CN105681214A - Large-scale network transmission optimization method and system - Google Patents
Large-scale network transmission optimization method and system Download PDFInfo
- Publication number
- CN105681214A CN105681214A CN201511007678.3A CN201511007678A CN105681214A CN 105681214 A CN105681214 A CN 105681214A CN 201511007678 A CN201511007678 A CN 201511007678A CN 105681214 A CN105681214 A CN 105681214A
- Authority
- CN
- China
- Prior art keywords
- node
- packet
- sent
- destination
- sends
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- 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 invention provides a large-scale network transmission optimization method and system and is characterized by that the method comprises following steps of connecting N nodes in a same network end to end, thus forming an annular structure, wherein each node is connected with adjacent nodes at two sides of the node and is connected with another adjacent nodes of the adjacent nodes. According to the method and the system provided by the invention, the data transmission method in combination of an annular network topology structure and a polling algorithm is provided; the large-scale network transmission is optimized; the problems such as disorder and congestion in data transmission are solved or partially solved; and the method and the system are applicable to data transmission among nodes in a big data environment.
Description
Technical field
The present invention relates to communication technical field, particularly relates to a kind of large scale network transmission optimization method and system.
Background technology
Beowulf (Beowulf) cluster is a kind of high performance parallel computer cluster topology, it is the multiple processor system of the loose coupling built by the computer system of one group of independence, in system, each process is by real-time performance communication, shared drive transmission information, thus realizing Distributed Parallel Computing. The design object of Beowulf cluster is the supercomputer system making the department of computer science of general university and small research mechanism have the ability to set up oneself, per second runs 1,000,000,000 instructions. Beowulf Clustering addresses how the problem by using common computer to reach high-performance, high real-time computing system. In recent years, Beowulf group system develops rapidly, the process performance being primarily due to the workstation system as clustered node is more and more stronger, processor and more efficient multi-CPU machine come into the market in a large number faster, and the introducing of LAN new technique and new agreement, communication between clustered node can obtain higher bandwidth and less delay, and group system is easier to be fused in existing network system than traditional parallel computer. At present, the developing instrument of group system is ripe day by day, and its price is also further cheap and is prone to build. It addition, the extensibility of group system is good, the performance of node is also easily via increasing internal memory or improving processor performance acquisition raising.
Messaging interface (MessagePassingInterface, MPI) is a kind of program message passing interface, it provides the multilingual collection of functions realizing one series interfaces. MPI is message passing mechanism conventional on Beowulf cluster, and concurrent program is frequently used MPI to carry out transmission and the reception of data. MPI standard defines one group of function, makes application program from a MPI process, message can be delivered to another MPI process.
In group system, when using MPI to carry out data transmission between node, generally there is following two transmission means:
Mode one adopts the mode of sequential delivery:
Fig. 1 shows that the 1st takes turns data transmission procedure, and in the process, node 1 to node 7 all sends data to node 0, and node 0 adopts Do statement to receive the data that each node sends successively.
Mode two adopts the data transmission order determined at random:
Fig. 2 shows and a certain takes turns in data transmission procedure, contingent data transmission state, node 0 and node 2 all send data to node 5, node 1 sends data to node 4, node 3 sends data to node 6, and node 4 sends data to node 0, and node 5 sends data to node 7, node 6 sends data to node 1, and node 7 sends data to node 2.
Prior art deficiency is in that:
When the mode adopting sequential delivery carries out the transmission of many-to-one data, under some application scenarios, because packet loss of link there will be the sharply decline of network performance, show as throughput far below link bandwidth, cause the waste of whole network bandwidth resources. And when big data quantity or many-one transmission, very easily there is congestion problems.
When the data transmission order that employing machine is determined carries out data transmission, it needs to be determined that data send sequence list, which not only adds the workload in data transmission, when network size is bigger, this table also can compare huge and complicated, multiple nodes send message to a node simultaneously, if traffic load not being scheduling or dispatching method being unreasonable, the node so receiving data needs frequently to process the request receiving data, and the request sending data cannot be processed, fairness and the problem such as congested occur in a network.
