processor can be CPU (Central Processing Unit, central processing unit).
As depicted in figs. 1 and 2, it is the application scenarios schematic diagram of the embodiment of the present application, unlike Fig. 1, wraps in fig. 2
Multiple load-balancing devices are included, by disposing this multiple load-balancing device, can solve the single-point of a load-balancing device
Failure, when a load-balancing device stops service (such as breaking down), it can also continue to carry by other load-balancing devices
For service, so as to ensure that business is not interrupted.
In fig. 1 and 2, multiple servers can be included in server cluster, this multiple server is client jointly
Same application, such as WEB (webpage) applications are provided.The corresponding virtual address of multiple servers in server cluster, and often
Individual server has corresponding true address.In fig. 2, at least two load-balancing devices are included in load balancing cluster, when negative
When load balancing cluster includes two load-balancing devices, then the two load-balancing devices are handled using active-standby mode, i.e.,
For one load-balancing device as main load-balancing device, another load-balancing device is used as standby load-balancing device.When negative
When carrying balancing cluster comprising the load-balancing device of three or more than three, then these load-balancing devices use clustered deploy(ment)
Pattern., can be by it when a load-balancing device stops service (such as breaking down) under above two deployment mode
Its load-balancing device continues offer service, so as to ensure that business is not interrupted.
Wherein, Fig. 1 and Fig. 2 is a kind of typical network topology structure, but is not unique topological structure, for other
Topological structure, processing mode is identical with the processing mode of the embodiment of the present application, will not be repeated here.For in load balancing cluster
Multiple load-balancing devices for, ECMP (Equal Cost Multipath can be used between multiple load-balancing devices
Routing, equal-cost route) load balancing mode processing.
For Fig. 1 and Fig. 2, the handling process of the embodiment of the present application is identical, is subsequently illustrated by taking Fig. 1 as an example.Above-mentioned
Under application scenarios, as shown in figure 3, the processing method of the message may comprise steps of：
Step 301, when receiving message, judge the message destination address whether be this load-balancing device ground
Location.If it is not, then perform step 302；If it is, perform step 305.
In one example, the address in the embodiment of the present application can be network address, that is, IP address, such as IPv4
Location or IPv6 addresses, describe for convenience, address is collectively referred to as in the embodiment of the present application.
In one example, the message that load-balancing device receives, can be the message that user end to server is sent,
It can also be the message that server is sent to client.If the message that user end to server is sent, then perform step
302；If the message that server is sent to client, then step 305 is performed.
Wherein, when client needs to access server, meeting sending destination location is the message of virtual address, therefore, pin
The message sent to user end to server, the destination address of the message are virtual addresses, rather than this load-balancing device
Address, therefore, if the destination address of message is not the address of this load-balancing device, illustrate that the message is client to clothes
The message that business device is sent, performs step 302.
Wherein, for load-balancing device when being E-Packeted to server, the source address of the message is load-balancing device
Address.Therefore, server is to client when sending message, the destination address of the message for load-balancing device address (i.e.
Receive the source address of message).Therefore, if the destination address of message is the address of this load-balancing device, the message is illustrated
It is the message that server is sent to client, performs step 305.
Step 302, it is the message that user end to server is sent to determine the message, and calculates the source address of the message
Corresponding hashed value.In one example, for the process of hashed value corresponding to the source address of the calculating message, in addition to：
Calculate the message source address and destination address corresponding to hashed value.
In the embodiment of the present application, when calculating hashed value, only with reference to the source address of message, or, the source address of message and
Destination address, and ignore the factors such as protocol number, source port mark, destination interface mark.Wherein, when calculating hashed value, for
The message that user end to server is sent, because the source address of message is constant, therefore may be referred to the source address of message, so,
All messages (such as TCP message or UDP messages icmp packet) sent for user end to server, same dissipate can be corresponded to
Train value.Because the source address of message is constant, destination address is constant, therefore may be referred to the source address and destination address of message, this
Sample, all messages sent for user end to server, can correspond to same hashed value.Because different clients are to server
The destination address of transmission message is identical, it is thus impossible to individually refer to the destination address of message.
In one example, the process of hashed value corresponding to the source address of the message is calculated, can be included but is not limited to
Following manner：Hash operation is carried out to the source address using default hash algorithm, obtains hashed value corresponding to the source address.
The default hash algorithm can include：The total quantity of hashed value obtained after Hash operation can be made, with the load balancing
The total quantity identical hash algorithm of multiple processors in equipment.Wherein, because default hash algorithm can make Hash operation
The total quantity of the hashed value obtained afterwards, it is identical with the total quantity of multiple processors on load-balancing device, therefore, each hash
Value can uniquely correspond to a processor, in this manner it is possible to determine the processor corresponding to each hashed value.Moreover, it is directed to
Each hashed value, corresponding only one processor is only can determine whether out, without determining multiple processors simultaneously.
