KR101966430B1 - System and Method for Determining Fog Server Number and Placement in Local Area Network Environment - Google Patents

Abstract

The present invention relates to a method and apparatus for determining the number and location of fog servers in a local network environment that can increase the service processing speed and minimize the network bandwidth usage by determining the number and location of the fog servers in consideration of the network topology and the flow status in fog computing A fog device which is an intermediate SDN switch and processes the service of the cloud at an edge; a fog device connected to all the fog devices to transmit and receive computing resource information, and to provide network topology information through an SDN controller; A fog manager for performing network status monitoring and searching for an optimal fog device, a client requesting a service to the fog server, and a data source having data necessary for service processing, wherein the client and the data sources Iru In the binary service flows to place the fog to the fog server device by selection of fog device the distance between the client and the data source device and the fog is minimized.

Description

[0001] The present invention relates to a method and apparatus for determining the number of fog servers and a location of a fog server in a local network environment,

The present invention relates to fog computing. More specifically, the present invention relates to a fog server in a local network environment that can increase service processing speed and minimize network bandwidth usage by determining the number and location of fog servers in consideration of network topology and flow status And an apparatus and a method for determining the position of the placement.

Cloud computing refers to the way a user accesses a central data center through a network and uses hardware, software, and data.

However, there are many problems with such cloud services.

Typically, the distance between the cloud and the user is so long that the latency is long. As a result, many users are using the cloud service because it is combined with the Internet (IoT) and the whole network traffic is increased.

To solve these problems, a new paradigm called fog computing has recently been proposed.

FogComputing is a model that provides some of the services provided by the cloud in the existing cloud computing concept in devices in the edge network.

By providing the functionality provided by the cloud in relatively near-edge devices, users can drastically reduce latency and reduce overall network traffic.

However, the fog devices lack computing resources relative to handling a large number of requests compared to a cloud server handling many users' services.

For this reason, if a service request is increased to a specific fog device, a load may be concentrated on the device, which may cause a problem in service processing.

And because users do not always use the same service, traffic on the network is constantly changing.

Even if the fog server is deployed, the flow to the service continuously changes, so running the service on the same fog server always leads to inefficient results.

Therefore, there is a need to develop a new technology that can reduce the traffic of the entire network and reduce the response time to the service request of the user by arranging the fog server at a location where the path of traffic movement can be minimized .

Korea Patent No. 10-1574026 Korean Patent Publication No. 10-2015-0120555

The present invention solves the problem of server placement in fog computing as described above. The present invention determines the number and location of fog servers in consideration of network topology and flow state, thereby increasing service processing speed and minimizing network bandwidth usage And to provide a device and method for determining the number and location of fog servers in a local network environment.

In the local network environment, a fog server having a minimum distance from the corresponding devices to a fog device is selected based on a specific service flow including a client and data sources. And an apparatus and a method for determining the position of the placement.

In the present invention, a fog server is disposed at a location where traffic can be minimized, thereby reducing the traffic of the entire network and reducing the response time for a service request of the user by processing the service. And an apparatus and a method for determining the position of the placement.

In the present invention, the number of fog servers and the number of fog servers in a local network environment in which a load can be distributed by selecting optimal fog devices to be executed by a fog server in consideration of the existence of various network flows, And an apparatus and a method for determining the position of the placement.

In the present invention, a fog server having a minimum distance between the client and the data sources and the fog device is selected in the service flow including the client and the data sources, and the fog server is disposed in the corresponding fog device, The present invention provides a device and method for determining the number and location of fog servers in a local network environment in which network resources can be efficiently utilized by allowing a greater bandwidth for a fog server.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

In order to achieve the above object, an apparatus for determining the number and location of a fog server in a local network environment according to the present invention is an intermediate SDN switch, and the fog devices for processing services of the cloud at the edge are connected to all the fog devices A fog manager that receives and maintains computing resource information, receives network topology information through an SDN controller, and monitors network status and finds an optimal fog device; a client that requests service from a fog server; And selecting a fog device having a minimum distance between the client and data sources and a fog device in a service flow composed of the client and the data sources and transmitting the fog server to the corresponding fog device And it characterized in that.

Here, the fog device serving as the SDN switch is characterized in that the fog server is executed as a container by using a container as a container-based virtualization tool.

And the number of fog devices in which the fog server can be executed is the largest as the number of service flows, and the number of fog devices in which the fog server can be executed is from one to the number of the service flows.

The number of cases in which the service flows can be divided by the number of batches is obtained and fog devices having the minimum number of hops used by the clients and data sources are found for each number of cases and stored in pairs. do.

The target fog device T, which is the fog devices having the minimum number of hops used by the client and the data sources, is used for each number of the fog servers, and the number of used hobs, And stores the device information.

Then, if a smaller number of used hops is found in the pair, the information is stored in the target fog device (T), and after all the operations are performed, the fog server is finally placed in the fog device stored in the target fog device .

A method for determining the number and location of a fog server in a local network environment according to the present invention for achieving another object is characterized in that a fog manager is connected to all fog devices to transmit and receive computing resource information, Determining a number of fog devices in which the fog server can be executed, calculating the number of cases in which the requested service flows can be divided by each batch number, (T), which are fog devices having the minimum number of hops used by clients and data sources, are stored in a pair of all the fog server devices The number of used hops in which the number of used hops is minimized and the number of used hops Storing the device information, and, after all the operations are performed, placing the fog server in the fog device stored in the target fog devices T at the end.

Here, the flow for the service is not fixed, but is repeatedly performed periodically at a predetermined time according to the request of the user, and the fog server is continuously arranged at another place.

And a fog server is disposed in one fog device when the number of used hogs in which the fog server is arranged in one fog device and the number of used hogs in which the fog server is arranged are equal to the number of used fog devices in one fog device, do.

If a smaller number of used hops is found in the pair, the information is stored in the target fog devices T. When all the operations are performed, the fog server is finally placed in the fog devices stored in the target fog devices T .

The apparatus and method for determining the number and location of fog servers in a local network environment according to the present invention have the following effects.

First, by considering the network topology and the flow state, the number of fog servers and the location of the fog server are determined, thereby increasing the service processing speed and minimizing the network bandwidth usage.

Second, a fog server is disposed at a position where the path of the traffic can be minimized to process the service, thereby reducing the traffic of the entire network and reducing the response time of the service request of the user.

Third, it is possible to distribute the load by selecting the optimal fog devices to be executed by the fog server in consideration of the existence of various network flows so that no load occurs in the specific fog server.

Fourth, since the computing resources of the fog devices to be mounted by the fog server are not sufficient, it is possible to save the computing resources of the entire network devices by arranging the fog servers only at necessary positions without disposing the fog servers at all the fog devices.

Fifth, by selecting fog devices having a minimum distance between the client and the data sources and the fog device in the service flow including the client and the data sources, the fog server is disposed in the corresponding fog device, thereby reducing traffic of the entire network, It is possible to utilize the network resources efficiently.

1 is a block diagram of an apparatus for determining the number and location of fog servers in a local network environment according to the present invention;
2 is a block diagram showing a flow example in which a fog server arrangement algorithm according to the present invention is applied
FIG. 3 shows a result obtained by applying the method for determining the number and position of fog servers according to the present invention
FIG. 4 is a flowchart of three cases in which the method for determining the number of fog servers and the positioning of the fog according to the present invention is applied
5 is a flowchart illustrating a method for determining the number of fog servers and their placement in a local network environment according to the present invention;

Hereinafter, a preferred embodiment of an apparatus and method for determining the number of fog servers and the location of a fog server in a local network environment according to the present invention will be described in detail.

The features and advantages of the apparatus and method for determining the number and placement of fog servers in a local network environment according to the present invention will be apparent from the following detailed description of each embodiment.

FIG. 1 is a block diagram of an apparatus for determining the number and location of fog servers in a local network environment according to the present invention.

The apparatus and method for determining the number and placement of fog servers in the local network environment according to the present invention are not required to place fog servers on all fog devices because the computing resources of the fog servers (network devices) Thereby saving the computing resources of the entire network devices.