Summary of the invention
In view of the above problems, the present invention proposes a kind of large scale network transmission optimization method and system, to can solve the problem that or part solves the problem such as unordered and congested in data transmission, mainly include two-part content: optimization design of network topology structure and transmission method.
For reaching above-mentioned purpose, the technical scheme is that and be achieved in that:
A kind of large scale network transmission optimization method, it is characterised in that described method includes:
Being joined end to end by N number of node in consolidated network, form circulus, wherein, each node is connected with its both sides adjacent node, and is connected with another adjacent node of described adjacent node,
Described each node, connected two nodes in side send packet respectively wherein.
In the transmission of data packets of described N number of node, also including: N number of node is numbered, the numbering of the destination node that the required packet sent of described each node arrives according to each packet is numbered.
Take turns in data transmission procedure non-final one, each node described is carried out twice sending action, each node packet to sending is ranked up by destination node distance, select packet corresponding to one or two destination node that the packet sent is closest, wherein, when node number is even number, last is taken turns each node and only sends a packet, when node number is odd number, last is taken turns each node and sends two packets.
If N number of node is clockwise or counterclockwise from 0 to N-1 serial number, if numbering respectively i and the j of the destination node of the said two packet that node k sends, k < i≤j, and 0≤k≤N-3, the packet being sent to i-node is dealt into k+1 node or k+2 node by node k, and the packet mailing to j node is sent to k+2 node or k+1 node;
The packet being sent to i-node is dealt into node N-1 or node 0 by node N-2, and the packet mailing to j node is sent to node 0 or node N-1; The packet being sent to i-node is dealt into node 0 or node 1 by node N-1, and the packet mailing to j node is sent to node 1 or node 0.
Described each node also can receive the packet transmitted from adjacent two nodes of opposite side, and described node receives packet according to arriving first the mode first received.
The present invention also provides for a kind of large scale network transmission optimization system, and described system includes:
Described system includes N number of node, and N number of node joins end to end, and forms circulus, and wherein, each node is connected with its both sides adjacent node, and is connected with another adjacent node of described adjacent node.
Described each node includes sending module, and for described each node, connected two nodes in side send packets wherein.
Described sending module also includes,
Sequencing unit: for N number of node is numbered, and according to the numbering of the destination node of each packet, needs are sent packet and be ranked up.
Described sending module is taken turns in transmission of data packets non-final one, for performing twice sending action, the packet that need to send is ranked up by destination node distance, select packet corresponding to one or two destination node that the packet sent is closest, when node number is even number, last sending module taking turns each node only sends a packet, and when node number is odd number, last sending module taking turns each node sends two each and every one packets.
Described system, also include, if N number of node is clockwise or counterclockwise from 0 to N-1 serial number, if numbering respectively i and the j of the destination node of the said two packet that node k sends, k < i≤j, and 0≤k≤N-3, described sending module also includes, for the packet being sent to i-node being dealt into k+1 node or k+2 node, the packet mailing to j node is sent to k+2 node or k+1 node;
The packet being sent to i-node is dealt into node N-1 or node 0 by the sending module of node N-2, and the packet mailing to j node is sent to node 0 or node N-1; The packet being sent to i-node is dealt into node 0 or node 1 by the sending module of node N-1, and the packet mailing to j node is sent to node 1 or node 0.
Described each node also includes receiver module, and the packet of two the nodes transmission that are connected for opposite side, described node receives packet according to arriving first the mode first received.
Have the beneficial effect that:
The present invention proposes a kind of data transmission method ring network topology structure and Round-Robin Algorithm combined and system, and large scale network transmission being optimized, it is possible to solve or part solves the problem such as unordered and congested in data transmission. The present invention can also realize the equilibrium of each node data amount and transmission bandwidth in network; Each node contacts with adjacent 4 nodes, takes into account efficiency of transmission and network robustness; Round-robin transmission, it is not necessary to the state of record data transmission in network, each node can realize simple efficient data according to the order of own data and send.
