CN109634902A - A kind of data transmission method suitable for PCIE bus, apparatus and system - Google Patents
A kind of data transmission method suitable for PCIE bus, apparatus and system Download PDFInfo
- Publication number
- CN109634902A CN109634902A CN201811557916.1A CN201811557916A CN109634902A CN 109634902 A CN109634902 A CN 109634902A CN 201811557916 A CN201811557916 A CN 201811557916A CN 109634902 A CN109634902 A CN 109634902A
- Authority
- CN
- China
- Prior art keywords
- node
- credit
- data
- mirror image
- bitmap
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4282—Bus transfer protocol, e.g. handshake; Synchronisation on a serial bus, e.g. I2C bus, SPI bus
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/40—Bus networks
- H04L12/40006—Architecture of a communication node
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2213/00—Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F2213/0026—PCI express
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention provides a kind of data transmission method suitable for PCIE bus, apparatus and system, is related to field of computer technology, and the method applied to first node includes: to obtain the credit to match with data to be transmitted from the credit mirror image for the second node being locally stored;Wherein the credit mirror image is used to characterize the credit pool that the second node is directed to the first node, the number of credits of the credit pool is to determine that the credit is used to indicate the memory address information of the second node according to the performance of the first node and second node;The data to be transmitted is sent to the memory headroom of the second node indicated by the credit to match with the data to be transmitted through PCIE bus.Compared with prior art, the coupling between each node is reduced, the distribution router platform for being suitable for more possessing the forward node of multifrequency nature is can be very good, the forwarding performance between high-performance node can be made to perform to maximum.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of data transmission methods suitable for PCIE bus, dress
It sets and system.
Background technique
PCIE bus is input/output (I/O) local bus standard of a new generation, is to replace the revolutionary character of pci bus total
Line architecture, PCIE bus have obtained large-scale use in processor system.Compared to pci bus, PCIE bus is using unique
Dual-channel transmission mode, be similar to full-duplex mode, substantially increase data transmission bauds;The interconnection of point-to-point is used simultaneously
Technology can flexible topological structure by equipment such as interchangers (Switch);In addition, PCIE bus also use it is some in net
Technology used in network communication, for example support a variety of data routing modes, substantially envisage the service occurred in data transmission
Quality (QoS, Quality of Service) problem.
Architecture based on PCIE bus mainly includes root component (RC, Root Complex), interchanger (Switch)
With various terminal equipment (EP, Endpoint).Wherein RC mainly complete memory domain to PCIE bus domain address conversion, RC's
Downstream port can mount interchanger to extend more ports PCIE, can also directly mount EP.Interchanger primarily serves expansion
The effect of exhibition system, it supports a upper port and multiple lower ports, can carry out data exchange between port, pass through interchanger
Bridge joint, PCIE bus can be into downlink spreader.
The existing forwarding scheme based on PCIE is forwarded usually using virtual channel, establishes equity between transmitting-receiving line card
Virtual channel carry out message transmissions, but the line card type class in router is more, and performance difference is larger, using virtual logical
The scheme in road will limit the forward efficiency between high-performance line card, to influence overall forward efficiency.
Summary of the invention
The present invention provides a kind of data transmission method and device suitable for PCIE bus, makes in the prior art for solving
The problem of will limit the forward efficiency between high-performance line card with the scheme of virtual channel.
In a first aspect, the embodiment of the present invention provides a kind of data transmission method suitable for PCIE bus, it is applied to first
In node, which comprises
The credit to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored;The wherein letter
It is used to characterize the credit pool that the second node is directed to the first node with mirror image, the number of credits of the credit pool is basis
What the performance of the first node and second node determined, the credit is used to indicate the memory address letter of the second node
Breath;
The data to be transmitted is sent to indicated by the credit to match with the data to be transmitted through PCIE bus
The memory headroom of the second node.
In the embodiment of the present invention, when first node sends data to be transmitted to second node, the letter from second node is needed
The memory address for being stored in second node is characterized with the matched credit of data to be transmitted, a credit with obtaining in mirror image
Information, first node by be stored in local second node for first node credit pool credit mirror image determine it is to be transmitted
The corresponding memory space of data, and second node is according to first segment for the number of credits in the credit pool of the first node
What the performance of point and second node determined, in the prior art by the virtual channel of fixed performance carry out first node and
Second node carries out transmission data and compares, and reduces the coupling between each node, can be very good to be suitable for possessing a variety of more
The distribution router platform of the forward node of characteristic, can make the forwarding performance between high-performance node perform to maximum.
Further, described that the letter to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored
With preceding, further includes:
The corresponding credit mirror image of credit pool that the second node that second node is sent is directed to the first node is received,
And store the credit mirror image.
In the embodiment of the present invention, the credit mirror image that the second node received is sent is saved, to obtain letter
Used time calls.
Further, the method also includes:
Receive and store the credit bitmap corresponding with the credit mirror image that the second node is sent, wherein the letter
The use state for marking the second node for credit in the credit pool of the first node is used for bitmap;
Correspondingly, the data to be transmitted is sent to the credit to match with the data to be transmitted institute through PCIE bus
After the memory headroom of the second node indicated, further includes:
It updates the credit bitmap of the second node and updated credit bitmap is sent to the second node, it is described
Updated credit bitmap is used to indicate the first node and has sent the data to be transmitted.
In the embodiment of the present invention, by credit bitmap come the financial status in the credit pool of indication node, and by more
Bitmap after new notifies second node, and data to be transmitted has been transmitted, compared with prior art, it is possible to reduce expense mentions
The utilization rate of high bandwidth.
It is further, described that updated credit bitmap is sent to after the second node, further includes:
Receive the credit bitmap that the second node that the second node is sent updates;
The credit bitmap updated according to the second node updates the credit mirror image for being stored in the local second node
And the corresponding credit bitmap of the credit mirror image.
In the embodiment of the present invention, second node needs credit bitmap being sent to first node after updating credit bitmap,
So that first node can be updated according to credit bitmap is stored in local credit mirror image and credit bitmap.
Further, described that the letter to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored
With, comprising:
From the credit mirror image for the second node being locally stored according to data to be transmitted priority orders obtain with it is to be passed
The credit that transmission of data matches.
In the embodiment of the present invention, it is contemplated that the different priorities having in data to be transmitted, it is corresponding according to priority application
Credit, guarantee preferential to send the high data to be transmitted of priority.
