CN110336755B - Novel ship network congestion control method - Google Patents
Novel ship network congestion control method Download PDFInfo
- Publication number
- CN110336755B CN110336755B CN201910576911.1A CN201910576911A CN110336755B CN 110336755 B CN110336755 B CN 110336755B CN 201910576911 A CN201910576911 A CN 201910576911A CN 110336755 B CN110336755 B CN 110336755B
- Authority
- CN
- China
- Prior art keywords
- network
- state information
- congestion
- data packet
- flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- H04L47/11—Identifying congestion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a novel method for controlling network congestion of a ship, which comprises the following steps: step one, acquiring a network service flow set of a ship network, and when a data packet in a certain service flow enters a local switch, pressing network state information into a state information stack of the data packet through the local switch, and processing and forwarding the data packet; step two, judging whether congestion occurs in the ship network; when the target server receives the data packet with the network state information, storing the data packet into a time sequence database, analyzing the state information of a plurality of continuous data packets, and evaluating whether congestion occurs in the ship network according to the time sequence; thirdly, the controller performs global scheduling to solve congestion; and the controller calculates the congestion degree and performs global scheduling according to the congestion degrees of different data links. The invention can effectively solve the congestion problem, improve the throughput of the whole network and ensure the timeliness of network transmission.
Description
Technical Field
The invention relates to the technical field of ship networks, in particular to a novel ship network congestion control method.
Background
China's ship network technology has undergone three generations of technical development, and the earliest source is the development and construction of a ship-based combat command system. In the early 80 s of the 20 th century, aiming at the defects of a point-to-point communication mode, a 1553B bus is adopted by a ship-based combat command system in China, the rate is 1Mbps, and a baseband transmission and master-slave communication working mode is adopted. In the early stage of 90 s in the 20 th century, in order to improve the communication speed, a commercial 10M Ethernet technology is applied and is adapted to a carrier-based combat system network through transformation. By the beginning of the 21 st century, with the further development of the carrier-based combat command system, the ship network is rapidly transited to the adoption of the switched Ethernet, and the speed is also developed from 10Mbps to 100Mbps/1000Mbps.
With the progress of the Ethernet technology, the network connection range of the ships is greatly expanded; according to the service range and the specific requirements, independent control of a plurality of subnetworks is achieved, flexible management and control evolution of the whole ship network is accelerated along with popularization of a Software Defined Network (SDN) technology, but in the operation process of the networks, due to the characteristics of sudden flow and high frequency in an actual use environment, the problem of congestion is easily caused, the network throughput is reduced due to the congestion problem, and the network environment is deteriorated. Therefore, the method combines the flexible and fine-grained control characteristics of the software defined network and the data plane programmable technology to research the congestion problem in the ship network, and has important theoretical and military application values.
The traditional network congestion processing methods, such as buffer allocation, packet elimination, flow control, packet suppression and the like, have the defects of slow monitoring timeliness, low network efficiency and the like, and influence the transmission of some control instructions needing quick response. In a ship network, the services are various, the burst rate is high, the problem of congestion is more easily caused, and the network environment is worsened. There is therefore a need for a practical and efficient way to control congestion problems.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a novel ship network congestion control method aiming at the defects in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the invention provides a novel ship network congestion control system, which comprises a local switch, a controller and a time sequence database which are arranged in a ship network, collects local network state information of the ship network, generates a control strategy by a congestion control method, and sends a data packet to a destination server; wherein:
the local switch presses the network state information into a state information stack of the data packet and carries the network state information along a data packet forwarding path;
the destination server pops up the state information in the state information stack after receiving the data packet and stores the state information in the time sequence database;
and the controller is provided with an application program for analyzing the network state information on the upper layer, judges whether congestion occurs or not according to analyzing the state information of a plurality of continuous data packets, and generates a new shunting rule and sends the new shunting rule to the data plane through the controller if the congestion occurs.
Further, the network status information of the present invention includes a switch ID and a port queue length.