Accompanying drawing explanation
Specific embodiments of the invention are described below with reference to accompanying drawings, wherein:
Fig. 1 illustrates the schematic diagram of scheme 1 data transmission procedure of prior art;
Fig. 2 illustrates the schematic diagram of scheme 2 data transmission procedure of prior art;
Fig. 3 illustrates inventive network topological structure schematic diagram;
Fig. 4 illustrates the embodiment of the present invention one schematic flow sheet;
Fig. 5 a illustrates the embodiment of the present invention two first round data transmission procedure figure;
Fig. 5 b illustrates that the embodiment of the present invention 2 second takes turns data transmission procedure figure;
Fig. 6 illustrates the structural representation of large scale network transmission optimization system in the embodiment of the present invention;
Detailed description of the invention
In order to make technical scheme and advantage clearly understand, below in conjunction with accompanying drawing, the exemplary embodiment of the present invention is described in more detail, obviously, described embodiment is only a part of embodiment of the present invention, rather than all embodiments is exhaustive. And when not conflicting, the embodiment in this explanation and the feature in embodiment can be combined with each other.
For the deficiencies in the prior art, the embodiment of the present invention proposes a kind of large scale network transmission optimization method and system, is illustrated below.
Fig. 3 illustrates that the present invention optimizes network topology structure, as it can be seen, include:
The connected mode of large scale network interior joint becomes ring type structure, and node connects each other, constitutes the loop closed, each node and its both sides adjacent node, and the next node of adjacent node is associated.
In this network structure, each node is connected with about 4 nodes, there is a path being directly connected between two adjacent nodes; Meanwhile, each node also forms the indirect communication path of left and right adjacent node as via node. Each node is set up with two nodes about and is contacted. In this network structure, data are in a network with set direction flowing.
In Fig. 3, eight nodes are example, 8 nodes in same network join end to end, composition circulus, its interior joint 0 is attached with node 1, node 2 and node 7 node 6 respectively, node 1 is attached with node 2, node 3 and node 0, node 7 respectively, and in turn, node 7 and node 0, node 1 and node 6, node 5 are attached.
Beneficial effect:
According to the Model Design network structure of ring-type, in such a configuration, network node degree more uniformly spreads, and the handling capacity of network entirety will be greatly improved. The network structure of ring-type, also being the basis of efficient transmission method, within such networks, each node is only adjacent the next node of node and adjacent node and sets up contact, data or control command one-way flow in a network, it is to avoid the conflict that transmitted in both directions is brought.
Fig. 4 illustrates embodiment one, large scale network transmission optimization method, as it can be seen, include:
Based on loop network structure, adopting Round-Robin Algorithm (RoundRobinScheduling), packet transmits successively according to node sequence in a network. The direction of data transmission adopts set order and direction, one-way flow.
Step 201, based on network topology structure in above-mentioned network, has N number of node, joins end to end with ring style, N number of node is numbered in described network, from 0 to N-1, this numbering can be clock-wise order numbering, it is also possible to be sequence counter-clockwise numbering; The embodiment of the present invention is described being numbered example clockwise.
Step 202, for each node in N number of node, must send packet to N-1 node, be ranked up according to the packet that beard and hair is sent by the numbering of the destination node of each packet, and namely the numbering of packet is corresponding with destination node numbering;
Step 203, take turns in transmission of data packets non-final one, the node that node k is directly attached thereto to two, the same side sends data (0≤k < N), if numbering respectively i and the j of the destination node of the said two packet that node k sends, k < i≤j, and the packet being sent to i-node is dealt into k+1 node or k+2 node by 0≤k≤N-3, node k, the packet mailing to j node is sent to k+2 node or k+1 node;
This same side is clockwise or counterclockwise, and the present invention is all described for example in a clockwise direction.
Step 204, is often taking turns in transmission of data packets, and node k receives packet from (k-1) and (k-2) the two adjacent node, and reception mode is first received for arriving first;
Step 205, it is assumed that the packet that k node receives, the numbering of its destination node respectively m and n, m≤n, this node first determines whether that whether m and n is equal to k;
Step 206, if it is determined that m=k or n=k, illustrates that packet m or n arrives destination node, and the transmitting procedure of this packet terminates.