Further, it is obtained from the credit mirror image for the second node being locally stored according to the priority orders of data to be transmitted
Before taking the credit to match with data to be transmitted, further includes:
Determine the priority of data to be transmitted and the credit mirror image of the second node can transmission state whether match;Its
In, it is described can transmission state be still not used credit in the credit mirror image according to second node quantity determine.
In the embodiment of the present invention, second node is for the quantity of not used credit in the credit pool of the first node
How much determine whether current first node carries out data transmission, or which data transmitted.
Second aspect, the embodiment of the present invention also provide a kind of data transmission method suitable for PCIE bus, are applied to the
In two nodes, comprising:
The second node is established for described first according to the performance of the performance of first node and the second node
The credit pool of node simultaneously determines credit mirror image, wherein the credit mirror image is for characterizing the second node for described first
The credit pool of node, the credit are used to indicate the memory address information of the second node;
The credit mirror image is sent to the first node, so that the first node stores the credit mirror image, and
The credit to match with data to be transmitted is obtained from the credit mirror image, through PCIE bus by the data to be transmitted be sent to
The memory headroom of the second node indicated by the credit that the data to be transmitted matches.
In the embodiment of the present invention, second node is established and is directed to according to the performance and the performance of oneself of opposite end first node
The credit pool of first node, and characterized using credit mirror image, credit mirror image is sent to first node, so that first node root
It is believed that with mirror image determine need obtain with the matched credit of data to be transmitted, in the prior art pass through fixed performance it is virtual
Channel carries out first node and second node carries out transmission data and compares, and reduces the coupling between each node, can be very
The distribution router platform of the good forward node for being suitable for more possessing multifrequency nature can make turn between high-performance node
Volatility performs to maximum.
Further, described that the second node is established according to the performance of first node and the performance of the second node
Credit pool and the corresponding credit mirror image of the determining credit pool for the first node, comprising:
Credit pool according to the second node for the first node determines credit mirror image and credit bitmap, described
Credit bitmap is in the use state for marking credit in credit pool of the second node for the first node;
It is described that the credit mirror image is sent to the first node, comprising:
The credit mirror image and the credit bitmap are sent to the first node.
In the embodiment of the present invention, using the use state of credit in credit bitmap characterization credit pool, credit position can be passed through
Figure is come pond state of making a credit inquiry.
Further, the described credit mirror image and the credit bitmap are sent to after the first node is also wrapped
It includes:
The updated credit bitmap that the first node is sent is received, the updated credit bitmap is described first
Second node described in node updates determines later for the credit bitmap of the first node;
Determine that the first node has sent data to be transmitted according to the updated credit bitmap.
In the embodiment of the present invention, determine that first node has transmitted data to be transmitted and transferred by credit bitmap
At compared with prior art, it is possible to reduce expense improves the utilization rate of bandwidth.
Further, described to determine that the first node has sent number to be transmitted according to the updated credit bitmap
According to later, further includes:
The data to be transmitted is being determined after exporting in the corresponding memory headroom of its corresponding credit, is updating described the
Two nodes are directed to the credit bitmap of the first node;
Updated credit bitmap is sent to the first node.
In the embodiment of the present invention, second node data to be transmitted from the corresponding memory headroom of its corresponding credit it is defeated
When out, i.e., second node determines that the corresponding memory address of the credit occupied in credit pool can discharge, so second node
Credit bitmap is updated, and updated credit bitmap is sent to first node, so that first node can be according to more
Credit bitmap after new obtains credit.
Further, described that the second node is established according to the performance of first node and the performance of the second node
For the credit pool of the first node, comprising:
The second node needle determined according to the worst node of performance in the first node and the second node
To the number of credits in the credit pool of the first node;
The credit pool that the second node is directed to the first node is established according to the number of credits.
In the embodiment of the present invention, in order to realize reciprocity transmission, the credit in the credit pool between the node of connection is established
Number is determined according to the node of performance difference in two nodes.
The third aspect, the embodiment of the present invention also provide a kind of data transmission device suitable for PCIE bus, are applied to the
In one node, comprising:
Credit acquiring unit matches for obtaining from the credit mirror image for the second node being locally stored with data to be transmitted
Credit;Wherein the credit mirror image is used to characterize the credit pool that the second node is directed to the first node, the credit
The number of credits in pond is determined according to the performance of the first node and second node, and the credit is used to indicate described the
The memory address information of two nodes;
Transmission unit matches for through PCIE bus being sent to the data to be transmitted with the data to be transmitted
The memory headroom of the second node indicated by credit.
Further, described device further includes receiving unit:
It is corresponding for the credit pool of the first node specifically for receiving the second node that second node is sent
Credit mirror image, and store the credit mirror image.
Further, the receiving unit is also used to:
Receive and store the credit bitmap corresponding with the credit mirror image that the second node is sent, wherein the letter
The use state for marking the second node for credit in the credit pool of the first node is used for bitmap;
Described device further include:
Updating unit, for updating the credit bitmap of the second node and being sent to updated credit bitmap described
Second node, the updated credit bitmap are used to indicate the first node and have sent the data to be transmitted.
Further, the receiving unit is also used to:
Receive the credit bitmap that the second node that the second node is sent updates;
The updating unit is also used to:
The credit bitmap updated according to the second node updates the credit mirror image for being stored in the local second node
And the corresponding credit bitmap of the credit mirror image.
Further, the credit acquiring unit is specifically used for:
From the credit mirror image for the second node being locally stored according to data to be transmitted priority orders obtain with it is to be passed
The credit that transmission of data matches.
Further, the credit acquiring unit is specifically used for:
Determine the priority of data to be transmitted and the credit mirror image of the second node can transmission state whether match;Its
In, it is described can transmission state be still not used credit in the credit mirror image according to second node quantity determine.
Fourth aspect, the embodiment of the present invention also provide a kind of data transmission device suitable for PCIE bus, are applied to the
In two nodes, comprising:
Credit pool determination unit, for establishing described according to the performance of first node and the performance of the second node
Two nodes for the first node credit pool and determine credit mirror image, wherein the credit mirror image is for characterizing described the
Two nodes are directed to the credit pool of the first node, and the credit is used to indicate the memory address information of the second node;
Credit mirror image transmission unit, for the credit mirror image to be sent to the first node, so that the first segment
Point stores the credit mirror image, and the credit to match with data to be transmitted is obtained from the credit mirror image, will through PCIE bus
The memory that the data to be transmitted is sent to the second node indicated by the credit to match with the data to be transmitted is empty
Between.
