KR101541168B1 - Route control method for flow of controller in software defined network - Google Patents

Abstract

According to an embodiment of the present invention, a path control method for a flow of a controller in a software defined network environment comprises the steps of: storing at least one path from a random edge switch disposed on the edge on a network to a different edge switch, and more specifically, allocating identification information for distinguishing a path of each edge switch pair, which is predetermined based on topology of the network, by each path and storing the same; determining a first edge switch and a second edge switch adjacent to a second host by receiving flow information from the first edge switch adjacent to a first host when a flow from the first host to the second host is generated; selecting one path from searched paths by searching for paths between the first edge switch and the second edge switch among stored paths; and transmitting information on a path selected by the first edge switch so the flow is processed through the selected path.

Description

[0001] The present invention relates to a method of controlling a flow of a controller in a software-defined network environment,

The present invention relates to a path control method for a flow of a controller in a software defined network environment.

Software Defined Network (SDN) is a technology for logically centralizing and controlling network functions. Control functions of existing network technology are distributed to individual devices and control the data transmission function of each device. On the other hand, in the software defined network, the control functions are separated from the individual devices and concentrated in separate control devices, and the data transmission functions are controlled in software through the control devices. In other words, the control area is centralized and mainly responsible for route determination and management of the entire network, and the data area is relatively simplified, mainly handling the flow according to the method specified in the control area.

Software-defined networks enable network administrators to program and manage their own networks, providing better and more automated network control.

With respect to the method using this software defined network, Korean Patent No. 10-1438212 discloses a software defined network deep packet analysis method and a software defined network system using the same, and Korean Patent No. 10-1465884 Discloses a stochastic controller selection method and apparatus in a software defined network. Korean Patent Laid-Open Publication No. 2014-0135000 discloses a method and apparatus for processing a service of a software defined networking-based communication system.

However, in software-defined networks, network control functions are centralized in the controller, which can result in control message overhead between the controller and the network equipment. That is, a communication delay may occur between the control area and the data area when managing the network.

In particular, this overhead and communication delay may occur when a new flow handling scheme is designated for the switch. Such overhead and communication delays are not suitable for high performance networks and there is a problem of delaying packet processing of flows until a processing scheme is determined.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a controller and a path control method for a flow that can control a link on a network to be used efficiently while minimizing processing delay caused by a control message.

It is to be understood, however, that the technical scope of the present invention is not limited to the above-described technical problems, and other technical problems may be present.

According to another aspect of the present invention, there is provided a method of controlling a flow of a controller in a software defined network environment, the method comprising: receiving at least one edge switch from an edge switch Allocating and storing identification information for identifying a path of each edge switch pair set in advance on the basis of the topology of the network; When the flow from the first host to the second host is generated, information of the flow is received from the first edge switch adjacent to the first host to determine the first edge switch and the second edge switch adjacent to the second host step; Searching for a path between the first edge switch and the second edge switch among the stored paths and selecting any one of the retrieved paths; And transmitting information about the selected path to the first edge switch such that the flow is processed through the selected path.

Further, a controller in a software defined network environment according to an embodiment of the present invention includes a memory for storing a program for controlling a flow path on a network; And a processor for executing the program.

At this time, according to the execution of the program, the processor stores at least one path from any edge switch disposed at the edge on the network to the other edge switch, and determines the path of each predetermined edge switch pair based on the topology of the network Allocating and storing identification information for each path; When the flow from the first host to the second host is generated, information of the flow is received from the first edge switch adjacent to the first host to determine the first edge switch and the second edge switch adjacent to the second host step; Searching for a path between the first edge switch and the second edge switch among the stored paths and selecting any one of the retrieved paths; And transmitting information about the selected path to the first edge switch such that the flow is processed through the selected path.

According to an embodiment of the present invention, a plurality of paths may be allocated to each flow, a path used may be changed according to a network state, and a basic path may be specified. In addition, the present invention can pre-calculate possible paths for application of fast path change and use it fixedly, thereby processing packets during generation of flow entries, and providing an improved flow processing speed when generating flows.

FIG. 1 is a flowchart for explaining a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention. Referring to FIG.
2 is a diagram illustrating an example of a topology showing a relationship between switches and paths used in a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention.
3 is a diagram illustrating an example of a path information table of a controller used in a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention.
4 is a diagram showing an example of a flow table and an identification information table of a first edge switch A used in a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention.
5 is a flowchart illustrating load balancing in a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification, when a part is referred to as being "connected" to another part, it includes not only "directly connected" but also "electrically connected" with another part in between . Also, when an element is referred to as "including" an element, it is to be understood that the element may include other elements as well as other elements, And does not preclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or combinations thereof.

FIG. 1 is a flowchart for explaining a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention. Referring to FIG.

At first, at least one path from any edge switch disposed on the edge of the network to the other edge switch is stored, and identification information for distinguishing the path of each edge switch pair set on the basis of the topology of the network is allocated for each path (S110).

Then, when a flow from the first host to the second host is generated, information of the flow is received from the first edge switch adjacent to the first host, and the first edge switch and the second edge switch adjacent to the second host (S120).

Subsequently, a path between the first edge switch and the second edge switch of the stored path is searched to select one of the searched paths (S130).

Next, information on the path selected by the first edge switch is transmitted so that the flow is processed through the selected path (S140).

Hereinafter, the above-described steps will be described in detail with reference to FIGS. 2 to 5 with respect to the path control method for the flow of the controller in the software-defined network environment.

2 is a diagram illustrating an example of a topology showing a relationship between switches and paths used in a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention. FIG. 4 is a diagram showing an example of a flow table and an identification information table of the first edge switch A. FIG.

In the present invention, in order to efficiently distribute the load to each link, a plurality of paths may be allocated between the edge switches A to G, and the path used may be changed according to the network state.

At this time, as shown in FIG. 2, the network topology 20 information can be used to determine a plurality of paths for each edge switch pair on the network in advance. Here, each of the edge switches A to G can be used as a transmission destination and a destination, and can be used as a common route to the flow of the controller 200. [ Further, each edge switch may include an identification information table and a flow table. In addition, the edge switch refers to a switch that is disposed on the outer periphery of a plurality of switches on the network topology 20 adjacent to the host.

In addition, in a software defined network environment, the controller and each of the switches exchange information using a predetermined protocol of the network, and the switch can generally receive control information from the controller via the secure channel and allocate and store the control information in the flow table . The flow table consists of rules, actions, and stats. The rules are used to check packet header information for destination management such as port and IP, and actions are used for encryption, blocking, and transmission. Function, and the stat can calculate the number of transmission packets and the transmission status statistics.

Referring to Figs. 2 and 3, a description will be given of a method in which paths are set in advance for each edge switch pair 310. The shortest path is determined for each edge switch pair 310 (x, y) The link on the path can be removed from the topology 20, and the shortest path can be determined again. This process can be repeated to determine all possible paths 320.

The step of storing S110 stores the identification information 330 for distinguishing each path 320 with respect to all predetermined paths 320 corresponding to each edge switch pair 310 based on the topology 20 of the network, And the path 320 information can be retained in the controller 200. [0033] FIG. As will be described in detail later, the shortest path among the paths 320 of each edge switch pair 310 may be allocated as a basic path and used for processing a flow in which a flow entry is not created.

In step S120, a flow from the first host 210 to the second host 220 may be newly generated. In this case, the first edge switch 230 can determine the first edge switch 230 by receiving the information of the flow from the first edge switch 230 adjacent to the first host 210. In addition, the second edge switch 240 can determine the second edge switch 240 by receiving information of the flow from the second edge switch 240 adjacent to the second host 220. [ Here, the information of the flow may include information of the header 410 of the flow.

The selecting step S130 may select any one of the retrieved paths 320 by searching the path 320 between the first edge switch 230 and the second edge switch 240 of the stored path 320 .

The selecting step S 130 may select a path 320 that represents the minimum link utilization 340 among the retrieved paths 320.

Each link utilization rate 340 corresponding to path 320 for each edge switch pair 310 may be reported periodically or at random by at least one of the edge switch pairs 310. [

In order to solve the communication delay between the control area and the data area generated at this stage, a method of minimizing the flow setup delay can be used. Also, the intermediate switches B, C, E, F, and G of each path 320 can be routed using the wildcard and the identification information 330 and 430.

The path selection process will be described in more detail. In the first edge switch 230, the flow can indicate the path in use by using the identification information table (b in FIG. 4) of the first edge switch 230. In generating the flow, the path can be applied by preferentially processing packets using the basic path A-G-D and changing only the flow table of the first edge switch 230 (Fig. 4A). At this time, a path ABCD having the lowest link utilization rate 340 of a plurality of paths 320 is selected, and a flow H1 having the action 420 as the identification information 430 g2 corresponding to the first edge switch 230 is selected → H2 (f1) can be added. In addition, when changing the route of the flow, the identification information table (b in Fig. 4) in the first edge switch 230 of the flow can be changed to process the route change of the flow.

In the transmission step S140, information on the path selected by the first edge switch 230 is processed so as to be processed through the path selected in the above-described selecting step S130 instead of the shortest path (basic path) can do.

When a flow f1 from H1 to H2 is generated, the first edge switch 230 uses the identification information 330 g1 of the basic path AGD, which is the shortest path, to transmit the packet of the flow f1 And transmits the flow header 410 of f1 to the controller 200 to query the processing method for the flow f1.

The controller 200 recognizes the first edge switch 230 and the second edge switch 240 to which the first host 210 and the second host 220 are connected, respectively. The controller 200 selects the identification information 330 g2 corresponding to the path ABCD having the lowest link utilization rate 340 of 0.3 among the paths 320 between the first and second edge switch pairs A and D , It is possible to select the path for processing the flow f1. Next, an entry for flow f1 is added to the flow table (a in Fig. 4) of the first edge switch 230 and the identification information 430 g2 corresponding to the path 320 ABCD through the action 420 By routing, routing information can be input to the first edge switch 230.

Here, the packet process will be described in detail. When the packet enters the edge switch, the header information of the packet can be checked by the rules of the flow table. At this time, if the header information of the packet matches the information of the flow header 410, the packet can be processed by the designated action 420. On the other hand, if the packet header information does not match the flow header 410 information, the edge switch may request the controller 200 for the processing method. The requested controller 200 can designate the processing of the packet and install it in the edge switch flow table to process the packet.

The method of controlling path 320 for the flow of controller 200 described above may further comprise receiving a respective link utilization rate 340 corresponding to path 320 for each edge switch pair 310 have. At this time, the above-described selection step S130 can be selected based on a result of comparing the link utilization rate 340 corresponding to the searched route with a predetermined threshold utilization rate. That is, in selecting a path, the controller 200 can select a route by taking into consideration not only the link utilization rate 340 but also the balancing of the load on the switch, which will be described with reference to FIG.

5 is a flowchart illustrating load balancing in a path control method for a flow of a controller in a software defined network environment according to an embodiment of the present invention.

Load balancing can be performed when the link utilization 340 shows a usage rate above the threshold utilization (threshold).

Referring to FIG. 5, the process of performing load balancing will be described. In operation S510, each edge switch periodically transmits status information to the controller 200. FIG.

The controller 200 may collect status information of each edge switch and flow and check all link status information on the network (S520).

If the link utilization rate 340 corresponding to each edge switch pair 310 is higher than the threshold utilization rate (threshold value), the step S520 is repeated and the link utilization rate 340 corresponding to each edge switch pair 310 is compared with the threshold utilization ratio (Threshold value), the step S540 may be performed (S530).

The controller 200 can recognize a flow used for the corresponding edge switch pair 310 and select a flow to change the path 320 (S540).

Here, the selected flow (to be selected) may be a flow in which there is a packet that has been transmitted recently within a specific period, and the number of bytes of the packet transmitted in the flow not subjected to load balancing is the maximum.

If the link utilization rate 340 corresponding to each edge switch pair 310 is higher than the threshold utilization rate (threshold value), the step S520 is repeated and the link utilization rate 340 corresponding to each edge switch pair 310 is compared with the threshold utilization ratio (Threshold value), the step S540 may be performed (S550).

Illustratively, when the link utilization rate 340 of the path 320 ABCD of the edge switch pair A and D is equal to or greater than a predetermined threshold utilization threshold, the first edge switch 230 is connected to the path 320 ABCD (330) g1 and g3, the part of the flow assigned to the identification information (330) g2 corresponding to the identification information (330) g2. The controller 200 notifies the controller 200 of the state change of the path 320 ABCD when the link utilization rate 340 is equal to or greater than the threshold use rate (threshold value), and the controller 200 notifies the controller 200 of the first and second edge switch pairs A and D, And can change from the oldest flow of the route. At this time, when the path has recently performed load balancing, the controller 200 may change the flow of the next rank. When the path change is completed, the load balancing target path and the load balancing execution time of the alternate path may be updated. This process can be repeated until the link utilization rate 340 falls below a predetermined threshold utilization rate (threshold value).

In the software defined network environment, the path control method for the flow of the controller may include the controller 20. In addition, the controller 20 may be composed of a memory and a processor.

The memory may store a program that controls the path of the flow on the network. In addition, the memory may store additional generated or transmitted and received information.

The processor may execute the program. At first, the processor stores at least one path from any edge switch to the other edge switch disposed at the edge on the network according to the execution of the program, and determines the path of each preset edge switch pair based on the topology of the network The identification information for each path can be allocated and stored.

Then, when a flow from the first host to the second host is generated, the processor receives the information of the flow from the first edge switch adjacent to the first host and generates a first edge switch and a second edge adjacent to the second host, The switch can be determined.

Then, the processor can select one of the paths retrieved by searching the path between the first edge switch and the second edge switch of the stored path.

Next, the processor may send information about the selected path to the first edge switch so that the flow is processed through the selected path.

In this software-defined network environment, the path control method for the flow of the controller can reduce the overhead and communication delay that occur when a new flow processing method is specified using a predetermined path. In addition, as the overhead and communication delay are reduced, the processing speed of the flow can be improved.

It will be understood by those skilled in the art that the foregoing description of the present invention is for illustrative purposes only and that those of ordinary skill in the art can readily understand that various changes and modifications may be made without departing from the spirit or essential characteristics of the present invention. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

20: Topology 200: Controller
210: first host 220: second host
230: first edge switch 240: second edge switch
310: switch pair 320: path
330, 430: identification information 340: link utilization rate
410: Flow Header 420: Action

Claims (6)

CLAIMS 1. A path control method for a flow of a controller in a software defined network environment,
Wherein at least one path from one edge switch to another edge switch disposed on an edge of the network is stored, and identification information for identifying a path of each edge switch pair set in advance on the basis of the topology of the network is stored for each path Allocating and storing the data;
When a flow from a first host to a second host is generated, receiving information of the flow from a first edge switch adjacent to the first host and transmitting the flow information from the first edge switch to a second edge adjacent to the second host Determining a switch;
Selecting a route indicating the minimum link utilization rate among the searched routes by searching a route between the first edge switch and the second edge switch among the stored routes; And
And transmitting information about the selected path to the first edge switch such that the flow is processed through the selected path,
Wherein the step of selecting includes performing load balancing and selecting a load balancing target path and an alternative path load balancing when the path selection is completed when each of the link utilization ratios corresponding to the searched path is higher than a preset threshold utilization ratio, Wherein the time is updated. delete The method according to claim 1,
Wherein each link utilization rate corresponding to the path for each edge switch pair is periodically or randomly reported by at least one of the pair of edge switches. The method according to claim 1,
Further comprising receiving each link utilization rate corresponding to a path for each of the edge switch pairs. The method according to claim 1,
Wherein the information of the flow includes header information of the flow. A controller in a software defined network environment,
A memory for storing a program for controlling a path of a flow on the network; And
And a processor for performing the program,
The processor, upon execution of the program,
Wherein at least one path from one edge switch to another edge switch disposed on an edge of the network is stored, and identification information for identifying a path of each edge switch pair set in advance on the basis of the topology of the network is stored for each path Allocating and storing the data;
When a flow from a first host to a second host is generated, receiving information of the flow from a first edge switch adjacent to the first host and transmitting the flow information from the first edge switch to a second edge adjacent to the second host Determining a switch;
Selecting a route indicating the minimum link utilization rate among the searched routes by searching a route between the first edge switch and the second edge switch among the stored routes; And
And transmitting the information about the selected path to the first edge switch so that the flow is processed through the selected path, wherein the step of selecting includes the step of determining whether each link utilization rate corresponding to the retrieved path satisfies a predetermined threshold utilization ratio The load balancing execution time of the load balancing target path and the alternative path is updated when the path selection is completed.

Images