Further, Hash operation is carried out to the source address using default hash algorithm, it is corresponding obtains the source address
Hashed value process, can specifically include but is not limited to following manner：By the source address to the total of the multiple processor
Quantity remainder, and determine that remainder is hashed value corresponding to the source address.Wherein, it is by configuring default hash algorithm：By source
Address can make the total quantity of hashed value obtained after Hash operation, with load-balancing device to the total quantity remainder of processor
On multiple processors total quantity it is identical.It is moreover, source address is simpler to the calculating process of the total quantity remainder of processor
Single, amount of calculation is smaller.
In one example, calculate the message source address and the destination address corresponding to hashed value process, tool
Body can include but is not limited to following manner：Hash is carried out to the source address and the destination address using default hash algorithm
Computing, obtain hashed value corresponding to the source address and the destination address.Wherein, the default hash algorithm can specifically wrap
Include：The total quantity of hashed value obtained after Hash operation can be made, it is total with multiple processors on the load-balancing device
Quantity identical hash algorithm.Wherein, the total quantity of the hashed value obtained due to presetting hash algorithm to make after Hash operation,
Identical with the total quantity of multiple processors on load-balancing device, therefore, each hashed value can uniquely correspond to a processing
Device, in this manner it is possible to determine the processor corresponding to each hashed value.Moreover, be directed to each hashed value, only can determine whether out pair
The only one processor answered, without determining multiple processors simultaneously.
Further, Hash operation is carried out to the source address and the destination address using default hash algorithm, obtained
The process of hashed value, can include but is not limited to corresponding to the source address and the destination address：By the source address and institute
State destination address and be converted to a new numerical value, and the total quantity remainder by the new numerical value to the multiple processor, and determine
Remainder is hashed value corresponding to the source address and the destination address.Wherein, it is by configuring default hash algorithm：By source
Location and destination address are converted to a new numerical value, and the total quantity remainder by new numerical value to processor, after can making Hash operation
The total quantity of obtained hashed value is identical with the total quantity of multiple processors on load-balancing device.By source address and purpose
Address conversion is a new numerical value, and new numerical value is fairly simple to the calculating process of the total quantity remainder of processor, amount of calculation compared with
It is small.
In one example, the multiple processor refers to all processing for data forwarding on load-balancing device
Device.For example, when including 8 processors for being used for data forwarding on load-balancing device, then this 8 processors are compiled
Number, these processors be respectively processor 0, processor 1 ... processors 7.
For " the total quantity remainder by the source address to the multiple processor, and determine that remainder is the source address
The process of corresponding hashed value ", it is assumed that source address 127.118.12.12, then can be directly by 1271181212 pairs of total quantitys 8
Remainder, and determine that remainder (being assumed to be 2) is hashed value corresponding to source address.
For " source address and the destination address are converted into a new numerical value, and by the new numerical value to described
The mistake of the total quantity remainder of multiple processors, and determine remainder hashed value corresponding to the source address and the destination address "
Journey, it is assumed that source address 127.118.12.12, destination address 192.168.0.1, then can by 127.118.12.12 and
192.168.0.1 a new numerical value is converted to, and by the new numerical value to the remainder of total quantity 8, and determine that remainder (being assumed to be 1) is
Hashed value corresponding to source address and destination address.
Wherein, when 127.118.12.12 and 192.168.0.1 are converted into a new numerical value, can be converted into
Numerical value 127118121219216801, either it is converted into numerical value 192168011271181212 or is converted into number
It is worth (1271181212+19216801) etc., the mode for source address and destination address to be converted to new numerical value can be according to reality
Border experience arbitrary disposition, it is without limitation in the embodiment of the present application.
Step 303, the processor corresponding to the hashed value is determined.In one example, determine that the hashed value institute is right
The process for the processor answered, it can specifically include but is not limited to following manner：Dissipated by what hashed value inquiry was pre-configured with
The mapping relations of train value and processor, obtain the processor corresponding to the hashed value.Wherein, due to hashed value total quantity with
The total quantity of multiple processors on load-balancing device is identical, therefore the mapping that can be pre-configured with hashed value and processor is closed
System, so that each hashed value can uniquely correspond to a processor, in this manner it is possible to determine corresponding to each hashed value
Processor.
Wherein it is possible to be pre-configured with hashed value and the mapping relations of processor on load-balancing device, it is used for for example, working as
When the quantity of the processor of data forwarding is 8, then the mapping relations can be as shown in table 1.Based on this mapping relations, obtaining
When hashed value is 1, it is processor 1 to determine processor corresponding to hashed value 1.
Table 1
| Hashed value | Processor |
| 0 | Processor 0 |
| 1 | Processor 1 |
| 2 | Processor 2 |
| 3 | Processor 3 |
| 4 | Processor 4 |
| 5 | Processor 5 |
| 6 | Processor 6 |
| 7 | Processor 7 |
Step 304, message is sent to the processor, so that the processor is handled the message.
For example, when it is determined that the processor corresponding to the hashed value is processor 1, then message (is connect in step 301
The message received) processor 1 is sent to, so that the processor 1 handles the message.