Particularly, by selecting fog devices having a minimum distance between the client and the data sources and the fog device in a service flow composed of clients and data sources, it is possible to reduce the traffic of the entire network by arranging the fog server in the corresponding fog device, And thus the network resources can be efficiently utilized.

A fog server is a server that makes services that are processed in a cloud server to be processed by a fog device at the edge of the network.

And, when a client requests a service, the fog server fetches necessary data from a data source located somewhere to process the service, and processes the data. Service flow.

In a video streaming service, for example, a user who wants to view a video becomes a client, a place where the video data is stored becomes a data source, a fog server fetches necessary video data from a data source, Flow is formed.

The apparatus for determining the number and location of fog servers in the local network environment according to the present invention comprises a fog device 300 that is an intermediate switch and processes the service of the cloud at the edge, A Fog manager 100 for receiving network topology information through an SDN controller and monitoring the network status and searching for an optimal foggy device, A requesting client 200, and a data source 400 having data necessary for service processing.

In the embodiment of FIG. 1, eight clients 200 and four data sources 400 are connected to four SDN switches (fog devices) 300, and all SDN switches are connected to the SDN controller to determine where to send the packet Lt; / RTI >

The SDN switch can also be used as a container-based virtualization tool.

Table 1 shows an algorithm for selecting a foggy device to place the fog server.

In Line 2, the number of fog devices that the fog server can run on is the maximum of the number of flows, and since it is pointless to run on more fog servers than that, the number of fog devices that the fog server can run from is one to the number of flows Lt; / RTI >

In Line 3, the number of cases in which flows can be divided by the number of batches is obtained. In Lines 5-7, the fog devices with the smallest number of hops are used by the client and data sources in each case.

The target fog device T stores information on the number of used hops and the devices to be placed at that time, according to the number of all fog server batches. If a smaller number of used hops is found in the pair, After all the work is done, the fog server is placed in the last fog device stored in T.

5 is a flowchart illustrating a method for determining the number of fog servers and their placement in a local network environment according to the present invention.

As shown in FIG. 5, a method for determining the number and location of fog servers in a local network environment according to the present invention is as follows. The Fog manager is connected to all fog devices to receive and maintain computing resource information, (S501).

Then, the number of fog devices in which the fog server can be executed is determined (S502)

Next, the number of cases in which the flows can be divided by the respective arrangement numbers is calculated (S503)

Then, in each case, the fog devices having the minimum number of hops used by the client and the data sources are searched for and stored in the pair (S504)

Then, the target fog devices T store the number of used hops in which the number of used hops is minimized according to all the fog server placement counts and the devices to be arranged at that time (S505)

Then, if a smaller number of used hops is found in the pair, the information is stored in T, and if all the operations are performed, the fog server is finally placed in the fog device stored in T (S506)

The number of fog servers and the positioning of the fog server to which the apparatus and method for determining the fog server number and the location of the fog server in the local network environment according to the present invention are applied will now be described in detail.

FIG. 2 is a configuration diagram illustrating an example of a flow of applying the fog server arrangement algorithm according to the present invention, and FIG. 3 is a result of applying the method of FIG. 2 for determining the number and location of fog servers according to the present invention.

And FIG. 4 is a flow chart of three cases in which the method for determining the number of fog servers and the positioning of fogs according to the present invention is applied.

The simple flow is the same as in FIG. 2, and the result of applying the fog server placement algorithm is as shown in FIG.

When placed in one fog device, the number of used hops becomes 8 and can be placed in one of B, D, G sets.

However, considering the arrangement of two fog devices, the number of used hops becomes 4 (2 + 2) when they are placed in B and G, And finally, the fog server is disposed in the two fog devices B and G to process the service.

After the placement is completed, the service of F1 is processed by B, and the service of F2 is processed by G.

However, since the flow to the service is not fixed, but it changes continuously according to the user's request, the algorithm is executed only once and not repeated.

A more detailed flow will be described. In the topology of FIG. 1, an arbitrary Flow <1>, a Flow <2> having a large number of Paths going through, and a Flow <3> Respectively.

Each flow is the same as in Fig.

Table 2 shows the results when selecting the optimal fog device based on three flows.

Since Flow <1> has the same number of hops as 14 fog servers when deployed on one and two fog devices, deploying on two BEs can waste resources.

Therefore, place it in E one. Flow <2> places the fog server on three fog devices, as in the case of BEG and BCG.

Similarly, Flow <3> places a fog server on three Bog fog devices.

The apparatus and method for determining the number and location of fog servers in the local network environment according to the present invention determine the number and location of fog servers in consideration of the network topology and the flow status.

The computing resources of the entire network devices can be saved by disposing the fog server only at the necessary positions without disposing the fog servers in all the fog devices since the computing resources of the fog devices to be mounted by the fog servers are insufficient.

Particularly, by selecting fog devices having a minimum distance between the client and the data sources and the fog device in a service flow composed of clients and data sources, it is possible to reduce the traffic of the entire network by arranging the fog server in the corresponding fog device, And thus it is possible to efficiently utilize network resources.

As described above, it will be understood that the present invention is implemented in a modified form without departing from the essential characteristics of the present invention.

It is therefore to be understood that the specified embodiments are to be considered in an illustrative rather than a restrictive sense and that the scope of the invention is indicated by the appended claims rather than by the foregoing description and that all such differences falling within the scope of equivalents thereof are intended to be embraced therein It should be interpreted.

100. Fog manager 200. Client
300. Fog device 400. Data source

Claims (10)

Fogg devices that serve as an intermediate SDN switch to service the cloud at the edge;
A fog manager connected to all the fog devices to receive and maintain computing resource information, receive network topology information through the SDN controller, monitor network status and find an optimal fog device;
A client requesting a service to the fog server;
A data source having data necessary for service processing,
Wherein a fog server having a minimum distance between the client and the data sources and the fog device is selected in the service flow including the client and the data sources and the fog server is disposed in the fog device. Apparatus for determining number and placement. The apparatus as claimed in claim 1, wherein the fog server, which is an SDN switch, is implemented as a container by using a container-based virtualization tool as a container. The method of claim 1, wherein the number of fog devices in which the fog server can be executed is the largest as the number of service flows,
Wherein the number of fog devices from which the fog server can be run is from one to the number of service flows. The method according to claim 1, wherein the number of cases in which the service flows can be divided by each batch number is obtained,
Wherein the fog devices having the minimum number of hops used by the client and the data sources are searched for and stored in a pair for each number of the fog servers. 5. The method according to claim 4, wherein the target fog device (T), which is fog devices having the minimum number of hops used by the client and data sources for each number of cases, determines the number of used hobs And information of devices to be placed at that time. The apparatus for determining the number and location of fog servers in a local network environment. 6. The method of claim 5, wherein if a smaller number of used hops is found in the pair, the information is stored in the target fog device (T) Server in a local network environment. &Lt; RTI ID = 0.0 &gt; 8. &lt; / RTI &gt; The Fog manager being connected to all fog devices to receive and maintain computing resource information and to receive network topology information through an SDN controller;
Determining a number of fog devices on which a fog server can be run;
Calculating the number of cases in which the requested service flows can be divided by the respective batch number, finding the fog devices having the minimum number of hops to be used by the client and the data sources in each case, and storing them in a pair ;
The number of used hops in which the number of used hops is minimized according to all the fog server placement counts and the devices to be arranged at that time are stored in the target fog devices T which are the fog devices in which the number of hops used by the clients, ;
And placing a fog server in a fog device stored in the target fog devices T after all the operations are performed. The method according to claim 7, wherein the flow for the service is not fixed, but the fog server is repeatedly arranged at regular intervals at a predetermined time according to the user's request. Method for positioning. The method according to claim 7, wherein, when the number of used hops where a fog server is arranged in one fog device and the number of used hogs in which a fog server is arranged are equal to the number of used fog devices in one fog device, The number of fog servers in a local network environment. The method as claimed in claim 7, wherein if a smaller number of used hops is found in the pair, the information is stored in the target fog devices (T) Wherein the fog server is located in a local network environment.

Images