CN111343090A - Rule pre-deployment-based software defined network efficient routing method - Google Patents
Rule pre-deployment-based software defined network efficient routing method Download PDFInfo
- Publication number
- CN111343090A CN111343090A CN202010128144.0A CN202010128144A CN111343090A CN 111343090 A CN111343090 A CN 111343090A CN 202010128144 A CN202010128144 A CN 202010128144A CN 111343090 A CN111343090 A CN 111343090A
- Authority
- CN
- China
- Prior art keywords
- flow
- switch
- network
- rule
- controller
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/74—Address processing for routing
- H04L45/745—Address table lookup; Address filtering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/25—Routing or path finding in a switch fabric
Abstract
The invention discloses a rule pre-deployment-based software defined network efficient routing method. Firstly, predicting data flow which may pass through a certain switch in the next period of time according to the state information of the current network, and configuring and deploying a corresponding wildcard rule on the switch to ensure that the correctly predicted data flow can be matched with a forwarding rule on the switch; and when the data stream arrives at the next period, judging whether the wildcard rule can be matched or not, when the access switch can not be matched with the wildcard rule, passively reporting to the controller, calculating and issuing an accurate matching rule according to the state of the whole network, and finishing the forwarding of the data stream. By combining the deployment of the active wildcard rule and the passive exact matching rule, the invention reduces the table resource consumption of the data layer, can reduce the interaction between the switch and the controller and reduces the load and the response time of the controller.
Description
Technical Field
The invention belongs to the technical field of Software Defined Network routing, and particularly relates to a Network efficient routing method based on rule pre-deployment in a Software Defined Network (SDN).
Background
With the development of information technology, traffic on both the cloud network and the internet has increased dramatically. On the one hand, with the development of information technology, many new network-based applications (such as search and content distribution) are emerging. Therefore, the network should be able to accommodate the large data flow of these applications. On the other hand, some hot events may also attract attention of many people through the network. For example, many viewers will watch some attractive sports playground live over the internet. In recent years, SDN has become a potential technology to better manage large numbers of flows. SDNs are typically divided into control planes and data planes. The control plane consists of a logically centralized controller, which may be a set of distributed controllers, responsible for managing the entire network. The data plane consists of a set of SDN switches. As more data flows arrive at the network, the switch will send more packets to the controller, which will result in higher controller load and response time.
There are three existing schemes for reducing controller load and response time:
(1) multiple controller schemes with static controller allocation mechanism: the control plane is implemented as a cluster of distributed controllers, with each switch connected/associated with only one controller. The switch transmits the data packets to the different controllers via messages, which helps to reduce the controller load compared to a single controller framework. However, since the traffic in the network fluctuates dynamically in space and time, some controllers may still be in a highly loaded state, even congested. In fact, once the SDN is deployed, its infrastructure is hardly changed in practice. At the same time, more controllers require a large amount of communication overhead in the control plane to maintain network consistency and increase management complexity.
(2) Dynamic controller allocation mechanism: the switch connected to the controller with the higher load is disconnected and reconnected to the controller with the lower load, thereby reducing the load on the controller. However, the dynamic controller allocation scheme has a disadvantage in service continuity. In fact, traffic dynamics often trigger switches to change their controller assignments, which may disrupt service continuity.
(3) Static routing scheme: forwarding rules are deployed in the switches in advance for all data flows in the network. Since each flow may match at least one pre-deployed rule upon arrival at the switch, the switch will not send any Packet-In message packets to the controller. Thus, the controller load is very low and the response time is also short. However, static routing cannot effectively handle dynamic changes in traffic and management policies. First, due to the existence of dynamic traffic intensity, the controller is unable to provide dynamic path control for the traffic flow, which may cause instantaneous congestion on some data links, resulting in packet loss and reduced throughput. Second, for many practical applications, network operators want to specify fine-grained policies to drive how the underlying switches forward, drop, and measure traffic. However, as the management policy of flows may change over time (e.g., due to host mobility or middleware placement), static routing schemes do not adapt well to these changes.
Disclosure of Invention
Aiming at the existing technical problems, the invention provides a rule pre-deployment-based software defined network efficient routing method, which comprises the steps of firstly actively configuring and deploying wildcard rules, and then passively routing data streams which cannot be matched with the wildcard rules dynamically, effectively using limited data layer resources and control layer processing capacity, and avoiding the problems of high table item consumption, high control load, response time and the like caused by routing in a software defined network.