Wherein, the message can include but is not limited to：UDP messages, TCP message, icmp packet.
In one example, the process handled for the processor the message, can include but is not limited to
Following situation：When the message is UDP messages or TCP message, if there is session corresponding to the message, then institute is utilized
State session and send the message；The session if there is no corresponding to the message, then the message is created on the processor
Corresponding session, and send the message using the session created.When the message is icmp packet, if there is the report
Session corresponding to text, then send the message using the session；The session if there is no corresponding to the message, then abandon institute
State message.
In one example, processor is after message (such as UDP messages, TCP message, icmp packet) is received, can be with
Utilize characteristic information (such as source address, destination address, source port mark, destination interface mark, five yuan of the protocol number in the message
Group information) inquire about whether have session corresponding to the message.
In one example, for creating the process of session corresponding to the message on a processor, processor can profit
The session corresponding to the characteristic information in the message establishes the message.
In one example, for the process using the session transmission message, then processor can be described in
Message described in session modification, and amended message is sent to server.
Below in conjunction with specific application scenarios, the processing procedure of above-mentioned message is described in detail.
1st, for first UDP messages/TCP message of same characteristic features information (such as five-tuple information), do not built on processor
Vertical UDP messages/session corresponding to TCP message, thus using UDP messages/TCP message characteristic information inquiry whether have corresponding to
During session, Query Result is without corresponding session, establishes session using the characteristic information of UDP messages/TCP message, the session
It can include but is not limited to：Characteristic information, session state information are (as state, connection wait state, connection closed shape are established in connection
State, receive data mode etc.), real server information (address of such as real server, port, upstate).Wherein, very
Real server is the server that processor is based on that load-balancing algorithm is UDP messages/TCP message selection.Processor utilizes the meeting
Destination address (is such as revised as the address of real server, is present treatment by source address modification by words modification UDP messages/TCP message
The address of device), amended UDP messages/TCP message is sent to server.
2nd, for follow-up UDP messages/TCP message (such as second UDP report of same characteristic features information (such as five-tuple information)
Text/TCP message, the 3rd UDP messages/TCP message etc.), meeting corresponding to UDP messages/TCP message is had built up on processor
Words, therefore, when whether having corresponding session using the characteristic information inquiry in UDP messages/TCP message, Query Result is to have pair
The session answered, processor using session modification UDP messages/TCP message (as by destination address with being revised as real server
Location, by the address that source address modification is present processor), and amended UDP messages/TCP message is sent to server.
Because first UDP messages/TCP message and follow-up UDP messages/TCP message are transmitted based on same session,
Therefore, same server can be sent to by processor, is handled by same server.
3rd, client is on the basis of UDP messages/TCP message is sent, and in actual applications, client is also possible to service
Device sends icmp packet, and such as the icmp packet of Source quench (source is closed) type, therefore, icmp packet is in UDP
Sent after message/TCP message, UDP messages/session corresponding to TCP message is now had been set up on processor.It is based on
This, for icmp packet, has built up session corresponding to icmp packet on processor, therefore, the spy in using icmp packet
When whether sign information inquiry has corresponding session, for Query Result to there is corresponding session, processor utilizes session modification ICMP
Destination address (is such as revised as the address of real server, by the address that source address modification is present processor) by message, and will repair
Icmp packet after changing is sent to server.In addition, processor inquiry whether have corresponding session when, if Query Result is
Without corresponding session, then it is a wrong message to illustrate icmp packet, and processor needs to abandon the icmp packet, no longer sent out
Give the icmp packet.
Step 305, it is the message that server is sent to client to determine message, is inquired about by the destination address of message advance
The address of configuration and the mapping relations of processor, determine the processor corresponding to the destination address of message.
In one example, for each processor on load-balancing device, the processor, which has, is pertaining only to the processing
One or more address (being referred to as local addresss) of device, and be pertaining only to the address of the processor and belong to other processing
The address of device is different.Based on this, reflecting for address and processor can be safeguarded in advance in the local address tables of load-balancing device
Relation is penetrated, is an example of local address tables as shown in table 2.
Table 2
| Address | Processor |
| 192.168.0.100 | Processor 0 |
| 192.168.0.101 | Processor 1 |
| 192.168.0.102 | Processor 2 |
| 192.168.0.103 | Processor 3 |
| 192.168.0.104 | Processor 4 |
| 192.168.0.105 | Processor 5 |
| 192.168.0.106 | Processor 6 |
| 192.168.0.107 | Processor 7 |
In one example, the source address of the message sent for each processor, the processor to server is at this
The address of device is managed, so that the destination address for the message that server returns is the address of the processor.Wherein, in processor to service
When device sends message, by the address by source address modification for present processor, so, server is in returned packet, destination
Location is exactly the address of the processor.So, the message that server returns can be sent to the processor, by the processor to report
Text is handled.The message of server, the message that server returns to client, by same are sent to for client
Processor processor.