The invention provides a novel method for controlling network congestion of a ship, which comprises the following steps:
step one, acquiring a network service flow set of a ship network, and when a data packet in a certain service flow enters a local switch, pressing network state information into a state information stack of the data packet through the local switch, and processing and forwarding the data packet;
step two, judging whether congestion occurs in the ship network; when the destination server receives the data packets with the network state information, the data packets are stored in a time sequence database, the state information of a plurality of continuous data packets is analyzed, and whether congestion occurs in the ship network is evaluated according to the time sequence;
thirdly, the controller performs global scheduling to solve congestion; the controller calculates the congestion degree and carries out global scheduling according to the congestion degrees of different data links.
Further, the specific method of the first step of the invention is as follows:
let D be the traffic flow set of the ship network, D = (D) 1 ,d 2 ,…,d n ) Wherein i is more than or equal to 1 and less than or equal to n;
when a certain traffic flow d i After the data packets enter a local switch of a ship and are processed, the global time t of each data packet entering is measured through metadata last_packet And storing the data packet into a memory of the switch as the latest time of the flow reaching the switch, and measuring the global time t of the data packet after the flow enters the ship switch new_packet Calculating the arrival time of the data packet by delta t = t new_packet -t last_packet (ii) a If Δ t is greater than the set time t set Then, a new sub information flow is considered to appear, namely flow _ ID +1; otherwise, the flow _ ID is not changed, and the newly arrived data packet and the previous data packet are in the same flow;
and performing hash calculation on the flow _ ID and the packet quintuple to generate a corresponding path _ ID, and identifying a forwarding path of the flow by taking the path _ ID as a unit so as to forward the flow.
Further, the specific method of the second step of the invention is as follows:
when the destination server receives the service flow d with state forwarding i And after the data packet, popping up the content in the state information stack and storing the content in a time sequence database Influx DB, analyzing according to a time sequence method, and evaluating whether congestion occurs in the network.
Further, the specific method of the third step of the invention is as follows:
for each link, by packet d i State information carried by state forwarding is reconstructed in the global view, and the length of a forwarding queue, namely the congestion degree, of a transmission path corresponding to each pair of ports is reconstructed, wherein the longer the forwarding queue is, the higher the congestion degree is;
the controller optimizes a network transmission path according to the goals of minimizing the service flow and the maximum queue length, and distributes a flow scheduling scheme; when part of links fail, the controller prunes the failed links in the global view, updates the cost of each path, and redistributes the traffic among a plurality of equivalent paths.
The invention has the following beneficial effects: the invention relates to a novel ship network congestion control method which comprises the following steps that (1) a flexible network control form is continued while a ship network overall architecture is constructed based on a software defined network, part of state processing capacity is released from a controller to a data plane, and the state is expressed through the data plane; (2) The fine granularity of the collection of the network state reaches the level of a data packet, the real-time performance is higher, and the current network state can be reflected in real time; (3) The controller is used for carrying out overall regulation and control, so that the congestion problem is solved, the throughput of the whole network is improved, and the timeliness of network transmission is guaranteed.
Drawings
The invention will be further described with reference to the following drawings and examples, in which:
FIG. 1 is a schematic diagram of the general structure of an embodiment of the present invention;
fig. 2 is a flow chart of a method of an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
As shown in fig. 1, which is a schematic diagram of a general network structure, in the ship network congestion control method based on data plane zone state forwarding adopted in the embodiment of the present invention, network state information is locally collected, and a controller excavates, analyzes and generates a policy.
The local switch presses state information (such as switch ID, port queue length and the like) into a state information stack of a data packet, the state information is carried along with a data packet forwarding path, after a destination server receives the data packet, the state information in the state information stack is popped up and stored into a time sequence database, an application program on the upper layer of the controller analyzes the state information, whether congestion occurs or not is judged according to analyzing the state information of a plurality of continuous data packets, and if the congestion occurs, a new shunting rule is generated and issued to a data plane through the controller.