Further, the credit pool determination unit is specifically used for:
Credit pool according to the second node for the first node determines credit mirror image and credit bitmap, described
Credit bitmap is in the use state for marking credit in credit pool of the second node for the first node;
The credit mirror image transmission unit is specifically used for:
The credit mirror image and the credit bitmap are sent to the first node.
Further, described device further include:
Receiving unit, the updated credit bitmap sent for receiving the first node, the updated credit
Bitmap is that the first node updates the second node for determining after the credit bitmap of the first node;According to institute
It states updated credit bitmap and determines that the first node has sent data to be transmitted.
Further, described device further includes updating unit, is specifically used for:
The data to be transmitted is being determined after exporting in the corresponding memory headroom of its corresponding credit, is updating described the
Two nodes are directed to the credit bitmap of the first node;
Updated credit bitmap is sent to the first node.
Further, the credit pool determination unit is specifically used for:
Determine that the second node is directed to according to the worst node of performance in the first node and the second node
Number of credits in the credit pool of the first node;
The credit pool that the second node is directed to the first node is established according to the number of credits.
The embodiment of the present invention also provides a kind of data transmission system suitable for PCIE bus, including as the third aspect is any
The first data transmission device and any second data transmission device of fourth aspect.
The embodiment of the present invention also provides a kind of electronic equipment, comprising:
At least one processor;And
The memory being connect at least one described processor communication;Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes, so that at least one described processor is able to carry out any the method in above-described embodiment.
The embodiment of the present invention also provides a kind of non-transient computer readable storage medium, and the non-transient computer is readable to deposit
Storage media stores computer instruction, and the computer instruction is for making the computer execute any side in above-described embodiment
Method.
In the embodiment of the present invention, when first node sends data to be transmitted to second node, need to be directed to from second node
It is obtained in the credit pool of first node and characterizes one with the matched credit of data to be transmitted, a credit and be stored in second node
Memory address information, first node by be stored in local second node for first node credit pool credit mirror image
Determine that the corresponding memory space of data to be transmitted i.e. first node is to be passed to receive to one memory headroom of second node application
Transmission of data, and second node is according to first node and the second section for the number of credits in the credit pool of the first node
What the performance of point determined, it is carried out with the virtual channel progress first node and second node in the prior art by fixed performance
Transmission data are compared, and the coupling between each node is reduced, and can be very good the forwarding section for being suitable for more possessing multifrequency nature
The distribution router platform of point, can make the forwarding performance between high-performance node perform to maximum.And use credit bitmap
Carry out the financial status in the credit pool of indication node, and second node is notified by updated bitmap, data to be transmitted is
Through being transmitted, compared with prior art, it is possible to reduce expense improves the utilization rate of bandwidth.And in transmission process, according to
The corresponding credit of priority application of data to be transmitted guarantees the high data to be transmitted of preferential transmission priority.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is a kind of distributed routing device schematic diagram provided in an embodiment of the present invention;
Fig. 2 is a kind of transmission side data suitable for PCIE bus applied to first node provided in an embodiment of the present invention
The flow diagram of method;
Fig. 3 is a kind of transmission side data suitable for PCIE bus applied to first node provided in an embodiment of the present invention
The flow diagram of method;
Fig. 4 is a kind of transmission side data suitable for PCIE bus applied to second node provided in an embodiment of the present invention
The flow diagram of method;
Fig. 5 is a kind of transmission side data suitable for PCIE bus applied to second node provided in an embodiment of the present invention
The flow diagram of method;
Fig. 6 is that a kind of data suitable for PCIE bus applied to first node provided in an embodiment of the present invention transmit dress
The structural schematic diagram set;
Fig. 7 is that a kind of data suitable for PCIE bus applied to second node provided in an embodiment of the present invention transmit dress
The structural schematic diagram set;
Fig. 8 is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into
It is described in detail to one step, it is clear that the described embodiments are only some of the embodiments of the present invention, rather than whole implementation
Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts
All other embodiment, shall fall within the protection scope of the present invention.
PCIE (Peripheral Component Interconnect-Express, peripheral components interconnection) equipment
Refer to and connect by PCIE interface with other equipment (such as computer motherboard), the equipment for carrying out data processing, common PCIE is set
Have video card, network interface card, solid state hard disk.PCIE interface belongs to the point-to-point binary channels high bandwidth transmission of high speed serialization, and what is connected sets
Back-up does not share bus bandwidth, main advantage is exactly message transmission rate height with bandwidth chahnel is exclusively enjoyed.
PCIE architecture mainly includes root component (Root Complex, hereinafter referred to as RC), interchanger
(SWITCH, hereinafter referred to as SW) and various terminal equipment (EndPoint, hereinafter referred to as EP), as shown in Figure 1.Wherein
RC mainly completes memory domain to the address conversion in PCIE bus domain, and the downstream port of RC can mount SW and extend more PCIE
Port can also directly mount EP;SW primarily serves the effect of expansion system, it supports a upper port and multiple lower ports,
Data exchange can be carried out between port, is bridged by SW, and PCIE can be into downlink spreader.
In Fig. 1, it is made of in system a RC, multiple PCIE SW and multiple EP.RC is generally by the master in router
Engine undertakes, and PCIE SW is extended PCIE topology, and EP can regard the bracing cable card in distribution router, this hair as
It is referred to as node in bright embodiment.Data interaction between node and node is based on PCIE topology.
In the prior art, data are transmitted between different nodes is carried out by virtual channel, in each of distributed system
The virtual channel of the data transmission of carrying control channel and service channel is established between line card, control channel and service channel is unified
It is carried with virtual channel, but in the prior art in the biggish situation of the performance difference of node, adaptability is poor, often performance
Low node will affect the PCIE forwarding performance of complete machine, cause the PCIE forwarding performance between high-performance node that can not play.
Based on above content, the embodiment of the present invention provides a kind of data transmission system suitable for PCIE bus, such as Fig. 1 institute
Show, any one EP in Fig. 1 is as first node, then the opposite end EP transmitted with the EP is as second node.
In embodiments of the present invention, when establishing connection, each node can be applied for end segment first node and second
The credit pool of point, and apply for the credit pool mirror image that certain memory headroom is directed to this end node for storing opposite end.Work as local terminal
When node and other nodes establish connection, can also it establish with the credit pool of other peer nodes and the other opposite ends of preservation for this
The credit pool mirror image of end node.
In embodiments of the present invention, credit is used to indicate the memory address information of second node, and credit is distributed system
In between each node backboard forwarding data minimum unit, be to be calculated according to the physical address and the base address PCIE of caching
One address parameter.