For example, when processor 1 sends message to server 1, for above-mentioned " processor utilizes session modification message "
The destination address of message can be revised as the address of the address, i.e. server 1 of real server by process, processor 1, and by message
Source address modification for present processor address, i.e. processor 1 address 192.168.0.101.So, server 1 is to visitor
When family end sends message, the destination address of the message is exactly the address 192.168.0.101 of processor 1.Load-balancing device exists
After receiving the message, processor can be obtained as processing by the destination address 192.168.0.101 inquiry tables 2 of the message
Device 1.
Step 306, message is sent to the processor of determination, so that the processor determined is handled message.
For example, when it is determined that the processor corresponding to the destination address of message is processor 1, then by message (i.e. step 301
In the message that receives) be sent to the processor 1 so that the processor 1 handles the message.
Wherein, the message can include but is not limited to：UDP messages, TCP message, icmp packet.
In one example, the process handled for the processor of determination the message, can include but unlimited
In following situation：If there is session corresponding to the message, then the processor sends the message using the session；If
In the absence of session corresponding to the message, then the processor abandons the message.
In one example, processor is after message (such as UDP messages, TCP message, icmp packet) is received, can be with
Utilize characteristic information (such as source address, destination address, source port mark, destination interface mark, five yuan of the protocol number in the message
Group information) inquire about whether have session corresponding to the message.
In one example, for the process using the session transmission message, then processor can be described in
Message described in session modification, and amended message is sent to client.
Below in conjunction with specific application scenarios, the processing procedure of above-mentioned message is described in detail.
1st, server can return to UDP messages/TCP message, clothes after UDP messages/TCP message is received to client
Business device, which returns to UDP messages/TCP message, to be carried out after client sends UDP messages/TCP message, is had built up on processor
UDP messages/session corresponding to TCP message.Therefore, processor receive server return UDP messages/TCP message when,
When whether having respective session using the characteristic information inquiry of UDP messages/TCP message, Query Result utilizes this to there is respective session
Session modification UDP messages/TCP message, and amended UDP messages/TCP message is sent to client.Whether have in inquiry
During corresponding session, if Query Result is without corresponding session, it is a wrong report to illustrate UDP messages/TCP message
Text, processor need to abandon the UDP messages/TCP message.
2nd, in actual applications, server is also possible to send icmp packet, and the ICMP reports that server is sent to client
Text is carried out after client sends UDP messages/TCP message, is had built up on processor corresponding to UDP messages/TCP message
Session.Therefore processor is in the characteristic information inquiry using icmp packet when receiving the icmp packet of server transmission
Not no when having corresponding session, Query Result utilizes the session modification icmp packet to there is corresponding session, and will be amended
Icmp packet is sent to client.When whether inquiry has corresponding session, if Query Result is without corresponding session,
It is a wrong message to illustrate icmp packet message, and processor needs to abandon the icmp packet.
In one example, server can include but is not limited to the icmp packet that client is sent：Destination interface is not
The ICMP error messages of reachable ICMP error messages, TTL (Time To Live, lifetime value) time-out.Wherein, service is worked as
After the serve port of device is closed, the server will return to destination after the message from client is received to client
The inaccessible ICMP error messages of mouth.When the network where server has loop, then server can be sent to client
The ICMP error messages of TTL time-out.Further, client is after ICMP error messages are received, it is possible to according to difference
Type of error, close connection or prompting user malfunction reason, in order to user investigate problem.
In one example, each message of session modification (such as UDP messages, TCP message, icmp packet) is utilized in processor
When, the source address modification of the message can be the destination address in session by processor, i.e. virtual address, and by the mesh of the message
Address be revised as the address of source address, i.e. client in session.
Obviously, by above-mentioned processing, client is sent to UDP messages/TCP message of server, icmp packet can be sent out
Same processor is given to be handled.Moreover, UDP messages/TCP message for being returned to client of server, icmp packet also can
The processor is sent to be handled.
Based on above-mentioned technical proposal, in the embodiment of the present application, multiple messages for same session can be sent to together
One processor, multiple messages are handled by the processor, so as to avoid business processing mistake.For example, for same
The message 1 and message 2 of session, load-balancing device are supplied to processor 1 after message 1 is received, by message 1, received
After message 2, message 2 is supplied to processor 1, message 1 and message 2 are sent to server 1 by processor 1, so as to avoid industry
Business processing mistake.Moreover, on load-balancing device, only multiple messages of same session can be carried out by a processor
Processing, rather than multiple processors compete and this multiple message are handled simultaneously, avoid the competition between multiple processors,
So as to make full use of substantial amounts of processor, the resource of processor is saved, improves the utilization rate of processor, lifts load balancing
Performance and disposal ability, lift the performance of business.
The process of message is sent to client for server, in the embodiment of the present application, it is also proposed that a kind of processing of message
Method, this method can apply to include on the load-balancing device of multiple processors, as shown in figure 4, the processing side of the message
Method may comprise steps of：
Step 401, when receiving the message that server is sent to client, inquired about by the destination address of message advance
The address of configuration and the mapping relations of processor, determine processor corresponding to the destination address of message.