The overall process flow herein is shown in fig. 2. Based on a belt state forwarding technology, firstly, collecting network state information on a data plane; and secondly, popping up and storing the state information collected by the terminal server into a database, and carrying out global flow scheduling by the APP on the upper layer of the controller according to the state information in the database.
The programmable data plane-based stateful forwarding technology can express the state of the ship network data plane, namely, the switch and the server can collect and express the network state, so that the state processing capacity of the data plane is improved, the controller can regulate and control congestion in the network according to the collected state information, and a flow forwarding rule is calculated and generated to be issued.
The invention is to control congestion in the whole network through a control plane. Firstly, collecting the state information of the current network element when a data packet flows through a switch, and forwarding the data packet with the state information, after a destination server receives the data packet, popping up the state information in a state information stack, storing the state information in a time sequence database, analyzing the state information in the database by an application program on the upper layer of a controller, judging whether congestion occurs according to analyzing the state information of a plurality of continuous data packets, and if so, generating a new shunting rule and issuing the new shunting rule to a data plane through the controller.
The novel ship network congestion control method provided by the embodiment of the invention comprises the following steps:
the method comprises the steps of firstly, obtaining and processing network state parameters. Let D be the traffic flow set of the ship network, D = (D) 1 ,d 2 ,...,d n ) Wherein i is more than or equal to 1 and less than or equal to n.
When a certain traffic flow d i After the data packets enter the ship switch for processing, the global time t of each data packet entering is measured through metadata last_packet And storing the data packet into a memory of the switch as the latest time of the flow reaching the switch, and measuring the global time t of the data packet after the flow enters the ship switch new_packet Calculating the time of arrival of the data packet at = t new_packet -t last_packet . If Δ t is greater than t set Then, a new sub information flow is considered to appear, namely flow _ ID +1; otherwise, the flow _ ID is not changed and the newly arrived packet is in the same flow as the previous packet.
The method comprises the steps of performing hash calculation on flow _ ID and a data packet quintuple (a network layer protocol number, a source/destination IP address and a source/destination port number), generating corresponding path _ ID, and identifying a forwarding path of the flow by taking the path _ ID as a unit so as to forward the flow.
And step two, judging whether congestion occurs in the network. When the destination server receives the service flow d with state forwarding i And after the data packet, popping up the content in the state information stack and storing the content in a time sequence database Influx DB, analyzing according to a time sequence method, and evaluating whether congestion occurs in the network.
For a particular flow d i The forwarding path of which passes through a switch s i We can get the switch forwarding d in the Influx DB i Queue length per packet, andforming a time series.
As shown in the above formulas (1) and (2),refers to the instantaneous rate of change of queue length at time t,the future queue length change rate predicted at the time t is referred to, alpha is an exponential weighting factor and represents the weight occupied by the historical information, and 0<α≤1,The number may be positive or negative, with positive numbers indicating the rate of increase and negative numbers indicating the rate of decrease.
As shown in the above-mentioned formula (3),for a switch s i The maximum queue length of the path corresponding to the flow is forwarded, beta is the percentage of the congestion threshold and is 0<Beta is less than or equal to 1, if the formula (3) is satisfied, the switch s is indicated i Congestion occurs.
When the traffic flow in the influx DB is analyzed to find congestion in a flow, the controller carries out global scheduling.
And step three, solving the congestion by global scheduling of the controller. For each link, by i The state information carried by the state forwarding can reconstruct the transmission corresponding to each pair of ports in the global viewThe length of the forwarding queue of the output path, i.e. the congestion level, the longer the forwarding queue, the higher the congestion level, and vice versa.