For example, in embodiments of the present invention, by taking second node as an example, the certain credit pool space of second node application,
It include the subspace for being directed to the first node that second node establishes connection in the credit pool space, the subspace is for filling the
The available credit information of data to be transmitted that one node is sent.
In second node, when second node is after establishing the credit for first node, it is notified that first node, the
One node saves the credit mirror image for the credit pool that second node is established for first node, and similarly, second node also receives the
The credit mirror image for the credit pool that the first node that one node is sent is established for second node.
In embodiments of the present invention, the size of credit pool is variable, and the initial credit quantity in credit pool is also variable
, the number of credits in credit pool is that the node for establishing transmission according to two determines, that is to say, that by taking second node as an example,
, can be different from the size of the sub- credit pool for other nodes for the sub- credit pool A of first node, and in sub- credit pool A
Initial credit quantity it is different from the initial credit quantity in other credit pools.
Optionally, in embodiments of the present invention, the maximum quantity of the open ended credit of the credit pool of second node is basis
What the performance of the third node in distributed routing device determined, third node is the optimal node of performance;Second node is directed to
Number of credits in the credit pool of first node is determined according to the worst node of performance in first node and second node.
That is, in embodiments of the present invention, there are N number of node, the property of each node in distributed routing device
Can be different, performance here is determined according to the processor processing capacity of each node and the thread process ability of each node
's.The credit pool size for choosing the optimal node of performance in N number of node is the benchmark that credit pool is arranged, and N number of node all applies for base
The credit pool of quasi- size.
And after first node and second node establish connection, it is thus necessary to determine that for the quantity of credit in the credit pool of opposite end,
If the performance of first node is higher than second node, the letter of first node and second node is determined according to the performance of second node
With the number of credits in pond.
By taking first node as an example, it is assumed that the performance of first node is A, and the performance of second node is B, and A is greater than B, according to A
It can determine that number of credits is 200kb, can determine that number of credits is 100kb according to the performance of B, when first node and the second section
When point establishes connection, first node is 100kb for the number of credits in the credit pool of second node, and second node is directed to first
Number of credits in the credit pool of node is 100kb.
When first node and third node establish connection, the performance of first node is A, and the performance of third node is A, according to
A can determine that number of credits is 200kb, then first node is 200kb for the number of credits in the credit pool of third node, the
Three nodes are 200kb for the number of credits in the credit pool of first node.
Optionally, in embodiments of the present invention, the letter to match with data to be transmitted is obtained except through credit mirror image
With, can also to second node send credit obtain information, second node according to credit obtain information determine credit, second node
Determining credit is sent to first node.
Optionally, in embodiments of the present invention, credit mirror image can be indicated with different representations, such as matrix,
The representation methods such as bitmap, vector can determine workable financial status in credit pool by the representation method.
Optionally, in embodiments of the present invention, it when second node is after establishing the credit for first node, is notified that
The credit bitmap of first node credit mirror image and the credit mirror image, first node protect credit mirror image and credit bitmap
It deposits.
Optionally, in embodiments of the present invention, bitmap can be used to indicate credit mirror image, 1 table of unappropriated credit
Show, the credit of occupancy is indicated with 0, that is to say, that in bitmap, whether 1 and 0 digital representation credit occupied, 1 and 0
The corresponding memory address information of position characterization credit in bitmap.
When first node needs to send data to be transmitted to second node, first node is from the second node being locally stored
Credit mirror image obtain the credit to match with data to be transmitted;First node sends the data to be transmitted through PCIE bus
To the memory headroom of second node indicated by the credit to match with data to be transmitted.
In embodiments of the present invention, first node obtains and number to be transmitted from the credit mirror image for the second node being locally stored
It is whether determining second node has enough credit and store data to be transmitted according to the credit to match, first node can basis
The data volume of data to be transmitted obtains credit from credit mirror image, for example, first node needs to transmit the to be transmitted of 4 2Kb
Data, a credit is 2KB, then first node applies for a credit first, then sends, is then recycled, then goes to apply
Another credit, then sends, until data to be transmitted is all sent to second node.
Optionally, for the ease of managerial credit pond, the theory of waterline management is introduced, is provided with min, low in credit pool,
Tri- kinds of ranks of high, for different ranks, second node can received data volume it is different, so first node is obtaining phase
When matched credit, it is thus necessary to determine that the current rank of credit pool.
It optionally, in embodiments of the present invention, include protocol massages and data message in data to be transmitted, if first node
After determining the rank of credit pool, determine whether to transmit according to the type of message to be transmitted and rank, when credit pool rank is low
When low waterline, stops sending general data message, send protocol massages;When credit pool rank is lower than min waterline, stop
General data message and protocol massages are sent, general data message and protocol massages are cached;When credit pool rank
When more than or equal to low waterline, general data message and protocol massages are sent.
It optionally, in embodiments of the present invention, include the preferential of data in the data to be transmitted that first node needs to send
Grade, for example, the priority of protocol massages is higher than data message in data to be transmitted, it also may include multiple preferential in data message
Grade successively obtains corresponding credit according to the priority of data to be transmitted, guarantees that the high data of priority are first transmitted.
After first node obtains credit corresponding with data to be transmitted, by data to be transmitted be sent to it is to be transmitted
The memory headroom of the second node indicated by the credit that data match needs to notify second node after transmission
Data to be transmitted is sent,
In the embodiment of the present invention, first node notifies second node by way of updating credit bitmap, i.e., when the second section
Point receives the credit bitmap of the update of first node transmission, it is determined that first node has transferred data to be transmitted.With it is existing
Need to add tail portion message after completing primary output in technology, according to the information of tail portion message, to confirm that the message of front is
No to have completed to compare, since the ratio of data message in the prior art and tail portion message is in 5:1, even lower, this can be led
It causes PCIE bandwidth availability ratio to reduce, is mentioned so can reduce expense using the mode for updating credit bitmap in embodiments of the present invention
High bandwidth utilization.
When second node credit bitmap upon a reception of an updated, then it can determine that data to be transmitted is received complete
Finish, that is to say, that when second node determines that credit bitmap has update, then it can determine that the credit in credit pool is used, and
It can be determined which storage address is occupied.If second node determines credit, bitmap does not update, and continues waiting for.
After second node determines that credit bitmap has update, corresponding credit is taken out from credit pool according to bitmap, each
Corresponding one of credit is by the data message of hardware acceptance, by data message composition chained list output, so that data message can
It is processed.
Optionally, in embodiments of the present invention, after second node is by data message composition chained list output, the credit, that is, quilt
Release, so second node is also required to update credit bitmap, optionally, second node, which updates credit bitmap, be can be by data
It is updated after message composition chained list output, is also possible to second node when setting time reaches, according to the number exported
According to updating credit bitmap.