In one example, for each processor on load-balancing device, the processor, which has, is pertaining only to the processing
One or more address (being referred to as local addresss) of device, and be pertaining only to the address of the processor and belong to other processing
The address of device is different.Based on this, reflecting for address and processor can be safeguarded in advance in the local address tables of load-balancing device
Relation is penetrated, is an example of local address tables as shown in table 2.
In one example, the source address of the message sent for each processor, the processor to server is at this
The address of device is managed, so that the destination address for the message that server returns is the address of the processor.
Step 402, message is sent to the processor, so that the processor is handled message.
Wherein, the message can include but is not limited to：UDP messages, TCP message, icmp packet.
In one example, the process handled for the processor of determination the message, can include but unlimited
In following situation：If there is session corresponding to the message, then the processor sends the message using the session；If
In the absence of session corresponding to the message, then the processor abandons the message.
The detailed process of step 401, step 402, referring to step 305 and step 306, it will not be repeated here.
Conceived based on the application same with the above method, a kind of processing dress of message is additionally provided in the embodiment of the present application
Put, the processing unit of the message can be applied on the load-balancing device including multiple processors.Wherein, the processing of the message
Device can be realized by software, can also be realized by way of hardware or software and hardware combining.Exemplified by implemented in software, make
It is by the processor of the load-balancing device where it, reading non-volatile storage for the device on a logical meaning
In corresponding computer program instructions formed.For hardware view, as shown in figure 5, the place of the message proposed for the application
A kind of hardware structure diagram of load-balancing device where reason device, except the processor shown in Fig. 5, nonvolatile memory
Outside, the load-balancing device can also include other hardware, such as be responsible for forwarding chip, network interface, the internal memory of processing message；
From hardware configuration, the load-balancing device is also possible to be distributed apparatus, may include multiple interface cards, so as to hard
Part aspect carries out the extension of Message processing.
As shown in fig. 6, the structure chart of the processing unit of the message proposed for the application, described device include：Judging unit
11, for when receiving message, judge the message destination address whether be the load-balancing device address；Calculate
Unit 12, for when the judged result of the judging unit 11 is no, it is determined that the message is user end to server hair
The message sent, and calculate hashed value corresponding to the source address of the message；Determining unit 13, for determining that the hashed value institute is right
The processor answered；Transmitting element 14, for the message to be sent into the processor, so that the processor is to the message
Handled.
In one example, the computing unit 12, it is additionally operable to calculating hashed value corresponding to the source address of the message
During, calculate the message source address and destination address corresponding to hashed value.
In one example, the determining unit 13, specifically for it is determined that processor corresponding to the hashed value
During, the hashed value and the mapping relations of processor that are pre-configured with by hashed value inquiry, obtain the hashed value institute
Corresponding processor.
In one example, the determining unit 13, it is additionally operable to when the judged result of the judging unit 11 is is, then
It is the message that server is sent to client to determine the message, and inquires about what is be pre-configured with by the destination address of the message
Address and the mapping relations of processor, determine the processor corresponding to the destination address of the message；The transmitting element 14, also
For the message to be sent to the processor of the determination, so that the processor of the determination is handled the message.
In one example, for each processor, the processor has one or more that be pertaining only to the processor
Address, and the address for being pertaining only to the processor is different from the address for belonging to other processors；
The source address of the message sent for each processor, the processor to server is the address of the processor, with
The destination address for making the message of server return is the address of the processor.
As shown in fig. 7, the structure chart of the processing unit of the message proposed for the application, is applied including multiple processors
On load-balancing device, described device includes：Determining unit 21, in the message for receiving server and being sent to client
When, the mapping relations of the address being pre-configured with and processor are inquired about by the destination address of the message, determine the message
Processor corresponding to destination address；Transmitting element 22, for the message to be sent into the processor, so that the processor
The message is handled.
In one example, for each processor, the processor has one or more that be pertaining only to the processor
Address, and the address for being pertaining only to the processor is different from the address for belonging to other processors；
The source address of the message sent for each processor, the processor to server is the address of the processor, with
The destination address for making the message of server return is the address of the processor.
Through the above description of the embodiments, those skilled in the art can be understood that the application can be by
Software adds the mode of required general hardware platform to realize, naturally it is also possible to which by hardware, but the former is more in many cases
Good embodiment.Based on such understanding, what the technical scheme of the application substantially contributed to prior art in other words
Part can be embodied in the form of software product, and the computer software product is stored in a storage medium, if including
It is dry to instruct to cause a computer equipment (be personal computer, server, or network equipment etc.) to perform this Shen
Method that please be described in each embodiment.It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment,
Unit or flow in accompanying drawing are not necessarily implemented necessary to the application.
It will be appreciated by those skilled in the art that the unit in device in embodiment can describe be divided according to embodiment
It is distributed in the device of embodiment, respective change can also be carried out and be disposed other than in one or more devices of the present embodiment.On
The unit for stating embodiment can be merged into a unit, can also be further split into multiple subelements.Above-mentioned the embodiment of the present application
Sequence number is for illustration only, does not represent the quality of embodiment.