Step 207, if it is determined that m is not equal to k or n is not equal to k, packet is put into k node data bag transmit queue, all packets that need to send are ranked up, in next round packet transmission process, node k mails to, by mailing to the packet that destination node is m node, one of them node (such as k+1 node or k+2 node) being attached thereto, and is sent to another one of them being attached thereto node (such as k+2 node or k+1 node) by mailing to the packet that destination node is n node.
In the large scale network of N number of node (N >=2), transmission of data packets has N+1 to take turns.
Wherein, those skilled in the art are often taking turns in transmission of data packets, and it is to adopt counterclockwise clockwise that the direction of transmission packet can adopt, the invention is not limited in this regard.
Wherein polling dispatching (RoundRobinScheduling) algorithm is exactly that namely scheduling performs i=(i+1) modn every time, and selects the i-th station server successively by servers different for request scheduling in the way of poll. The advantage of algorithm is its terseness, and it is without recording the state of current all connections, so it is a kind of stateless scheduling.
The embodiment of the present invention one, for 8 nodes, is described in detail by the embodiment of the present invention two.
In same network, there are 8 nodes, join end to end with ring style, be numbered from 0-7, be numbered according to the packet that institute's beard and hair is sent by the numbering of the destination node of each packet.
Wherein, the packet that node 0 need to send is according to packet 1, packet 2, packet 3 ... packet 7 is numbered, namely packet 1 correspondence mails to the packet of destination node 1, packet 2 correspondence mails to the packet of destination node 2, by that analogy, packet 7 correspondence mails to the packet of destination node 7.
The packet that node 1 need to send is according to packet 2, data 3 ... packet 7, packet 0 are numbered, and packet 2 correspondence mails to the packet of destination node 2, and by that analogy, packet 0 correspondence mails to the packet of destination node 0. ;
By that analogy, the packet that node 6 need to send is according to packet 7, packet 0 ... packet 6 is numbered.
First round data transmission procedure is as shown in Figure 5 a, node 0 sends packet 1 and packet 2 respectively to node 1 and node 2, node 1 sends packet 2 and packet 3 respectively to node 2 and node 3, and by that analogy, node 7 sends packet 0 and packet 1 respectively to node 0 and node 1.
Node 0 is respectively received packet 0 from node 7 and node 6, and its interior joint 0 receives packet according to arriving first the principle first received, if the packet of node 7 arrives first, just first receives, after the packet of node 6 arrives, then receives.
Second takes turns data transmission procedure as shown in Figure 5 b, node 0 sends packet 3 and packet 4 respectively to node 1 and node 2, node 1 sends packet 4 and packet 5 respectively to node 2 and node 3, and by that analogy, node 7 sends packet 2 and packet 3 respectively to node 0 and node 1.
Node 0 is respectively received packet 2 from node 7 and node 6, and its interior joint 0 receives packet according to arriving first the principle first received.
Third round sends process, owing to each node all have received the packet that other nodes are sent and destination node is not this node, such as the packet mailing to node 2 that node 0 have received node 6, node 7 is sent, therefore, packet corresponding to the node destination node that chosen distance is nearest from the packet that need to send is transmitted, namely node 0 sends packet 2 and packet 2 respectively to node 1 and node 2, and node 1 sends packet 3 and packet 3 respectively to node 2 and 3
By that analogy, the packet received is ranked up by each node according to set rule, and the data carrying out next round send, until all data have arrived destination node all, whole data transmission procedure is 9 take turns.
Above-described embodiment has even number node, odd number packet, in the end one take turns in transmission of data packets, only during one packet of transmission, when one of them node is to connected one packet of two node-node transmission, if package number and with connected node numbering consistent, be made directly transmission.
Based on above-described embodiment, when consolidated network has even number node, during odd number packet, in the end one take turns in transmission of data packets, only during one packet of transmission, when one of them node is to connected one packet of two node-node transmission, if package number and with connected node numbering inconsistent, node can transmit packet simultaneously and arrive two nodes being connected respectively, and has connected node to continue downwards to transmit, and wherein destination node can receive data according to arriving first the principle first received.