Optionally, second node is after determining data message output, application and the consistent credit of data message quantity, weight
Newly it is added in the credit pool for first node.
In second node after updating credit bitmap, updated credit bitmap is sent to first node, first node
Update storage the credit bitmap and credit mirror image in local second node.
Based on above content, the embodiment of the present invention provides a kind of data transmission method suitable for PCIE bus, is applied to
In first node, as shown in Figure 2, comprising:
Step 201, the credit to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored;Its
Described in credit mirror image be used for characterize the second node be directed to the first node credit pool, the credit number of the credit pool
Amount is to determine that the credit is used to indicate the interior of the second node according to the performance of the first node and second node
Deposit address information;
Step 202, the data to be transmitted is sent to the credit to match with the data to be transmitted through PCIE bus
The memory headroom of the indicated second node.
Further, described that the letter to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored
With preceding, further includes:
The corresponding credit mirror image of credit pool that the second node that second node is sent is directed to the first node is received,
And store the credit mirror image.
Further, the method also includes:
Receive and store the credit bitmap corresponding with the credit mirror image that the second node is sent, wherein the letter
The use state for marking the second node for credit in the credit pool of the first node is used for bitmap;
Correspondingly, the data to be transmitted is sent to the credit to match with the data to be transmitted institute through PCIE bus
After the memory headroom of the second node indicated, further includes:
It updates the credit bitmap of the second node and updated credit bitmap is sent to the second node, it is described
Updated credit bitmap is used to indicate the first node and has sent the data to be transmitted.
It is further, described that updated credit bitmap is sent to after the second node, further includes:
Receive the credit bitmap that the second node that the second node is sent updates;
The credit bitmap updated according to the second node updates the credit mirror image for being stored in the local second node
And the corresponding credit bitmap of the credit mirror image.
Further, described that the letter to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored
With, comprising:
From the credit mirror image for the second node being locally stored according to data to be transmitted priority orders obtain with it is to be passed
The credit that transmission of data matches.
Further, it is obtained from the credit mirror image for the second node being locally stored according to the priority orders of data to be transmitted
Before taking the credit to match with data to be transmitted, further includes:
Determine the priority of data to be transmitted and the credit mirror image of the second node can transmission state whether match;Its
In, it is described can transmission state be still not used credit in the credit mirror image according to second node quantity determine.
One of embodiment is applied to the data suitable for PCIE bus of first node to facilitate the understanding of the present invention
Transmission method, at this for example, the optimal node of performance is third node, corresponding credit pool in distributed routing device
Size is 200kb, and first node application credit pool, the size of the credit pool is 200kb, second node application credit pool, the letter
It is 200kb with the size in pond, the performance of first node is A, and the performance of second node is B, and A is greater than B, initial phase, root
It can determine that number of credits is 100kb according to the performance of B, first node is for the number of credits in the credit pool of second node
100kb, second node are 100kb for the number of credits in the credit pool of first node.When first node is needed to the second section
Point transmission 6kb data to be transmitted, and each credit be 2kb when, i.e. first node is needed to 3 credits of second node application,
Then detailed process is as shown in Figure 3, comprising:
Step 301, first data is sent to second node to first credit of second node application by first node;
Step 302, second data is sent to second node to second credit of second node application by first node;
Step 303, third data are sent to second node to second node application third credit by first node;
Step 304, first node updates the credit bitmap of second node, by the corresponding bitmap point in the credit position of occupancy
Value becomes 0;
Step 305, updated bitmap is sent to second node by first node.
Based on above content, the embodiment of the present invention provides a kind of data transmission method suitable for PCIE bus, is applied to
In second node, as shown in Figure 4, comprising:
Step 401, the second node is established according to the performance of the performance of first node and the second node to be directed to
The credit pool of the first node simultaneously determines credit mirror image, wherein the credit mirror image is directed to for characterizing the second node
The credit pool of the first node, the credit are used to indicate the memory address information of the second node;
Step 402, the credit mirror image is sent to the first node, so that the first node stores the credit
Mirror image, and the credit to match with data to be transmitted is obtained from the credit mirror image, through PCIE bus by the data to be transmitted
It is sent to the memory headroom of the second node indicated by the credit to match with the data to be transmitted.
Further, described that the second node is established according to the performance of first node and the performance of the second node
Credit pool and the corresponding credit mirror image of the determining credit pool for the first node, comprising:
Credit pool according to the second node for the first node determines credit mirror image and credit bitmap, described
Credit bitmap is in the use state for marking credit in credit pool of the second node for the first node;
It is described that the credit mirror image is sent to the first node, comprising:
The credit mirror image and the credit bitmap are sent to the first node.
Further, the described credit mirror image and the credit bitmap are sent to after the first node is also wrapped
It includes:
The updated credit bitmap that the first node is sent is received, the updated credit bitmap is described first
Second node described in node updates determines later for the credit bitmap of the first node;
Determine that the first node has sent data to be transmitted according to the updated credit bitmap.
Further, described to determine that the first node has sent number to be transmitted according to the updated credit bitmap
According to later, further includes:
The data to be transmitted is being determined after exporting in the corresponding memory headroom of its corresponding credit, is updating described the
Two nodes are directed to the credit bitmap of the first node;
Updated credit bitmap is sent to the first node.
Further, described that the second node is established according to the performance of first node and the performance of the second node
For the credit pool of the first node, comprising:
The second node needle determined according to the worst node of performance in the first node and the second node
To the number of credits in the credit pool of the first node;
The credit pool that the second node is directed to the first node is established according to the number of credits.
One of embodiment is applied to the data suitable for PCIE bus of second node to facilitate the understanding of the present invention
Transmission method, at this for example, the optimal node of performance is third node, corresponding credit pool in distributed routing device
Size is 200kb, and first node application credit pool, the size of the credit pool is 200kb, second node application credit pool, the letter
It is 200kb with the size in pond, the performance of first node is A, and the performance of second node is B, and A is greater than B, initial phase, root
It can determine that number of credits is 100kb according to the performance of B, first node is for the number of credits in the credit pool of second node
100kb, second node are 100kb for the number of credits in the credit pool of first node.When first node is needed to the second section
Point transmission 6kb data to be transmitted, and each credit be 2kb when, i.e. first node is needed to 3 credits of second node application,
Second node is as shown in Figure 5 in transmission process, comprising:
Step 501, second node establishes the task for receiving scheduling;
Step 502, second node determines whether credit bitmap has update, if so, thening follow the steps 503, otherwise executes step
Rapid 501;
Step 503, second node takes out corresponding credit according to credit bitmap from credit pool, and each credit is one corresponding
By the data message of hardware acceptance;
Step 504, second node exports the data message received composition chained list;
Step 505, second node updates credit bitmap;
Step 506, updated credit bitmap is sent to first node by second node.
Based on same design, the embodiment of the present invention also provides a kind of data transmission device suitable for PCIE bus, answers
For in first node, as shown in Figure 6, comprising:
Credit acquiring unit 601, for being obtained and data to be transmitted phase from the credit mirror image for the second node being locally stored
Matched credit;Wherein the credit mirror image is used to characterize the credit pool that the second node is directed to the first node, described
The number of credits of credit pool is to determine that the credit is used to indicate institute according to the performance of the first node and second node
State the memory address information of second node;
Transmission unit 602, for being sent to the data to be transmitted and the data to be transmitted phase through PCIE bus
The memory headroom of the second node indicated by the credit matched.
Further, described device further includes receiving unit 603:
It is corresponding for the credit pool of the first node specifically for receiving the second node that second node is sent
Credit mirror image, and store the credit mirror image.
Further, the receiving unit 603 is also used to:
Receive and store the credit bitmap corresponding with the credit mirror image that the second node is sent, wherein the letter
The use state for marking the second node for credit in the credit pool of the first node is used for bitmap;
Described device further include:
Updating unit 604, for updating the credit bitmap of the second node and being sent to updated credit bitmap
The second node, the updated credit bitmap are used to indicate the first node and have sent the data to be transmitted.
Further, the receiving unit 603 is also used to:
Receive the credit bitmap that the second node that the second node is sent updates;
The updating unit 604 is also used to:
The credit bitmap updated according to the second node updates the credit mirror image for being stored in the local second node
And the corresponding credit bitmap of the credit mirror image.
Further, the credit acquiring unit 601 is specifically used for:
From the credit mirror image for the second node being locally stored according to data to be transmitted priority orders obtain with it is to be passed
The credit that transmission of data matches.
Further, the credit acquiring unit 601 is specifically used for:
Determine the priority of data to be transmitted and the credit mirror image of the second node can transmission state whether match;Its
In, it is described can transmission state be still not used credit in the credit mirror image according to second node quantity determine.
The embodiment of the present invention also provides a kind of data transmission device suitable for PCIE bus, is applied in second node,
As shown in fig. 7, comprises:
Credit pool determination unit 701, for establishing institute according to the performance of first node and the performance of the second node
Second node is stated for the credit pool of the first node and determines credit mirror image, wherein the credit mirror image is for characterizing institute
The credit pool that second node is directed to the first node is stated, the credit is used to indicate the memory address letter of the second node
Breath;
Credit mirror image transmission unit 702, for the credit mirror image to be sent to the first node, so that described first
Node stores the credit mirror image, and the credit to match with data to be transmitted is obtained from the credit mirror image, through PCIE bus
The data to be transmitted is sent to the memory of the second node indicated by the credit to match with the data to be transmitted
Space.
Further, the credit pool determination unit 701 is specifically used for:
Credit pool according to the second node for the first node determines credit mirror image and credit bitmap, described
Credit bitmap is in the use state for marking credit in credit pool of the second node for the first node;
The credit mirror image transmission unit 702 is specifically used for:
The credit mirror image and the credit bitmap are sent to the first node.
Further, described device further include:
Receiving unit 703, the updated credit bitmap sent for receiving the first node, the updated letter
It is that the first node updates the second node for determining after the credit bitmap of the first node with bitmap;According to
The updated credit bitmap determines that the first node has sent data to be transmitted.
Further, described device further includes updating unit 704, is specifically used for:
The data to be transmitted is being determined after exporting in the corresponding memory headroom of its corresponding credit, is updating described the
Two nodes are directed to the credit bitmap of the first node;
Updated credit bitmap is sent to the first node.
Further, the credit pool determination unit 701 is specifically used for:
Determine that the second node is directed to according to the worst node of performance in the first node and the second node
Number of credits in the credit pool of the first node;
The credit pool that the second node is directed to the first node is established according to the number of credits.
The embodiment of the invention provides a kind of electronic equipment, including at least one processing unit and at least one storage list
Member, wherein the storage unit is stored with computer program, when described program is executed by the processing unit, so that described
Processing unit executes the step of data transmission method.As shown in figure 8, for the hardware of electronic equipment described in the embodiment of the present invention
Structural schematic diagram, the electronic equipment are specifically as follows desktop computer, portable computer, smart phone, tablet computer etc..Tool
Body, which may include memory 801, the computer program of processor 802 and storage on a memory, the place
The step of reason device 802 realizes any data transmission method in above-described embodiment when executing described program.Wherein, memory 801
It may include read-only memory (ROM) and random access memory (RAM), and provide in memory 801 and store to processor 802
Program instruction and data.
In Fig. 8 for being connected by bus.Program instruction that processor 802 calls memory 801 to store and according to obtaining
The program instruction obtained executes data transmission method provided by the above embodiment.
The embodiment of the present application provides a kind of computer program product, and the computer program product is non-temporary including being stored in
Calculation procedure on state computer readable storage medium, the computer program include program instruction, when described program instructs quilt
When computer executes, the computer is made to execute any of the above-described data transmission method.
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.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (25)
1. a kind of data transmission method suitable for PCIE bus is applied in first node characterized by comprising
The credit to match with data to be transmitted is obtained from the credit mirror image for the second node being locally stored;The wherein credit mirror
As the credit pool for being directed to the first node for characterizing the second node, the number of credits of the credit pool is according to
What the performance of first node and second node determined, the credit is used to indicate the memory address information of the second node;
Described in the data to be transmitted is sent to indicated by the credit to match with the data to be transmitted through PCIE bus
The memory headroom of second node.
2. the method according to claim 1, wherein described obtain from the credit mirror image for the second node being locally stored
Before taking the credit to match with data to be transmitted, further includes:
It receives the second node that second node is sent and is directed to the corresponding credit mirror image of credit pool of the first node, and deposit
Store up the credit mirror image.
3. according to the method described in claim 2, it is characterized in that, the method also includes:
Receive and store the credit bitmap corresponding with the credit mirror image that the second node is sent, wherein the credit position
Figure is for marking the second node for the use state of credit in the credit pool of the first node;
Correspondingly, the data to be transmitted is sent to indicated by the credit to match with the data to be transmitted through PCIE bus
The second node memory headroom after, further includes:
It updates the credit bitmap of the second node and updated credit bitmap is sent to the second node, the update
Credit bitmap afterwards is used to indicate the first node and has sent the data to be transmitted.
4. according to the method described in claim 3, it is characterized in that, described be sent to described second for updated credit bitmap
After node, further includes:
Receive the credit bitmap that the second node that the second node is sent updates;
According to the second node update credit bitmap update be stored in the local second node credit mirror image and
The corresponding credit bitmap of the credit mirror image.
5. method according to any one of claims 1 to 4, which is characterized in that from the credit mirror for the second node being locally stored
As obtaining the credit to match with data to be transmitted, comprising:
It is obtained and number to be transmitted from the credit mirror image for the second node being locally stored according to the priority orders of data to be transmitted
According to the credit to match.
6. according to the method described in claim 5, it is characterized in that, from the credit mirror image for the second node being locally stored according to
The priority orders of data to be transmitted obtain before the credit to match with data to be transmitted, further includes:
Determine the priority of data to be transmitted and the credit mirror image of the second node can transmission state whether match;Wherein,
It is described can transmission state be still not used credit in the credit mirror image according to second node quantity determine.
7. a kind of data transmission method suitable for PCIE bus is applied in second node characterized by comprising
The second node is established for the first node according to the performance of the performance of first node and the second node
Credit pool and determine credit mirror image, wherein the credit mirror image is for characterizing the second node for the first node
Credit pool, the credit is used to indicate the memory address information of the second node;
The credit mirror image is sent to the first node, so that the first node stores the credit mirror image, and from institute
State credit mirror image and obtain the credit to match with data to be transmitted, through PCIE bus by the data to be transmitted be sent to it is described
The memory headroom of the second node indicated by the credit that data to be transmitted matches.
8. the method according to the description of claim 7 is characterized in that the performance according to first node and second section
The performance of point establishes the second node for the credit pool of the first node and determines the corresponding credit mirror of the credit pool
Picture, comprising:
Credit pool according to the second node for the first node determines credit mirror image and credit bitmap, the credit
Bitmap is in the use state for marking credit in credit pool of the second node for the first node;
It is described that the credit mirror image is sent to the first node, comprising:
The credit mirror image and the credit bitmap are sent to the first node.
9. according to the method described in claim 8, it is characterized in that, described send out the credit mirror image and the credit bitmap
After giving the first node further include:
The updated credit bitmap that the first node is sent is received, the updated credit bitmap is the first node
Update what the second node determined later for the credit bitmap of the first node;
Determine that the first node has sent data to be transmitted according to the updated credit bitmap.
10. according to the method described in claim 9, it is characterized in that, described determine institute according to the updated credit bitmap
It states after first node sent data to be transmitted, further includes:
Determining that the data to be transmitted after exporting in the corresponding memory headroom of its corresponding credit, updated second section
Point is directed to the credit bitmap of the first node;
Updated credit bitmap is sent to the first node.
11. according to any method of claim 7-10, which is characterized in that the performance and institute according to first node
The performance for stating second node establishes the credit pool that the second node is directed to the first node, comprising:
Institute is directed to according to the second node that the worst node of performance in the first node and the second node determines
State the number of credits in the credit pool of first node;
The credit pool that the second node is directed to the first node is established according to the number of credits.
12. a kind of data transmission device suitable for PCIE bus is applied in first node characterized by comprising
Credit acquiring unit, for obtaining the letter to match with data to be transmitted from the credit mirror image for the second node being locally stored
With;Wherein the credit mirror image is used to characterize the credit pool that the second node is directed to the first node, the credit pool
Number of credits is to determine that the credit is used to indicate second section according to the performance of the first node and second node
The memory address information of point;
Transmission unit, for the data to be transmitted to be sent to the credit to match with the data to be transmitted through PCIE bus
The memory headroom of the indicated second node.
13. device according to claim 12, which is characterized in that described device further includes receiving unit:
The corresponding credit of credit pool for being directed to the first node specifically for receiving the second node that second node is sent
Mirror image, and store the credit mirror image.
14. device according to claim 13, which is characterized in that the receiving unit is also used to:
Receive and store the credit bitmap corresponding with the credit mirror image that the second node is sent, wherein the credit position
Figure is for marking the second node for the use state of credit in the credit pool of the first node;
Described device further include:
Updating unit, for updating the credit bitmap of the second node and updated credit bitmap being sent to described second
Node, the updated credit bitmap are used to indicate the first node and have sent the data to be transmitted.
15. device according to claim 14, which is characterized in that the receiving unit is also used to:
Receive the credit bitmap that the second node that the second node is sent updates;
The updating unit is also used to:
According to the second node update credit bitmap update be stored in the local second node credit mirror image and
The corresponding credit bitmap of the credit mirror image.
16. any device of 2-15 according to claim 1, which is characterized in that the credit acquiring unit is specifically used for:
It is obtained and number to be transmitted from the credit mirror image for the second node being locally stored according to the priority orders of data to be transmitted
According to the credit to match.
17. device according to claim 16, which is characterized in that the credit acquiring unit is specifically used for:
Determine the priority of data to be transmitted and the credit mirror image of the second node can transmission state whether match;Wherein,
It is described can transmission state be still not used credit in the credit mirror image according to second node quantity determine.
18. a kind of data transmission device suitable for PCIE bus is applied in second node characterized by comprising
Credit pool determination unit, for establishing second section according to the performance of first node and the performance of the second node
Credit pool and determining credit mirror image of the point for the first node, wherein the credit mirror image is for characterizing second section
Point is directed to the credit pool of the first node, and the credit is used to indicate the memory address information of the second node;
Credit mirror image transmission unit, for the credit mirror image to be sent to the first node, so that the first node is deposited
The credit mirror image is stored up, and obtains the credit to match with data to be transmitted from the credit mirror image, it will be described through PCIE bus
Data to be transmitted is sent to the memory headroom of the second node indicated by the credit to match with the data to be transmitted.
19. device according to claim 18, which is characterized in that the credit pool determination unit is specifically used for:
Credit pool according to the second node for the first node determines credit mirror image and credit bitmap, the credit
Bitmap is in the use state for marking credit in credit pool of the second node for the first node;
The credit mirror image transmission unit is specifically used for:
The credit mirror image and the credit bitmap are sent to the first node.
20. device according to claim 19, which is characterized in that described device further include:
Receiving unit, the updated credit bitmap sent for receiving the first node, the updated credit bitmap
It is that the first node updates the second node for determining after the credit bitmap of the first node;According to it is described more
Credit bitmap after new determines that the first node has sent data to be transmitted.
21. device according to claim 20, which is characterized in that described device further includes updating unit, is specifically used for:
Determining that the data to be transmitted after exporting in the corresponding memory headroom of its corresponding credit, updated second section
Point is directed to the credit bitmap of the first node;
Updated credit bitmap is sent to the first node.
22. any device of 8-21 according to claim 1, which is characterized in that the credit pool determination unit is specifically used for:
Determine the second node for described according to the worst node of performance in the first node and the second node
Number of credits in the credit pool of first node;
The credit pool that the second node is directed to the first node is established according to the number of credits.
23. a kind of data transmission system suitable for PCIE bus characterized by comprising such as any institute of claim 12-17
The first data transmission device and the second data transmission device as described in claim 18-22 is any stated.
24. a kind of terminal device, which is characterized in that including at least one processing unit and at least one storage unit,
In, the storage unit is stored with computer program, when described program is executed by the processing unit, so that the processing is single
First perform claim requires the step of 1~6 any claim the method or perform claim to require 7~11 any claims
The step of the method.
25. a kind of computer-readable medium, which is characterized in that it is stored with the computer program that can be executed by terminal device, when
When described program is run on the terminal device, so that the terminal device perform claim requires side described in 1~6 any claim
The step of the step of method or perform claim require 7~11 any claim the method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811557916.1A CN109634902B (en) | 2018-12-19 | 2018-12-19 | Data transmission method, device and system suitable for PCIE bus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811557916.1A CN109634902B (en) | 2018-12-19 | 2018-12-19 | Data transmission method, device and system suitable for PCIE bus |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109634902A true CN109634902A (en) | 2019-04-16 |
CN109634902B CN109634902B (en) | 2020-08-25 |
Family
ID=66075465
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811557916.1A Active CN109634902B (en) | 2018-12-19 | 2018-12-19 | Data transmission method, device and system suitable for PCIE bus |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109634902B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112468416A (en) * | 2020-10-23 | 2021-03-09 | 曙光网络科技有限公司 | Network flow mirroring method and device, computer equipment and storage medium |
CN113315709A (en) * | 2021-03-22 | 2021-08-27 | 阿里巴巴新加坡控股有限公司 | Address cache creating method, routing and addressing method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102263698A (en) * | 2011-08-11 | 2011-11-30 | 福建星网锐捷网络有限公司 | Method for establishing virtual channel, method of data transmission and line card |
CN102523168A (en) * | 2011-12-23 | 2012-06-27 | 福建星网锐捷网络有限公司 | Method and apparatus for message transmission |
CN103618618A (en) * | 2013-11-13 | 2014-03-05 | 福建星网锐捷网络有限公司 | Line card fault recovery method and related device based on distributed PCIE system |
CN104394100A (en) * | 2014-11-07 | 2015-03-04 | 深圳市国微电子有限公司 | Credit allocation method and switch |
CN104767606A (en) * | 2015-03-19 | 2015-07-08 | 华为技术有限公司 | Data synchronization device and method |
-
2018
- 2018-12-19 CN CN201811557916.1A patent/CN109634902B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102263698A (en) * | 2011-08-11 | 2011-11-30 | 福建星网锐捷网络有限公司 | Method for establishing virtual channel, method of data transmission and line card |
CN102523168A (en) * | 2011-12-23 | 2012-06-27 | 福建星网锐捷网络有限公司 | Method and apparatus for message transmission |
CN103618618A (en) * | 2013-11-13 | 2014-03-05 | 福建星网锐捷网络有限公司 | Line card fault recovery method and related device based on distributed PCIE system |
CN104394100A (en) * | 2014-11-07 | 2015-03-04 | 深圳市国微电子有限公司 | Credit allocation method and switch |
CN104767606A (en) * | 2015-03-19 | 2015-07-08 | 华为技术有限公司 | Data synchronization device and method |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112468416A (en) * | 2020-10-23 | 2021-03-09 | 曙光网络科技有限公司 | Network flow mirroring method and device, computer equipment and storage medium |
CN112468416B (en) * | 2020-10-23 | 2022-08-30 | 曙光网络科技有限公司 | Network flow mirroring method and device, computer equipment and storage medium |
CN113315709A (en) * | 2021-03-22 | 2021-08-27 | 阿里巴巴新加坡控股有限公司 | Address cache creating method, routing and addressing method and device |
Also Published As
Publication number | Publication date |
---|---|
CN109634902B (en) | 2020-08-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107347205B (en) | A kind of network slice selection method, apparatus and system | |
CN113473500A (en) | Time delay sensitive network communication method and device thereof | |
CN108667638A (en) | A kind of network service configuration method and network management device | |
US11770327B2 (en) | Data distribution method, data aggregation method, and related apparatuses | |
JP2019500821A (en) | Packet forwarding | |
CN109634902A (en) | A kind of data transmission method suitable for PCIE bus, apparatus and system | |
CN111371694B (en) | Shunting method, device and system, processing equipment and storage medium | |
CN109194578B (en) | Method and device for opening private line service | |
CN108173782A (en) | The method, apparatus and storage medium of transmitting data stream in virtual private cloud | |
CN106130925A (en) | Link scheduling method, equipment and the system of a kind of SDN | |
CN109257282A (en) | A kind of data transmission method and device | |
CN104618157A (en) | Network management method, device and system | |
AU2024201415A1 (en) | Device, method, and system that virtualize a network | |
CN108738078A (en) | A kind of transmission side data subtraction unit, equipment and readable storage medium storing program for executing | |
CN105763483B (en) | A kind of file transmitting method and device | |
CN107733809A (en) | A kind of VSM system messages forwarding port method to set up and device | |
CN106294191B (en) | The method for handling table, the method and apparatus for accessing table | |
CN107689923A (en) | Message processing method and router | |
DE102018006687A1 (en) | TECHNIQUES FOR SELECTING NON-MINIMUM WAYS AND THREADING THE CONNECTION SPEEDS TO INCREASE THE THROUGHPUT IN A NETWORK | |
CN105704037B (en) | A kind of list item store method and controller | |
CN110190988B (en) | Service deployment method and device | |
CN108235158B (en) | Method and device for processing multicast service of optical network | |
JP2022503762A (en) | Cellular telecommunications network | |
CN112714146A (en) | Resource scheduling method, device, equipment and computer readable storage medium | |
CN107888398B (en) | A kind of service bearer system and control 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 |