Disclosed above is only several specific embodiments of the application, and still, the application is not limited to this, any ability
What the technical staff in domain can think change should all fall into the protection domain of the application.
Claims (24)
1. a kind of processing method of message, it is characterised in that described applied on the load-balancing device including multiple processors
Method comprises the following steps：
When receiving message, judge the message destination address whether be the load-balancing device address；
If it is not, then determining that the message is the message that user end to server is sent, and calculate the source address pair of the message
The hashed value answered；
Determine the processor corresponding to the hashed value；
The message is sent to the processor, so that the processor is handled the message.
2. according to the method for claim 1, it is characterised in that hashed value corresponding to the source address for calculating the message
Process, in addition to：
Calculate the message source address and destination address corresponding to hashed value.
3. according to the method for claim 2, it is characterised in that
The process of hashed value corresponding to the source address of the message is calculated, is specifically included：Using default hash algorithm to the source
Address carries out Hash operation, obtains hashed value corresponding to the source address；
Calculate the message source address and destination address corresponding to hashed value process, specifically include：Calculated using default Hash
Method carries out Hash operation to the source address and the destination address, obtains dissipating corresponding to the source address and the destination address
Train value；
Wherein, the default hash algorithm includes：The total quantity of hashed value obtained after Hash operation can be made, it is and the multiple
The total quantity identical hash algorithm of processor.
4. according to the method for claim 3, it is characterised in that
It is described that Hash operation is carried out to the source address using default hash algorithm, obtain hashed value corresponding to the source address
Process, specifically include：Total quantity remainder by the source address to the multiple processor, and determine that remainder is the source address
Corresponding hashed value；
It is described that Hash operation is carried out to the source address and the destination address using default hash algorithm, obtain the source address
With the destination address corresponding to hashed value process, specifically include：The source address and the destination address are converted to one
Individual new numerical value, and the total quantity remainder by the new numerical value to the multiple processor, and determine remainder for the source address and
Hashed value corresponding to the destination address.
5. according to the method for claim 1, it is characterised in that the mistake for determining the processor corresponding to the hashed value
Journey, specifically include：
The hashed value and the mapping relations of processor being pre-configured with by hashed value inquiry, are obtained corresponding to the hashed value
Processor.
6. according to the method for claim 1, it is characterised in that when the message be UDP UDP messages or
During person's transmission control protocol TCP message, process that the processor is handled the message specifically includes：
If there is session corresponding to the message, then the message is sent using the session；
The session if there is no corresponding to the message, then session corresponding to the message, and profit are created on the processor
The message is sent with the session of establishment.
7. according to the method for claim 1, it is characterised in that when the message is reported for Internet Control Message agreement ICMP
Wen Shi, the process that the processor is handled the message, is specifically included：
If there is session corresponding to the message, then the message is sent using the session；
The session if there is no corresponding to the message, then abandon the message.
8. according to the method for claim 1, it is characterised in that whether the destination address for judging the message is described
After the address of load-balancing device, methods described also includes：
If it is, determining that the message is the message that server is sent to client, and pass through the destination address of the message
The mapping relations of the address being pre-configured with and processor are inquired about, determine the processor corresponding to the destination address of the message；
The message is sent to the processor of the determination, so that the processor of the determination is handled the message.
9. according to the method for claim 8, it is characterised in that
For each processor, the processor has one or more address for being pertaining only to the processor, and is pertaining only at this
The address for managing device is different from the address for belonging to other processors.
10. according to the method for claim 9, it is characterised in that
The source address of the message sent for each processor, the processor to server is the address of the processor, so that clothes
The destination address for the message that business device returns is the address of the processor.
11. according to the method for claim 8, it is characterised in that the processor of the determination is handled the message
Process, specifically include：
If there is session corresponding to the message, then the message is sent using the session；
The session if there is no corresponding to the message, then abandon the message.
12. according to the method described in claim any one of 8-11, it is characterised in that the message includes：UDP messages or
TCP message or icmp packet.
A kind of 13. processing method of message, it is characterised in that applied on the load-balancing device including multiple processors, institute
The method of stating comprises the following steps：
When receiving the message that server is sent to client, the ground being pre-configured with is inquired about by the destination address of the message
Location and the mapping relations of processor, determine processor corresponding to the destination address of the message；
The message is sent to the processor, so that the processor is handled the message.
14. according to the method for claim 13, it is characterised in that
For each processor, the processor has one or more address for being pertaining only to the processor, and is pertaining only at this
The address for managing device is different from the address for belonging to other processors.
15. according to the method for claim 14, it is characterised in that
The source address of the message sent for each processor, the processor to server is the address of the processor, so that clothes
The destination address for the message that business device returns is the address of the processor.
16. according to the method for claim 13, it is characterised in that the mistake that the processor is handled the message
Journey, specifically include：
If there is session corresponding to the message, then the message is sent using the session；
The session if there is no corresponding to the message, then abandon the message.