The controller optimizes a network transmission path according to the goals of minimizing the service flow and the maximum queue length, and distributes a flow scheduling scheme; when part of links fail, the controller prunes the failed links in the global view, updates the cost of each path and redistributes the flow among a plurality of equivalent paths.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (3)
1. A novel ship network congestion control method is characterized by being realized through a novel ship network congestion control system, wherein the system comprises a local switch, a controller and a time sequence database which are arranged in a ship network, collects local network state information of the ship network, generates a control strategy through a congestion control method, and sends a data packet to a target server; wherein:
the local switch presses the network state information into a state information stack of the data packet and carries the network state information along a data packet forwarding path;
the destination server pops up the state information in the state information stack after receiving the data packet and stores the state information in the time sequence database Influx DB;
the controller is provided with an application program for analyzing network state information on the upper layer, analyzes the state information of a plurality of continuous data packets according to the Influx DB of the time sequence database to judge whether congestion occurs, and if so, generates a new distribution rule and sends the new distribution rule to the data plane through the controller;
the method comprises the following steps:
step one, acquiring a network service flow set of a ship network, and when a data packet in a certain service flow enters a local switch, pressing network state information into a state information stack of the data packet through the local switch, and processing and forwarding the data packet;
step two, judging whether congestion occurs in the ship network; when the destination server receives the data packets with the network state information, the data packets are stored in a time sequence database, the state information of a plurality of continuous data packets is analyzed, and whether congestion occurs in the ship network is evaluated according to the time sequence;
step three, when the traffic flow in the influx DB is analyzed and the congestion is found in a flow, the controller carries out global scheduling to solve the congestion; the controller calculates the congestion degree and carries out global scheduling according to the congestion degrees of different data links;
the specific method of the first step comprises the following steps:
let D be the traffic flow set of the ship network, D = (D) 1 ,d 2 ,...,d n ) Wherein i is more than or equal to 1 and less than or equal to n;
when a certain traffic flow d i After the data packets enter a local switch of a ship and are processed, the global time t of each data packet entering is measured through metadata last_packet And storing the data packet into a memory of the switch as the latest time of the flow reaching the switch, and measuring the global time t of the data packet after the flow enters the ship switch new_packet Calculating the arrival time of the data packet by delta t = t new_packet -t last_packet (ii) a If Δ t is greater than the set time t set Then, a new sub information flow is considered to appear, namely flow _ ID +1; otherwise, the flow _ ID is not changed, and the newly arrived data packet and the previous data packet are in the same flow;
performing hash calculation on the flow _ ID and the packet quintuple to generate a corresponding path _ ID, and identifying a forwarding path of the flow by taking the path _ ID as a unit to forward;
the specific method of the second step is as follows:
when the destination server receives the service flow d with state forwarding i After the data packetPopping up the content in the state information stack and storing the content in an Influx DB of a time sequence database, analyzing according to a time sequence method, and evaluating whether congestion occurs in a network;
for a particular flow d i The forwarding path of which passes through a switch s i Obtaining the forwarding d of the switch in the Influx DB i The queue length of each data packet and forming a time sequence;
as shown in the above formulas (1) and (2),refers to the instantaneous rate of change of queue length at time t,the future queue length change rate predicted at the moment t is referred to, alpha is an exponential weighting factor and represents the weight occupied by the historical information, and 0<α≤1,The number can be positive or negative, the positive number represents the increasing rate, and the negative number represents the decreasing rate;
2. The new vessel network congestion control method of claim 1, wherein the network state information comprises switch ID and port queue length.
3. The novel ship network congestion control method according to claim 1, wherein the specific method in step three is:
for each link, by packet d i State information carried by state forwarding is reconstructed in the global view, and the length of a forwarding queue, namely the congestion degree, of a transmission path corresponding to each pair of ports is reconstructed, wherein the longer the forwarding queue is, the higher the congestion degree is;
the controller optimizes a network transmission path according to the goals of minimizing the service flow and the maximum queue length, and distributes a flow scheduling scheme; when part of links fail, the controller prunes the failed links in the global view, updates the cost of each path, and redistributes the traffic among a plurality of equivalent paths.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910576911.1A CN110336755B (en) | 2019-06-28 | 2019-06-28 | Novel ship network congestion control method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910576911.1A CN110336755B (en) | 2019-06-28 | 2019-06-28 | Novel ship network congestion control method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110336755A CN110336755A (en) | 2019-10-15 |
CN110336755B true CN110336755B (en) | 2023-01-03 |
Family
ID=68144591
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910576911.1A Active CN110336755B (en) | 2019-06-28 | 2019-06-28 | Novel ship network congestion control method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110336755B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110972151B (en) * | 2019-12-12 | 2022-08-09 | 鹏城实验室 | Maritime communication multi-level control method, storage medium and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106027412A (en) * | 2016-05-30 | 2016-10-12 | 南京理工大学 | TCP (Transmission Control Protocol) congestion control method based on congestion queue length |
CN106961387A (en) * | 2017-03-30 | 2017-07-18 | 中国科学院信息工程研究所 | A kind of link type DDoS defence methods migrated certainly based on forward-path and system |
CN107528791A (en) * | 2017-10-23 | 2017-12-29 | 北京交通大学 | A kind of jamming control method and equipment |
-
2019
- 2019-06-28 CN CN201910576911.1A patent/CN110336755B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106027412A (en) * | 2016-05-30 | 2016-10-12 | 南京理工大学 | TCP (Transmission Control Protocol) congestion control method based on congestion queue length |
CN106961387A (en) * | 2017-03-30 | 2017-07-18 | 中国科学院信息工程研究所 | A kind of link type DDoS defence methods migrated certainly based on forward-path and system |
CN107528791A (en) * | 2017-10-23 | 2017-12-29 | 北京交通大学 | A kind of jamming control method and equipment |
Also Published As
Publication number | Publication date |
---|---|
CN110336755A (en) | 2019-10-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110932989B (en) | Elephant flow path monitoring and scheduling method based on SDN data center network | |
CN105847151A (en) | Multi-constraint QoS routing strategy design method for software defined network | |
CN110891019B (en) | Data center flow scheduling method based on load balancing | |
US11888744B2 (en) | Spin-leaf network congestion control method, node, system, and storage medium | |
CN112260950B (en) | Cloud storage network multicast stream scheduling method based on service priority differentiation | |
CN112350949B (en) | Rerouting congestion control method and system based on flow scheduling in software defined network | |
CN103532877B (en) | A kind of dispatching method ensureing wireless sensor network information real-time Transmission | |
CN111083733A (en) | Congestion control method and system for wireless sensor network | |
CN106911516B (en) | A kind of software defined network controller portion arranging method | |
CN106209618A (en) | A kind of communication mixed networking method and system improving intelligence adapted electric energy effect | |
CN115277574B (en) | Data center network load balancing method under SDN architecture | |
Wang et al. | An ACO-based elephant and mice flow scheduling system in SDN | |
CN105357124A (en) | MapReduce bandwidth optimization method | |
CN110336755B (en) | Novel ship network congestion control method | |
CN113573419A (en) | Multi-hop network channel access method considering multi-priority service | |
Li et al. | Design of a service caching and task offloading mechanism in smart grid edge network | |
CN114448899A (en) | Method for balancing network load of data center | |
CN114257554A (en) | Scheduling method for improving TSN (traffic service network) BE (BE) stream bandwidth utilization rate | |
CN117240788A (en) | SDN-based data center network congestion control method | |
CN110943927B (en) | Named data network transmission control method, device and equipment based on time delay management | |
CN115604311B (en) | Cloud fusion computing system oriented to service network and self-adaptive routing method | |
CN116567089A (en) | Internet of things interaction system and method based on distributed intra-network collaborative computing | |
CN110891027A (en) | Named data network transmission control method, device and equipment based on queue scheduling | |
Li et al. | PARS-SR: A scalable flow forwarding scheme based on Segment Routing for massive giant connections in 5G networks | |
Pan et al. | Orchestrating probabilistic in-band network telemetry for network monitoring |
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 |