Based on above-described embodiment, when consolidated network has even number node, during odd number packet, in the end one take turns in transmission of data packets, only during one packet of transmission, when one of them node is to connected one packet of two node-node transmission, if package number and inconsistent with connected node numbering, node can only to the nearest node-node transmission packet of being connected, until packet is sent to destination node.
Embodiment three,
In same network, there are 8 nodes, join end to end with ring style, be numbered from 0-7, be numbered according to the packet that institute's beard and hair is sent by the numbering of the destination node of each packet.
Wherein, the packet that node 0 need to send is according to packet 1, packet 2, packet 3 ... packet 7 is numbered, namely packet 1 correspondence mails to the packet of destination node 1, packet 2 correspondence mails to the packet of destination node 2, by that analogy, packet 7 correspondence mails to the packet of destination node 7.
The packet that node 1 need to send is according to packet 2, data 3 ... packet 7, packet 0 are numbered, and packet 2 correspondence mails to the packet of destination node 2, and by that analogy, packet 0 correspondence mails to the packet of destination node 0.
By that analogy, the packet that node 6 need to send is according to packet 7, packet 0 ... packet 6 is numbered.
Packet 1 is sent to node 1 by first round data transmission procedure node 0, and packet 2 is sent to node 2 by node 1, and by that analogy, packet 0 is sent to node 0 by node 7;
Second takes turns data transmission procedure node 0 is sent to node 1 by packet 2, and packet 3 is sent to node 2 by node 1, and by that analogy, packet 1 is sent to node 0 by node 7;
Packet 1 is sent to node 1 by third round data transmission procedure node 0, and packet 2 is sent to node 2 by node 1, and by that analogy, packet 2 is sent to node 0 by node 7;
By that analogy, in the present embodiment, each node can according to unidirectional data transmission bag clockwise or counterclockwise.
Embodiment four,
In same network, there are 9 nodes, join end to end with ring style, be numbered from 0-8, be numbered according to the packet that institute's beard and hair is sent by the numbering of the destination node of each packet.
Wherein, the packet that node 0 need to send is according to packet 1, packet 2, packet 3 ... packet 8 is numbered, namely packet 1 correspondence mails to the packet of destination node 1, packet 2 correspondence mails to the packet of destination node 2, by that analogy, packet 8 correspondence mails to the packet of destination node 8.
The packet that node 1 need to send is according to packet 2, data 3 ... packet 8, packet 0 are numbered, and packet 2 correspondence mails to the packet of destination node 2, and by that analogy, packet 0 correspondence mails to the packet of destination node 0.
By that analogy, the packet that node 8 need to send is according to packet 9, packet 0 ... packet 7 is numbered.
First round data transmission procedure, node 0 sends packet 1 and packet 2 respectively to node 1 and node 2, node 1 sends packet 2 and packet 3 respectively to node 2 and node 3, and by that analogy, node 8 sends packet 0 and packet 1 respectively to node 0 and node 1.
Node 0 is respectively received packet 0 from node 7 and node 8, and its interior joint 0 receives data according to arriving first the principle first received, if the packet of node 7 arrives first, just first receives, after the packet of node 8 arrives, then receives.
Second takes turns data transmission procedure, node 0 sends packet 3 and packet 4 respectively to node 1 and node 2, node 1 sends packet 4 and packet 5 respectively to node 2 and node 3, and by that analogy, node 8 sends packet 2 and packet 3 respectively to node 0 and node 1.
Node 0 is respectively received packet 2 from node 7 and node 8, and its interior joint 0 receives packet according to arriving first the principle first received.
Having 9 nodes in above-described embodiment, each node has 8 packets, in transmission of data packets, often transmits two packets in wheel transmitting procedure all simultaneously.
It should be noted that sending direction is not defined by the scheme of the embodiment of the present invention, it is possible to carry out data transmission in the direction of the clock and also can carry out data transmission counterclockwise.
Based on same inventive concept, the embodiment of the present invention additionally provides a kind of large scale network transmission optimization system, owing to the principle of these equipment solution problem is similar to a kind of large scale network transmission optimization method, therefore the enforcement of these equipment may refer to the enforcement of method, repeats part and repeats no more.
