KR101508349B1 - Device and method for managing of the server in content delivery network - Google Patents

Abstract

A method for managing a server in a content delivery network (CDN) according to the present invention comprises: a step for receiving network information of the content delivery network; a step for calculating a cache server disposition coefficient of each node from the network information considering the number of user terminal requests or cooperation among neighboring nodes; and a step for selecting a node for a new cache server using the calculated cache server disposition coefficient.

Description

TECHNICAL FIELD [0001] The present invention relates to a server management apparatus and a server management method in a content delivery network,

The present invention relates to a server management apparatus and a server management method in a content delivery network.

In a network before a content delivery network exists, when a user terminal requests content, it is received from a source server. Therefore, this causes a phenomenon in which users' requests are pushed to a content source server, thereby causing a serious reduction in quality of service for content delivery.

Therefore, recently, a content delivery network (CDN) has been proposed. In order to improve the quality of service for delivering contents, a plurality of servers storing contents are positioned at an edge of a network close to a user terminal.

However, these servers are generally unable to store all the content that exists in the network due to storage limitations and the amount of myriad content. Therefore, each cache server has to provide services with restricted contents, and the content that the server does not have has to be served from the original server of the contents, which causes the service time delay.
As a prior art document related to the present invention described below, there is a method of scheduling distribution of a cache server network and a content file in Korean Patent Application Publication No. 2007-0003920 (published on Jan. 5, 2007).

The present invention proposes a server management apparatus and a server management method in a content delivery network in which a cache server can be placed in consideration of a relationship with neighboring nodes.

A method of managing a server in a content delivery network (CDN), the method comprising: receiving network information of the content delivery network; Calculating a cache server placement coefficient of each node in consideration of the cooperation relationship with the number or peripheral nodes, and determining the node to which the cache server is to be placed using the calculated cache server placement coefficient .

The step of receiving the network information may include receiving network topology information of an area in which the server is to be located and calculating hop count information based on the received network topology information.

The cache server placement coefficient may be calculated using the number of user terminals requesting content for each node and the hop count information.

The calculating step may include calculating the casser server placement coefficient at all nodes present in the network.

The determining may include determining to preferentially place a cache server on a node with a high cache server placement factor.

A method for managing a server in a content delivery network (CDN), comprising the steps of: checking network information of the content delivery network; Calculating a cost according to the placement or the cost of transferring the content, and determining the node to which the cache server is to be placed using the calculated cost.

The confirming may include calculating hop count information between the nodes using the network information or checking the number of requests of the user terminals of each node.

The step of calculating the cost may include calculating a distribution cost generated when the cache server is arranged using the hop count information and transmitting the content between the servers using the hop count information and the number of requests of the user terminal, And calculating a transmission cost incurred when the content is delivered to the user terminal.

The transmission cost may be the same as in the case where the user terminal receives the content from the associated cache server, the user terminal receives the content from the cache server in cooperation with the associated cache server, or the user terminal receives the content from the source server And may include any one of them.

Controlling the content to be transmitted to the cache server disposed at the determined node, and controlling the content to be transmitted from the content having a low transmission cost.

A server management apparatus in a content delivery network according to the present invention includes a receiver for receiving network information of a content delivery network (CDN), a cache server batch coefficient calculating unit for calculating a cache server batch coefficient using the network information, And a controller for controlling the cache server placement coefficient or the cost to determine a node to which the cache server is to be placed.

The calculating unit may include a coefficient calculating unit that calculates the cache server placement coefficient using a hop count between nodes and a request number of a user terminal of each node.

The coefficient calculator may calculate the cache server batch coefficient in proportion to a hop count between nodes and in proportion to the number of requests of the user terminal.

The calculating unit may further include a cost calculating unit that calculates a distribution cost generated when the cache server is disposed or a transmission cost generated when the content is delivered from the further arranged cache server.

The control unit can control to preferentially arrange the cache server in the node having the high cache server placement coefficient.

The control unit may control the cache server to be disposed in a node having a high number of requests from the user terminal and a small hop count from neighboring nodes.

According to the present invention, a cache server can be arranged in consideration of a number of requests and a hop count of a user terminal, thereby reducing costs and improving the quality of service provision.