The technical scheme of the invention is as follows:
a high-efficiency routing method of a software defined network based on rule pre-deployment comprises the following steps:
s01: predicting and predicting data streams which are likely to arrive at the next time period of each switch according to the state information of the current network, and configuring and deploying corresponding wildcard rules on all switches, wherein the wildcard rules specify macro streams, and each wildcard can be matched with the streams which pass through the switches and have the same characteristics;
s02: when the data flow of the next time period reaches the switch, firstly, the wildcard rules which are configured and deployed in the switch in advance are matched, and if the data flow is matched, the data flow directly carries out forwarding routing according to the wildcard rules; if not, the exchanger packs Packet-In information, and reports the data flow information to the control layer, and the control layer calculates and issues an accurate matching rule according to the state of the whole network, thereby realizing dynamic routing and completing the forwarding of the data flow.
In a preferred technical solution, the set of macro flows divided by each switch in the step S01 in units of wildcards satisfies two constraints: 1) each flow passing through the switch belongs to a set of flows; 2) the flow sets corresponding to any two wildcards do not intersect.
In a preferred embodiment, the step S01 further includes periodically collecting information of each data stream in the network through a sampling packet, and predicting the information of the macro stream according to the information of each data stream.
In the preferred technical scheme, the controller calculates the flow size of the macro flow according to the information of the network flow statistics to obtain the number of single flows in each macro flow.
In a preferred technical scheme, the switch selects a forwarding route according to a matching result of a data header, and updates statistical information when a data stream passes through the switch.
In a preferred embodiment, after the data flow selects the subset of the macro flows according to the wildcard rule in step S02, the static routing entry without traffic is deleted by prediction, and then the controller installs the updated flow entry to each switch and sets the idle timeout and the hard timeout to zero.
In a preferred embodiment, a boolean variable is used in step S01
Indicating whether the controller has deployed a wildcard rule on path p for the macro-flow t,
indicating that the controller is to deploy wildcard rules for the macro-flow t on path p;
indicating that the macro flow t is not configured to deploy wildcard rules on path p.
In a preferred technical solution, in step S01, the time that the data flow passes through in the network is divided into fixed time periods, at the beginning of each time period, the control layer sends a Read-State request to all SDN switches of all data layers through a network protocol to collect flow information, after receiving the Read-State request, the SDN switches pack the data flow information, generate a response packet, and send the response packet to the controller, and after collecting the responses of all SDN switches, the controller assembles the data flow information of the whole network to obtain the State information of the current network.
Compared with the prior art, the invention has the advantages that:
(1) flow table resource consumption is low: compared with the existing fine-grained routing scheme, the method can greatly reduce the consumption of the table resource in the data layer switch by actively deploying the wildcard rule for the macro flow.
(2) The control layer has low load and short response time: the method stores the forwarding rule of the macro flow into the corresponding switch table entry in advance by deploying the wildcard, thereby greatly reducing the load of information interaction between a control layer and a data layer when the data flow arrives and reducing the response time of the controller.
(3) The expandability is strong: the method realizes hybrid routing by combining active wildcard rule deployment and passive exact matching rule deployment, reduces resource consumption of a control layer and a data layer, and accordingly can optimize the network by utilizing different targets, such as maximizing network throughput rate.
Drawings
The invention is further described with reference to the following figures and examples:
FIG. 1 is a schematic diagram of a rule pre-deployment based software defined networking architecture of the present invention;
FIG. 2 is a workflow diagram of an active wildcard rule installation of the framework of the present invention;
FIG. 3 is a workflow diagram of passive exact match rule installation of the framework of the present invention;
fig. 4 is a network topology diagram of the present embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
Example (b):
the preferred embodiments of the present invention will be further described with reference to the accompanying drawings.