As shown in Figure 6, the present invention also provides for a kind of large scale network transmission optimization system, and described system includes:
Described system includes N number of node, and N number of node joins end to end, and forms circulus, and wherein, each node is connected with its both sides adjacent node, and is connected with another adjacent node of described adjacent node.
Described each node includes sending module, and for described each node, connected two nodes in side send packets wherein.
Described sending module also includes,
Sequencing unit: for N number of node is numbered, and according to the numbering of the destination node of each packet, needs are sent packet and be ranked up.
Described sending module is taken turns in data transmission procedure non-final one, it is additionally operable to perform twice sending action, the packet that need to send is ranked up by destination node distance, select packet corresponding to one or two destination node that the packet sent is closest, when node number is even number, last sending module taking turns each node only sends a packet, and when node number is odd number, last sending module taking turns each node sends two packets.
Described system, also include, for setting N number of node as clockwise or counterclockwise from 0 to N-1 serial number, if numbering respectively i and the j of the destination node of the said two packet that node k sends, k < i≤j, and 0≤k≤N-3, the packet mailing to j node, for the packet being sent to i-node is dealt into k+1 node or k+2 node, is sent to k+2 node or k+1 node by described sending module;
The packet being sent to i-node is dealt into node N-1 or node 0 by the sending module of node N-2, and the packet mailing to j node is sent to node 0 or node N-1;The packet being sent to i-node is dealt into node 0 or node 1 by the sending module of node N-1, and the packet mailing to j node is sent to node 1 or node 0.
Described each node also includes receiver module, and the packet of two the nodes transmission that are connected for opposite side, described node receives packet according to arriving first the mode first received.
In above-described embodiment, existing Functional Unit device blocks all can be adopted to implement. Such as, processing module can adopt existing data handling components, at least, the location-server adopted just possesses realize this Functional Unit device in existing location technology; As for receiver module, then it is the components and parts that all possess of any one equipment possessing signal transfer functions; Meanwhile, what A, n parameter calculating that processing module carries out, intensity adjustment etc. adopted is all existing technological means, and those skilled in the art design and develop can realize through accordingly.
For convenience of description, each several part of apparatus described above is divided into various module or unit to be respectively described with function. Certainly, the function of each module or unit can be realized in same or multiple softwares or hardware when implementing the present invention.
Those skilled in the art are it should be appreciated that embodiments of the invention can be provided as method, system or computer program. Therefore, the present invention can adopt the form of complete hardware embodiment, complete software implementation or the embodiment in conjunction with software and hardware aspect. And, the present invention can adopt the form at one or more upper computer programs implemented of computer-usable storage medium (including but not limited to disk memory, CD-ROM, optical memory etc.) wherein including computer usable program code.
The present invention is that flow chart and/or block diagram with reference to method according to embodiments of the present invention, equipment (system) and computer program describe. It should be understood that can by the combination of the flow process in each flow process in computer program instructions flowchart and/or block diagram and/or square frame and flow chart and/or block diagram and/or square frame. These computer program instructions can be provided to produce a machine to the processor of general purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device so that the instruction performed by the processor of computer or other programmable data processing device is produced for realizing the device of function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions may be alternatively stored in and can guide in the computer-readable memory that computer or other programmable data processing device work in a specific way, the instruction making to be stored in this computer-readable memory produces to include the manufacture of command device, and this command device realizes the function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, make on computer or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computer or other programmable devices provides for realizing the step of function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
Although preferred embodiments of the present invention have been described, but those skilled in the art are once know basic creative concept, then these embodiments can be made other change and amendment. So, claims are intended to be construed to include preferred embodiment and fall into all changes and the amendment of the scope of the invention.
Claims (12)
1. a large scale network transmission optimization method, it is characterised in that described method includes:
Being joined end to end by N number of node in consolidated network, form circulus, wherein, each node is connected with its both sides adjacent node, and is connected with another adjacent node of described adjacent node.
2. the method for claim 1, it is characterised in that also include: described each node, connected two nodes in side send packet respectively wherein.