1 is a view schematically showing a structure of a content delivery network (CDN).
2 is a block diagram briefly showing a server management apparatus according to an embodiment of the present invention.
3 is a diagram illustrating an example in which a user terminal receives content from an associated cache server.
4 is a diagram illustrating an example of receiving content from another cache server in cooperation with a cache server.
5 is a diagram showing an example of receiving contents from an original server.
6 is a flowchart illustrating a server management method according to an embodiment of the present invention.
Figure 7 is a diagram illustrating the Casser server placement coefficients of nodes for a content delivery network.
8 is a flowchart illustrating a server management method according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. 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 an element is referred to as "comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise. Also, the terms " part, "" module," and " module ", etc. in the specification mean a unit for processing at least one function or operation and may be implemented by hardware or software or a combination of hardware and software have.

A server management apparatus and a server management method in a content delivery network according to an embodiment of the present invention will now be described in detail with reference to FIGS. 1 to 8. FIG.

1 is a view schematically showing a structure of a content delivery network.

Referring to FIG. 1, a content delivery network (CDN) 10 includes a source server 12 storing a content source, a plurality of cache servers 14) exists.

And, the content delivery network can greatly reduce the delay for the content service through the cooperation relationship between the cache servers 14. Thus, one embodiment of the present invention contemplates cooperation between cache servers in deploying the cache server 14 in a content delivery network.

That is, the server management apparatus 100 according to an exemplary embodiment of the present invention provides an environment in which a cache server that reflects the number of user requests of a node to which a cache server is to be installed and a relationship with a neighboring node is arranged.

Therefore, it is possible to reduce the processing time of the source server as much as possible and to transmit the content to the user through the cooperation between the cache servers, even when the specific cache server does not have the content, from the neighboring cache server having the content. Thus, as opposed to not cooperating, cooperation between cache servers allows the cache servers to handle more user terminal requests.

2 is a block diagram briefly showing a server management apparatus according to an embodiment of the present invention. At this time, the server management apparatus 100 only shows a schematic configuration necessary for explanation according to the embodiment of the present invention, and is not limited to this configuration.

Referring to FIG. 2, the server management apparatus 100 according to an embodiment of the present invention may add a cache server in consideration of cooperation relationships of cache servers in a content delivery network (CDN) And transmits the content in consideration of the content. The server management apparatus 100 according to an exemplary embodiment of the present invention includes a receiving unit 110, a calculating unit 120, a determining unit 130, and a controlling unit 140.

The receiving unit 110 receives the network information of the content delivery network. The receiving unit 110 receives the network topology information of the area where the server is to be placed, and the controller 140 grasps the hop count information.

The receiving unit 110 may determine the number of the user terminals requesting contents for each node from the received network information, and may provide the control unit 140 with the number of the user terminals.

The calculation unit 120 calculates a cache server placement coefficient using the network information or calculates a cost based on the placement of the cache server. Here, the cache server placement coefficient is a numerical value indicating a tendency to place the cache server. The cache server placement coefficient indicates a higher value as the number of requests from a user terminal in a specific node or the hop count between nodes is smaller.

The calculating unit 120 includes a cost calculating unit 122 and a coefficient calculating unit 124 according to an embodiment of the present invention.

The cost calculation unit 122 calculates a dissemination cost generated when the cache server is disposed or calculates a transmission cost generated when the content is delivered from the arranged cache servers.

At this time, the objective function P for optimizing the cost generated by additionally arranging the cache server is expressed by Equation 1 below.

Here, N is a set of nodes, K is a set of contents, S _j is a total storage size of the jth cache node, and λ _i is a request number of the user terminal in the i th node. P _ik is the probability that the user terminal will request the k th content in the ith node, b _k is the size of the k th content, and c _ij is the hop count cost for delivering the content from node j to node i.

C _j0 is the hop count cost for delivering content from the source server to the jth cache node, T is the allowable number of cache nodes, _Uj is the number of services per unit time that the jth cache node can provide .

The distribution cost is the cost of installing the cache server. The placement of the cache server in a specific node means that the contents to be stored in the cache server are consumed by consuming the hop count cost from the source server to the node where the cache server is to be placed do.

And, the transmission cost is the hop count cost that occurs when the content is delivered to the user terminal. This transmission cost may be related to the case where the user terminal receives content from the associated cache server (local), the user terminal receives the content from the cache server in cooperation with the associated cache server (Cooperation) Or the origin of the content.

3 is a diagram showing an example in which a user terminal 20 receives contents from an associated cache server 14 and FIG. 4 shows an example of contents (contents) from another cache server 14b cooperating with the cache server 14a. FIG. 5 is a diagram illustrating an example of receiving content from the source server 12. As shown in FIG.

The coefficient calculating unit 124 calculates the cache server placement coefficient using the inter-node hop count and the number of requests of the user terminal of each node. At this time, the cache server placement coefficient? _J is inversely proportional to the hop count between nodes as shown in Equation (2) below, and is proportional to the number of requests from the user terminal.

은 노드의 집합이고,

는 노드 j에서의 요청 수이며,

는 노드 i에서 노드 j까지의 홉카운트 이다.here,

Is a set of nodes,

Is the number of requests at node j,

Is the hop count from node i to node j.

Therefore, the cache server placement coefficient (ρ _j ) shows a higher value as the number of requests of the user terminal at a specific node is small or the hop count between nodes is small, so that the cache server deployment coefficient is closer to neighboring nodes.

The determining unit 130 may determine a node to which the cache server is additionally arranged using the cache server placement coefficient or may determine a node to which the cache server is to be placed considering various costs associated with addition of the cache server.

The control unit 140 controls the determination unit 130 to determine whether or not the cache server is additionally disposed, and controls the content to be stored in the cache server. At this time, the control unit 140 controls to preferentially arrange the cache server in the node having the high cache server placement coefficient.

The control unit 140 can control the cache server to be disposed in a node having a large number of user terminals and a small hop count with neighboring nodes.

Referring now to FIG. 6, a process for managing a server management device in a content delivery network according to an embodiment of the present invention to calculate a cache server placement coefficient and add a server will be described in detail.

6 is a flowchart illustrating a server management method according to an embodiment of the present invention. At this time, the following flowchart will be described using the same reference numerals in conjunction with the configurations of Figs. 1 to 5.

Referring to FIG. 6, the server management apparatus 100 receives network information of the content delivery network and calculates hop count information based on the received network information (S100, S110).

At this time, the server management apparatus 100 receives the network topology information of the region in which the server is to be located, and calculates the hop count information based on the received network topology information.

Then, the server management apparatus 100 calculates the cache server placement coefficient of each node using the number of requests and the hop count information of the user terminal (S120).

Here, the server management apparatus 100 calculates a cache server placement coefficient by using the number of user terminals requesting content for each node and the hop count information, and for each node that exists on the network and is operable as a cache server .

In addition, the server management apparatus 100 determines the node to which the cache server is to be placed using the calculated cache server placement coefficient, and determines to place the cache server preferentially in the node having the high cache server placement coefficient (S130).

Figure 7 is a diagram illustrating the Casser server placement coefficients of nodes for a content delivery network.

Referring to FIG. 7, the server management apparatus 100 calculates the cache server placement coefficients rho 1 to rho 9 by each of the nodes. When the cache server placement coefficient? 5 of the specific node 30 is the largest, the server management apparatus 100 determines to preferentially arrange the cache server in the specific node 30. Therefore, the server management apparatus 100 can install a cache server from a node having a high cache server placement coefficient.

Referring now to FIG. 8, a process for managing a server management apparatus in a content delivery network according to an embodiment of the present invention to add a server in consideration of cost will be described in detail.

8 is a flowchart illustrating a server management method according to another embodiment of the present invention. At this time, the following flowchart will be described using the same reference numerals in conjunction with the configurations of Figs. 1 to 5.

First, the server management apparatus 100 receives the network information of the content delivery network (S200).

The server management apparatus 100 calculates the distribution cost generated when the cache server is arranged using the hop count information between the nodes and uses the hop count information and the number of requests from the user terminal to calculate the content (S210, S220). The transmission cost is calculated when the content is transmitted to the user terminal.

Here, the transmission cost may be a value obtained when the content is received from the cache server to which the user terminal is associated, when the user terminal receives the content from the cache server cooperating with the associated cache server, or when the user terminal receives the content from the source server Or the like.

Then, the server management apparatus 100 determines a node where the cache server is to be placed using the cache server placement coefficient (S230).

Then, when the cache server is located at the corresponding position, the server management apparatus 100 controls to store the content in the arranged cache server, and controls to store contents having a low transmission cost at step S240.

)를 계산하고, 이를 통해 전송 비용(

)이 제일 낮은 콘텐츠부터 캐시 서버에 저장하도록 제어한다. The server management apparatus 100 stores the transmission cost when each content is stored in the cache server to store the content in the installed cache server

) Is calculated, and the transmission cost (

) From the lowest content to the cache server.

As described above, the server management apparatus and the server management method according to an embodiment of the present invention can control the arrangement of the cache servers in consideration of the number of requests from the user terminals and the relationship with the neighboring nodes, And the like.

The server management apparatus according to an embodiment of the present invention uses a cache server batch coefficient that considers both the number of users and the hop count of neighboring nodes in selecting a node to which a cache server is to be added.

The server management apparatus according to an embodiment of the present invention can arrange a cache server in consideration of cooperation between cache servers in a content delivery network and control contents to be stored in consideration of various costs. It is highly likely to be used in the environment.

The embodiments of the present invention described above are not implemented only by the apparatus and method, but may be implemented through a program for realizing the function corresponding to the configuration of the embodiment of the present invention or a recording medium on which the program is recorded.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

110: Receiving unit 120:
130: Decision section 140:

Claims (16)

In a method for managing a server in a content delivery network (CDN)
Receiving network information of the content delivery network;
Calculating a cache server placement coefficient of each node in consideration of the number of requests from the user terminal or cooperation relationship with neighboring nodes from the network information, and
Determining a node to which the cache server is to be placed using the calculated cache server placement coefficient
/ RTI >
Wherein the calculating step comprises:
The cache server placement coefficient is inversely proportional to a hop count between nodes and proportional to the number of user terminals of each node
A server management method in a content delivery network. The method of claim 1,
Wherein the receiving the network information comprises:
Receiving network topology information of an area in which the server is to be located, and calculating hop count information based on the received network topology information
Wherein the server management method comprises: 3. The method of claim 2,
Wherein the cache server placement coefficient comprises:
The number of the user terminals requesting content for each node, and the hop count information. 4. The method of claim 3,
Wherein the calculating step comprises:
Computing the cache server placement factor at every node in the network
Wherein the server management method comprises: 5. The method of claim 4,
Wherein the determining comprises:
Determining to place a cache server preferentially on a node having a higher cache server placement coefficient
Wherein the server management method comprises: In a method for managing a server in a content delivery network (CDN)
Confirming network information of the content delivery network, calculating hop count information between nodes using the network information, or confirming the number of requests of user terminals of each node,
Calculating a cost according to the additional arrangement of the server or a cost of transmitting the content using the network information, and
Determining a node to which a cache server is to be allocated using the calculated cost
/ RTI >
Wherein the calculating the cost comprises:
Calculating a distribution cost generated when the cache server is arranged using the hop count information, and
Calculating a transmission cost generated when the content is delivered between the server and the user terminal using the hop count information and the number of requests of the user terminal,
Wherein the server management method comprises: delete delete The method of claim 6,
The transmission cost,
When the user terminal receives the content from the associated cache server, when the user terminal receives the content from the cache server in cooperation with the associated cache server, or when the user terminal receives the content from the source server A server management method in a content delivery network. The method of claim 9,
Controlling the content to be transmitted to the cache server disposed at the determined node,
Further comprising the steps of: A receiving unit for receiving network information of a content delivery network (CDN)
A calculation unit for calculating a cache server placement coefficient by using the network information or calculating a cost based on the placement of the cache server,
A cache server placement coefficient or the cost to determine a node to which the cache server is to be placed,
/ RTI >
The calculating unit calculates,
A coefficient calculation unit for calculating the cache server placement coefficient using a hop count between nodes and a request number of a user terminal of each node,
Lt; / RTI >
Wherein the coefficient calculating section calculates,
The cache server placement coefficient is inversely proportional to a hop count between nodes and is proportional to the number of requests of the user terminal
A server management device in a content delivery network. delete delete 12. The method of claim 11,
The calculating unit calculates,
A cost calculation unit for calculating a distribution cost generated when a cache server is disposed or a transmission cost generated when a content is delivered from a further arranged cache server,
The server management apparatus comprising: The method of claim 14,
Wherein,
Wherein the cache server preferentially places the cache server in a node having a higher cache server placement coefficient. 12. The method of claim 11,
Wherein,
Wherein the cache server is arranged in a node having a large number of requests from the user terminal and a small hop count with neighboring nodes.

Images