As shown in fig. 1, a software-defined network efficient routing method based on rule pre-deployment designs a new network architecture, and a hybrid routing architecture combining static routing and dynamic routing. The framework comprises an active wildcard rule installation module and a passive exact matching rule installation module. Firstly, predicting data flow which may pass through a certain switch in the next period of time according to the state information of the current network, and actively configuring and deploying a corresponding wildcard rule on the switch to ensure that the correctly predicted data flow can be matched with a forwarding rule on the switch; and when the data stream arrives at the next period, judging whether the wildcard rule can be matched or not, when the access switch can not be matched with the wildcard rule, passively reporting the data stream to the controller by using Packet-In information, and calculating and issuing a proper accurate matching rule by the controller according to the state of the whole network, thereby realizing dynamic routing and finishing the forwarding of the data stream. The invention realizes the hybrid routing in the software defined network by combining the deployment of the active wildcard rule and the passive exact matching rule, reduces the consumption of table resource in the data layer switch, can reduce the interaction between the switch and the controller and reduces the load and the response time of the controller. The invention has good expansibility and wide application prospect.
A high-efficiency routing method of a software defined network based on rule pre-deployment comprises the following steps:
s01: predicting and predicting data streams which are likely to arrive at the next time period of each switch according to the state information of the current network, and configuring and deploying corresponding wildcard rules on all switches, wherein the wildcard rules specify macro streams, and each wildcard can be matched with the streams which pass through the switches and have the same characteristics;
s02: when the data flow of the next time period reaches the switch, firstly, the wildcard rules which are configured and deployed in the switch in advance are matched, and if the data flow is matched, the data flow directly carries out forwarding routing according to the wildcard rules; if not, the exchanger packs Packet-In information, and reports the data flow information to the control layer, and the control layer calculates and issues an accurate matching rule according to the state of the whole network, thereby realizing dynamic routing and completing the forwarding of the data flow.
As shown in fig. 2, firstly, dividing the time that data flows in a network pass through into fixed time periods (for example, a data center is 1 minute or a backbone network is 5 minutes), at the beginning of each time period, a control layer sends Read-State collected traffic information to all SDN switches of all data layers through a network protocol, and after receiving a Read-State request, the SDN switches package the data flow information to generate a response packet and send the response packet to a controller. And after collecting the responses of all the SDN switches, the controller assembles data flow information of the whole network to obtain the state information of the current network.
Wildcards specify a coarser stream granularity, i.e., a macro-stream. Each wildcard may match flows having the same characteristics through the switch, such as matching all flows having the same source switch and destination switch addresses, so that one wildcard rule in an SDN switch may match forwarding multiple data flows.
And periodically collecting information of each data stream in the network through a sampling packet, and predicting the information of the macro stream according to the information of each data stream. Since the size of the flow table in the switch is limited, it is impractical to measure information of each flow only through the flow table. Flow-based measurements (e.g., NetFlow and sFlow) can provide generic support for different measurement tasks by sampling packets and overcome the problem of insufficient switch flow entries.
And selecting a forwarding route by the SDN switch according to a matching result of a data header (packet header) by adopting a software defined network of an OpenFlow protocol. And measures data flow information as the flow passes through the switch. The Read-State request sent by the controller belongs to Multipart Messages defined in OpenFlow.
And the controller calculates the flow of the macro flow according to the information of the network flow statistics to obtain the number of the single flow in each macro flow. Due to prediction error and traffic uncertainty, we should maintain a certain slack controller capacity for unpredictable data flows in the network to avoid long controller response times.
After the data flow selects a subset of the macro flows according to the wildcard rule, i.e., static routing, the static routing entries with no traffic are first deleted by prediction, then the controller installs the updated flow entries to each switch and sets the idle timeout and hard timeout to zero.
The set of macro-flows divided by each switch in wildcard units satisfies two constraints: 1) each flow passing through the switch belongs to a set of flows; 2) the flow sets corresponding to any two wildcards do not intersect.
Details of the active wildcard rule and passive exact match rule installation are as follows. The controller calculates a feasible path set for each macro flow, and saves flow table entry resources by installing wildcard rules in the network. We use Boolean variables
To indicate whether the controller is actively deploying wildcard rules on the path p for the macro-flow t.
Indicating that the controller is to actively deploy wildcard rules for the macro-flow t on the path p;
indicating that the macro flow t is not configured to deploy wildcard rules on path p. After the flow reaches the entrance switch, the flow is passively reported to the controller through Packet-In information to realize dynamic routing, and the controller installs a corresponding accurate matching rule on the path of the flow to realize the routing of the flow.
As shown in fig. 3, when a data flow reaches a switch, a flow table is queried, wildcard rule matching configured and deployed in the switch in advance is performed first, and if the data flow is matched with the wildcard rule, the data flow is directly forwarded and routed according to the wildcard rule; if not, the exchanger packs Packet-In information, and reports the data flow information to the control layer, and the control layer calculates and issues an accurate matching rule according to the state of the whole network, thereby realizing dynamic routing and completing the forwarding of the data flow.
As shown in fig. 4, the network topology of this embodiment is composed of 7 switches supporting the OpenFlow protocol, 6 terminal devices, and one SDN controller (e.g., RYU). Each switch is connected to the controller, which is not shown for simplicity. The RYU controller runs on a server with a kernel of i7-8700k and a memory of 32 GB. To verify the feasibility and efficiency of hybrid routing, 300 data flows were generated in the network and were subject to 2-8 distribution, i.e., containing 20% large flows and 80% small flows.
The workflow in fig. 2 and 3 is illustrated with partial flows in the network. Assuming that there is a partial stream from end device h1- > h2, the specific workflow is as follows:
(1) assuming that forwarding rules have already been deployed on the path v1- > v6- > v5- > v4 of the current time period h1- > h2, when the data flow reaches the ingress switch v1, the forwarding rules can be matched and routed to the terminal h 2. Meanwhile, the controller requests the switch to collect information in the network, such as flow information, link information and the like, through protocol messages, and the switch packs the information and then sends the information to the controller;
(2) after receiving the information sent by the switch, the controller predicts the data flow of the next time period of the network by combining the state of the network in the current time period. Assuming that the controller predicts that the data flow of the slave terminal h1- > h3 exists in the next time period, actively issuing a wildcard rule on the path switch v1- > v6- > v5 through an OpenFlow protocol message, and completing routing in the network by matching the wildcard rule for all the macro flows of the slave terminal h1- > h3 in the next time period;
(3) suppose that the controller predicts that there are more data flows from terminals h1- > h2 in the network in the next period of time, and that congestion may occur in the current network because of too many data flows. The controller recalculates a path, i.e., v1- > v2- > v3- > v4, based on the network information. And the switches are deployed with wildcard rules with higher priority, and when the macro flow arrives at the next time period, the routing can be carried out according to a new path, thereby avoiding the further network congestion and effectively improving the network performance;
(4) when the flow of the next time period arrives, rule matching in the switch is firstly carried out, and a part of data flow can be matched with the wildcard rule predicted and deployed in the last time period. And a part of the streams fail to match the wildcard rule due to prediction error, for example, the data streams from terminals h1- > h6 exist in the next time period, and the exact matching rule needs to be calculated passively. And packing the information of the data Packet by using the Packet-In message and sending the information to the controller, and sending the flow table entry to the corresponding switch by the controller after calculation according to the network state to realize dynamic routing.
In order to illustrate the efficiency of the framework of the present invention, this embodiment uses different routing schemes for comparison with the network topology shown in fig. 4, and the specific implementation steps are as follows:
(1) if all the implementation in the network deploys the exact matching table entries in advance, the routing is static. Flow entries are deployed in advance for all data flows in the network, e.g., traffic of h1- > h4 is forwarded to the destination by paths v1-v6-v5 through exact match entries. The scheme adopts the accurate matching table items, can realize the routing of all data flows in the network, but because the method adopts the static routing, the static routing can not be adjusted according to the link state in the network, and the network congestion condition can occur, thereby causing the link utilization rate in the network to be higher;
(2) if all the passive deployment accurate matching table items are implemented in the network; i.e. dynamic routing. Unlike static routing, flow entries are not deployed ahead of time before the data flow reaches the ingress switch. All data flows pack flow information at an entrance switch through a Ppeak-In message and send the flow information to a controller, and the controller calculates and sends a flow table to a path switch. The method can enable the network to have better performance (such as link utilization rate), 3 different paths of the data flow h1- > h2 can be selected, and the dynamic routing can select the optimal path for routing according to the network state. However, in the method, all data streams are reported to the controller, so that interaction between a data layer and a control layer is increased, a large load is brought to the controller, and the response time of the controller is increased;
(3) the invention effectively reduces the interaction between the data layer and the control layer by actively installing the wildcard rule for the macro flow. For example, a wildcard rule is deployed for the macro flow h2- > h6 on the path v4- > v3- > v7, a wildcard rule is deployed for the macro flow h2- > h5 on the path v4- > v5- > v6, and so on. In addition, the invention also passively installs the accurate matching rule for the data stream which can not be matched with the wildcard, realizes dynamic routing and improves the network performance, such as the network throughput rate and the like.
The network performance using the different schemes described above is shown in the following table:
as can be seen from the above embodiments, the scheme of the present invention divides the rule deployment in the network into two phases: active wildcard table entry installation and passive exact match table entry installation. Meanwhile, the invention breaks through the thinking mode of completely using static routing or dynamic routing in the network before, provides a mixed routing scheme, and can improve the performance of the software defined network while using wildcard rules (reducing the resource consumption of flow table items) as much as possible.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.
Claims (8)
1. A software defined network efficient routing method based on rule pre-deployment is characterized by comprising the following steps:
s01: predicting and predicting data streams which are likely to arrive at the next time period of each switch according to the state information of the current network, and configuring and deploying corresponding wildcard rules on all switches, wherein the wildcard rules specify macro streams, and each wildcard can be matched with the streams which pass through the switches and have the same characteristics;
s02: when the data flow of the next time period reaches the switch, firstly, the wildcard rules which are configured and deployed in the switch in advance are matched, and if the data flow is matched, the data flow directly carries out forwarding routing according to the wildcard rules; if not, the exchanger packs Packet-In information, and reports the data flow information to the control layer, and the control layer calculates and issues an accurate matching rule according to the state of the whole network, thereby realizing dynamic routing and completing the forwarding of the data flow.
2. The method for efficient routing of software-defined networks based on rule pre-deployment according to claim 1, wherein the set of macro-flows divided by each switch in units of wildcards in step S01 satisfies two constraints: 1) each flow passing through the switch belongs to a set of flows; 2) the flow sets corresponding to any two wildcards do not intersect.
3. The method for efficient routing in software-defined networks based on rule pre-deployment according to claim 1, wherein the step S01 further comprises collecting information of each data flow in the network periodically through sampling packets, and predicting information of macro flow according to the information of each data flow.
4. The efficient routing method for the software-defined network based on rule pre-deployment as claimed in claim 3, wherein the controller calculates the flow size of the macro flow according to the information of the network flow statistics to obtain the number of single flows in each macro flow.
5. The method as claimed in claim 1, wherein the switch selects forwarding routes according to the matching result of the data headers and updates the statistical information when the data flow passes through the switch.
6. The method for efficient routing of software-defined networks based on rule pre-deployment as claimed in claim 1, wherein in step S02, after the data flow selects the subset of macro flows according to the wildcard rule, the static routing entry without traffic is first deleted by prediction, and then the controller installs the updated flow entry to each switch and sets the idle timeout and the hard timeout to zero.
7. The method for efficient routing of software defined networks based on rule pre-deployment according to claim 1, wherein Boolean variables are used in step S01
Indicating whether the controller has deployed a wildcard rule on path p for the macro-flow t,
indicating that the controller is to deploy wildcard rules for the macro-flow t on path p;
indicating that the macro flow t is not configured to deploy wildcard rules on path p.
8. The method as claimed in claim 1, wherein in step S01, the elapsed time of the data flow in the network is divided into fixed time periods, the control layer sends a Read-State request to all SDN switches of all data layers through a network protocol to collect traffic information right at the beginning of each time period, the SDN switches package the data flow information after receiving the Read-State request, generate a response packet and send the response packet to the controller, and the controller collects the responses of all SDN switches and assembles the data flow information of the entire network to obtain the State information of the current network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128144.0A CN111343090A (en) | 2020-02-28 | 2020-02-28 | Rule pre-deployment-based software defined network efficient routing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128144.0A CN111343090A (en) | 2020-02-28 | 2020-02-28 | Rule pre-deployment-based software defined network efficient routing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111343090A true CN111343090A (en) | 2020-06-26 |
Family
ID=71180575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010128144.0A Pending CN111343090A (en) | 2020-02-28 | 2020-02-28 | Rule pre-deployment-based software defined network efficient routing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111343090A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9042234B1 (en) * | 2012-10-31 | 2015-05-26 | Big Switch Networks, Inc. | Systems and methods for efficient network traffic forwarding |
| CN104994065A (en) * | 2015-05-20 | 2015-10-21 | 上海斐讯数据通信技术有限公司 | Access control list operation system and method based on software-defined network |
| US20150339587A1 (en) * | 2014-05-21 | 2015-11-26 | Prophetstor Data Services, Inc. | Adaptive fuzzy rule controlling system for software defined storage system for controlling performance parameter |
| WO2017005328A1 (en) * | 2015-07-09 | 2017-01-12 | Nec Europe Ltd. | Pre-scheduled and pre-configured time-dependent bulk rerouting in transport networks |
| CN107995121A (en) * | 2017-11-27 | 2018-05-04 | 中国科学技术大学苏州研究院 | Flow statistical method in software defined network based on asterisk wildcard |
-
2020
- 2020-02-28 CN CN202010128144.0A patent/CN111343090A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9042234B1 (en) * | 2012-10-31 | 2015-05-26 | Big Switch Networks, Inc. | Systems and methods for efficient network traffic forwarding |
| US20150339587A1 (en) * | 2014-05-21 | 2015-11-26 | Prophetstor Data Services, Inc. | Adaptive fuzzy rule controlling system for software defined storage system for controlling performance parameter |
| CN104994065A (en) * | 2015-05-20 | 2015-10-21 | 上海斐讯数据通信技术有限公司 | Access control list operation system and method based on software-defined network |
| WO2017005328A1 (en) * | 2015-07-09 | 2017-01-12 | Nec Europe Ltd. | Pre-scheduled and pre-configured time-dependent bulk rerouting in transport networks |
| CN107995121A (en) * | 2017-11-27 | 2018-05-04 | 中国科学技术大学苏州研究院 | Flow statistical method in software defined network based on asterisk wildcard |
Non-Patent Citations (2)
| Title |
|---|
| GONGMING ZHAO，等: ""Hybrid routing by joint optimization of per-flow routing and tag-based routing in software-defined networks"", 《TSINGHUA SCIENCE AND TECHNOLOGY ( VOLUME: 23, ISSUE: 4, AUGUST 2018)》 * |
| 庄怀东: ""基于SDN的数据中心网络流量调度与负载均衡研究"", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑 2017年第03期》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10348571B2 (en) | Methods and apparatus for accessing dynamic routing information from networks coupled to a wide area network (WAN) to determine optimized end-to-end routing paths | |
| US8717875B2 (en) | Condensed core-energy-efficient architecture for WAN IP backbones | |
| Rezaee et al. | SDN-based quality of service networking for wide area measurement system | |
| CN110198234B (en) | Method for jointly deploying functions of virtual switch and virtual network in software defined network | |
| Trestian et al. | OFLoad: An OpenFlow-based dynamic load balancing strategy for datacenter networks | |
| WO2017078922A1 (en) | Apparatus and method for network flow scheduling | |
| US10892994B2 (en) | Quality of service in virtual service networks | |
| KR20150017723A (en) | Congestion control in packet data networking | |
| Lin et al. | DTE-SDN: A dynamic traffic engineering engine for delay-sensitive transfer | |
| CN109005126B (en) | Data stream processing method, device and computer readable storage medium | |
| Yan et al. | A survey of low-latency transmission strategies in software defined networking | |
| CN112954069A (en) | Method, device and system for accessing mobile equipment to SD-WAN (secure digital-Wide area network) | |
| Li et al. | Software defined traffic engineering for improving Quality of Service | |
| US20230216775A1 (en) | Packet forwarding method, forwarding indication information advertising method, advertisement packet advertising method, and device | |
| CN105306365A (en) | Power communication network and capacity expansion and destroy-resistant routing path determination method thereof | |
| Liu et al. | A port-based forwarding load-balancing scheduling approach for cloud datacenter networks | |
| CN113746750A (en) | Method, network node, system and storage medium for forwarding data message | |
| CN107995121B (en) | Method for counting flow in software defined network based on wildcard | |
| CN102932268A (en) | Data packet transmitting method and device | |
| Kaur et al. | Load balancing in software defined networking: A review | |
| Duliński et al. | MPLS-based reduction of flow table entries in SDN switches supporting multipath transmission | |
| Isyaku et al. | Dynamic routing and failure recovery approaches for efficient resource utilization in OpenFlow-SDN: a survey | |
| JP2012182605A (en) | Network control system and administrative server | |
| CN111245724A (en) | SDN load balancing routing method based on virtual switch deployment | |
| CN111343090A (en) | Rule pre-deployment-based software defined network efficient routing 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200626 |
|
| RJ01 | Rejection of invention patent application after publication |