3. method as claimed in claim 2, it is characterised in that also include: N number of node is numbered, the numbering of the destination node that the required packet sent of described each node arrives according to each packet is numbered.
4. method as claimed in claim 3, it is characterized in that, take turns in data transmission procedure non-final one, each node described is carried out twice sending action, each node packet to sending is ranked up by destination node distance, selects packet corresponding to one or two destination node that the packet sent is closest;
When node number is even number, last is taken turns each node and only sends a packet, and when node number is odd number, last is taken turns each node and sends two packets.
5. method as claimed in claim 4, it is characterized in that, if N number of node is clockwise or counterclockwise from 0 to N-1 serial number, if numbering respectively i and the j of the destination node of the said two packet that node k sends, k < i≤j, and the packet being sent to i-node is dealt into k+1 node or k+2 node by 0≤k≤N-3, node k, the packet mailing to j node is sent to k+2 node or k+1 node;
The packet being sent to i-node is dealt into node N-1 or node 0 by node N-2, and the packet mailing to j node is sent to node 0 or node N-1; The packet being sent to i-node is dealt into node 0 or node 1 by node N-1, and the packet mailing to j node is sent to node 1 or node 0.
6. method as claimed in claim 3, it is characterised in that described each node also can receive the packet of two the nodes transmission that are connected from opposite side, and described node receives packet according to arriving first the mode first received.
7. a large scale network transmission optimization system, it is characterised in that
Described system includes N number of node, and N number of node joins end to end, and forms circulus, and wherein, each node is connected with its both sides adjacent node, and is connected with another adjacent node of described adjacent node.
8. system as claimed in claim 7, it is characterised in that described each node includes sending module, for described each node, connected two nodes in side send packet respectively wherein.
9. system as claimed in claim 8, it is characterised in that described sending module also includes,
Sequencing unit: for N number of node is numbered, and according to the numbering of the destination node of each packet, needs are sent packet and be ranked up.
10. system as claimed in claim 9, it is characterized in that, described sending module is taken turns in transmission of data packets non-final one, for performing twice sending action, the packet that need to send is ranked up by destination node distance, select packet corresponding to one or two destination node that the packet sent is closest
When node number is even number, last is taken turns each node sending module and only sends a packet, and when node number is odd number, last sending module taking turns each node sends two packets.
11. system as claimed in claim 8, it is characterized in that, if N number of node is serial number clockwise or counterclockwise, if numbering respectively i and the j of the destination node of the said two packet that node k sends, 0≤k < i≤j, and 0≤k≤N 1, described sending module also includes, for the packet being sent to i-node being dealt into k+1 node or k+2 node, the packet mailing to j node is sent to k+2 node or k+1 node;
The packet being sent to i-node is dealt into node N-1 or node 0 by the sending module of node N-2, and the packet mailing to j node is sent to node 0 or node N-1; The packet being sent to i-node is dealt into node 0 or node 1 by the sending module of node N-1, and the packet mailing to j node is sent to node 1 or node 0.