17. according to the method described in claim any one of 13-16, it is characterised in that the message includes：User Datagram Protocol
Discuss UDP messages or transmission control protocol TCP message or Internet Control Message agreement icmp packet.
A kind of 18. processing unit of message, it is characterised in that applied on the load-balancing device including multiple processors, institute
Device is stated to specifically include：
Judging unit, for when receiving message, judging whether the destination address of the message is the load-balancing device
Address；
Computing unit, for when the judging unit judged result for it is no when, it is determined that the message is client to service
The message that device is sent, and calculate hashed value corresponding to the source address of the message；
Determining unit, for determining the processor corresponding to the hashed value；
Transmitting element, for the message to be sent into the processor, so that the processor is handled the message.
19. device according to claim 18, it is characterised in that
The computing unit, it is additionally operable to during hashed value corresponding to the source address of the message is calculated, calculate the report
Hashed value corresponding to the source address and destination address of text.
20. device according to claim 18, it is characterised in that the determining unit, specifically for it is determined that described dissipate
The hashed value and the mapping of processor being pre-configured with during processor corresponding to train value by hashed value inquiry are closed
System, obtains the processor corresponding to the hashed value.
21. device according to claim 18, it is characterised in that
The determining unit, it is additionally operable to when the judged result of the judging unit is is, it is determined that the message is server
The message sent to client, and the address being pre-configured with and the mapping pass of processor are inquired about by the destination address of the message
System, determines the processor corresponding to the destination address of the message；
The transmitting element, it is additionally operable to the message being sent to the processor of the determination, so that the processor of the determination
The message is handled.
22. device according to claim 21, it is characterised in that
For each processor, the processor has one or more address for being pertaining only to the processor, and is pertaining only at this
The address for managing device is different from the address for belonging to other processors；
The source address of the message sent for each processor, the processor to server is the address of the processor, so that clothes
The destination address for the message that business device returns is the address of the processor.
A kind of 23. processing unit of message, it is characterised in that applied on the load-balancing device including multiple processors, institute
Device is stated to specifically include：
Determining unit, for when receiving the message that server is sent to client, being looked into by the destination address of the message
The mapping relations of the address being pre-configured with and processor are ask, determine processor corresponding to the destination address of the message；
Transmitting element, for the message to be sent into the processor, so that the processor is handled the message.
24. device according to claim 23, it is characterised in that
For each processor, the processor has one or more address for being pertaining only to the processor, and is pertaining only at this
The address for managing device is different from the address for belonging to other processors；
The source address of the message sent for each processor, the processor to server is the address of the processor, so that clothes
The destination address for the message that business device returns is the address of the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610833366.6A CN107846364A (en) | 2016-09-19 | 2016-09-19 | A kind for the treatment of method and apparatus of message |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610833366.6A CN107846364A (en) | 2016-09-19 | 2016-09-19 | A kind for the treatment of method and apparatus of message |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107846364A true CN107846364A (en) | 2018-03-27 |
Family
ID=61657292
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610833366.6A Pending CN107846364A (en) | 2016-09-19 | 2016-09-19 | A kind for the treatment of method and apparatus of message |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107846364A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111010342A (en) * | 2019-11-21 | 2020-04-14 | 天津卓朗科技发展有限公司 | Distributed load balancing implementation method and device |
| CN111131439A (en) * | 2019-12-20 | 2020-05-08 | 浪潮电子信息产业股份有限公司 | iSCSI-based message transmission method, device, equipment and storage medium |
| CN112104566A (en) * | 2020-09-18 | 2020-12-18 | 网易（杭州）网络有限公司 | Load balancing processing method and device |
| WO2021008591A1 (en) * | 2019-07-16 | 2021-01-21 | 华为技术有限公司 | Data transmission method, device, and system |
| CN112653740A (en) * | 2020-12-11 | 2021-04-13 | 北京金山云网络技术有限公司 | Load balancing method and device supporting QUIC connection migration and computer product |
| CN115086274A (en) * | 2022-06-10 | 2022-09-20 | 北京启明星辰信息安全技术有限公司 | Network flow distribution method, device, equipment and storage medium |
| CN116436855A (en) * | 2023-06-12 | 2023-07-14 | 建信金融科技有限责任公司 | Data information processing method, device, electronic equipment and medium |
| TWI826253B (en) * | 2022-08-24 | 2023-12-11 | 瑞昱半導體股份有限公司 | Network control method and network interface card |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102624611A (en) * | 2011-12-31 | 2012-08-01 | 成都市华为赛门铁克科技有限公司 | Method, device, processor and network equipment for message dispersion |
| CN102811169A (en) * | 2012-07-24 | 2012-12-05 | 成都卫士通信息产业股份有限公司 | Virtual private network (VPN) implementation method and system for performing multi-core parallel processing by using Hash algorithm |
| US20150372917A1 (en) * | 2012-08-17 | 2015-12-24 | International Business Machines Corporation | Load balancing overlay network traffic using a teamed set of network interface cards |
| CN105554065A (en) * | 2015-12-03 | 2016-05-04 | 华为技术有限公司 | Method, conversion unit and application unit for message processing |
-
2016
- 2016-09-19 CN CN201610833366.6A patent/CN107846364A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102624611A (en) * | 2011-12-31 | 2012-08-01 | 成都市华为赛门铁克科技有限公司 | Method, device, processor and network equipment for message dispersion |
| CN102811169A (en) * | 2012-07-24 | 2012-12-05 | 成都卫士通信息产业股份有限公司 | Virtual private network (VPN) implementation method and system for performing multi-core parallel processing by using Hash algorithm |
| US20150372917A1 (en) * | 2012-08-17 | 2015-12-24 | International Business Machines Corporation | Load balancing overlay network traffic using a teamed set of network interface cards |
| CN105554065A (en) * | 2015-12-03 | 2016-05-04 | 华为技术有限公司 | Method, conversion unit and application unit for message processing |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021008591A1 (en) * | 2019-07-16 | 2021-01-21 | 华为技术有限公司 | Data transmission method, device, and system |
| CN111010342A (en) * | 2019-11-21 | 2020-04-14 | 天津卓朗科技发展有限公司 | Distributed load balancing implementation method and device |
| CN111010342B (en) * | 2019-11-21 | 2023-04-07 | 天津卓朗科技发展有限公司 | Distributed load balancing implementation method and device |
| CN111131439A (en) * | 2019-12-20 | 2020-05-08 | 浪潮电子信息产业股份有限公司 | iSCSI-based message transmission method, device, equipment and storage medium |
| CN112104566A (en) * | 2020-09-18 | 2020-12-18 | 网易（杭州）网络有限公司 | Load balancing processing method and device |
| CN112104566B (en) * | 2020-09-18 | 2024-02-27 | 网易（杭州）网络有限公司 | Processing method and device for load balancing |
| CN112653740A (en) * | 2020-12-11 | 2021-04-13 | 北京金山云网络技术有限公司 | Load balancing method and device supporting QUIC connection migration and computer product |
| CN115086274A (en) * | 2022-06-10 | 2022-09-20 | 北京启明星辰信息安全技术有限公司 | Network flow distribution method, device, equipment and storage medium |
| CN115086274B (en) * | 2022-06-10 | 2023-12-22 | 北京启明星辰信息安全技术有限公司 | Network traffic distribution method, device, equipment and storage medium |
| TWI826253B (en) * | 2022-08-24 | 2023-12-11 | 瑞昱半導體股份有限公司 | Network control method and network interface card |
| CN116436855A (en) * | 2023-06-12 | 2023-07-14 | 建信金融科技有限责任公司 | Data information processing method, device, electronic equipment and medium |
| CN116436855B (en) * | 2023-06-12 | 2023-09-12 | 建信金融科技有限责任公司 | Data information processing method, device, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107846364A (en) | A kind for the treatment of method and apparatus of message | |
| US10917351B2 (en) | Reliable load-balancer using segment routing and real-time application monitoring | |
| US11070447B2 (en) | System and method for implementing and managing virtual networks | |
| US9253245B2 (en) | Load balancer and related techniques | |
| CN103795805B (en) | Distributed server load-balancing method based on SDN | |
| KR102138619B1 (en) | Message clustering method and load balancer based on server cluster | |
| US8923296B2 (en) | System and methods for managing network packet forwarding with a controller | |
| CN103155500B (en) | For the method and system of the stateless load balance of network service flow | |
| CN104272708B (en) | It is distributed with the stateless first order grouping to server farm and is distributed to the secondary data packets of the stateful second level grouping distribution of some server in group | |
| US10855682B2 (en) | Virtual address for controller in a controller cluster | |
| CN105577723B (en) | Virtualize the method and apparatus that load balancing is realized in network | |
| JP6364106B2 (en) | Method, system and computer-readable medium for routing Diameter messages in a Diameter signaling router | |
| JP2013510506A (en) | Method and system for using overlay to secure connection over network | |
| CN109525684B (en) | Message forwarding method and device | |
| CN110417924A (en) | Message processing method and distributed apparatus in distributed apparatus | |
| CN101827039B (en) | Method and equipment for load sharing | |
| JP2020510938A (en) | System and method for a compute node management protocol | |
| CN107846433A (en) | A kind of synchronous methods, devices and systems of session information | |
| CN109474713A (en) | Message forwarding method and device | |
| CN110324244A (en) | A kind of method for routing and server based on Linux virtual server | |
| CN107547339A (en) | A kind of gateway media access control MAC address feedback method and device | |
| JP2010003273A (en) | Sip message distribution method and sip message distribution device | |
| US7711780B1 (en) | Method for distributed end-to-end dynamic horizontal scalability | |
| CN115632980A (en) | Method and device for realizing routing configuration, storage medium and electronic equipment | |
| CN117318974A (en) | Non-translated port oversubscription for proxy devices |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180327 |
|
| RJ01 | Rejection of invention patent application after publication |