12. system as claimed in claim 6, it is characterised in that described each node also includes receiver module, for receiving the packet of two the nodes transmission that are connected from opposite side, described node receives packet according to arriving first the mode first received.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511007678.3A CN105681214B (en) | 2015-12-28 | 2015-12-28 | A kind of large scale network transmission optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511007678.3A CN105681214B (en) | 2015-12-28 | 2015-12-28 | A kind of large scale network transmission optimization method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105681214A true CN105681214A (en) | 2016-06-15 |
| CN105681214B CN105681214B (en) | 2018-11-27 |
Family
ID=56189669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511007678.3A Active CN105681214B (en) | 2015-12-28 | 2015-12-28 | A kind of large scale network transmission optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105681214B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1738225A (en) * | 2005-09-01 | 2006-02-22 | 上海交通大学 | Ring topology data bus bandwidth reuse method |
| CN102595456A (en) * | 2011-01-06 | 2012-07-18 | 上海戴河易信息科技有限公司 | ZigBee loop sensor network |
| CN102970245A (en) * | 2012-11-21 | 2013-03-13 | 北京奇虎科技有限公司 | Data transmission method and system |
| CN102984084A (en) * | 2012-11-21 | 2013-03-20 | 北京奇虎科技有限公司 | Method and system for transmitting data among nodes in computer clusters |
| CN104378447A (en) * | 2014-12-03 | 2015-02-25 | 深圳市鼎元科技开发有限公司 | Non-migration distributed storage method and non-migration distributed storage system on basis of Hash ring |
| CN104811325A (en) * | 2014-01-24 | 2015-07-29 | 华为技术有限公司 | Cluster node controller monitoring method, related device and controller |
| CN105187308A (en) * | 2015-05-07 | 2015-12-23 | 深圳市迪菲特科技股份有限公司 | Resource node searching method and device |
-
2015
- 2015-12-28 CN CN201511007678.3A patent/CN105681214B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1738225A (en) * | 2005-09-01 | 2006-02-22 | 上海交通大学 | Ring topology data bus bandwidth reuse method |
| CN102595456A (en) * | 2011-01-06 | 2012-07-18 | 上海戴河易信息科技有限公司 | ZigBee loop sensor network |
| CN102970245A (en) * | 2012-11-21 | 2013-03-13 | 北京奇虎科技有限公司 | Data transmission method and system |
| CN102984084A (en) * | 2012-11-21 | 2013-03-20 | 北京奇虎科技有限公司 | Method and system for transmitting data among nodes in computer clusters |
| CN104811325A (en) * | 2014-01-24 | 2015-07-29 | 华为技术有限公司 | Cluster node controller monitoring method, related device and controller |
| CN104378447A (en) * | 2014-12-03 | 2015-02-25 | 深圳市鼎元科技开发有限公司 | Non-migration distributed storage method and non-migration distributed storage system on basis of Hash ring |
| CN105187308A (en) * | 2015-05-07 | 2015-12-23 | 深圳市迪菲特科技股份有限公司 | Resource node searching method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105681214B (en) | 2018-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Uy et al. | A comparison of AMQP and MQTT protocols for Internet of Things | |
| CN107087019B (en) | Task scheduling method and device based on end cloud cooperative computing architecture | |
| CN101828372B (en) | Parallelizing the TCP behavior of a network connection | |
| Nguyen et al. | RTEthernet: Real‐time communication for manufacturing cyberphysical systems | |
| WO2016187813A1 (en) | Data transmission method and device for photoelectric hybrid network | |
| CN107241767A (en) | The method and device that a kind of mobile collaboration is calculated | |
| WO2013102253A1 (en) | System and method for providing p2p based reconfigurable computing and structured data distribution | |
| CN104378308A (en) | Method and device for detecting message sending rate | |
| CN105022717A (en) | Network on chip resource arbitration method and arbitration unit of additional request number priority | |
| CN109547162B (en) | Data communication method based on two sets of one-way boundaries | |
| EP3855690A1 (en) | Network element supporting flexible data reduction operations | |
| CN108471629A (en) | The control method of business service quality, equipment and system in transmission network | |
| CN112019363B (en) | Method, device and system for determining service transmission requirement | |
| CN110928694B (en) | Computer system | |
| CN109379303A (en) | Parallelization processing framework system and method based on improving performance of gigabit Ethernet | |
| CN103384181A (en) | Data packet transmission method and device | |
| CN103748845A (en) | Packet sending and receiving method, device and system | |
| Xing et al. | MPTCP meets big data: Customizing transmission strategy for various data flows | |
| CN105763297A (en) | Cloud computing system-based remote data optimized transmission method and device | |
| Sun et al. | Republic: Data multicast meets hybrid rack-level interconnections in data center | |
| Mamidala et al. | Efficient SMP-aware MPI-level broadcast over InfiniBand's hardware multicast | |
| CN110971533A (en) | Method, server apparatus, client apparatus, and medium for data communication | |
| CN114095901A (en) | Communication data processing method and device | |
| CN103138894A (en) | Method and nodes and data network of data transmission | |
| CN105681214A (en) | Large-scale network transmission optimization method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |