KR102541077B1 - Method and apparatus for caching content in 5G MEC-based information-centric network - Google Patents

Abstract

The present invention discloses a method and apparatus for caching content in a 5G MEC-based information-centric network. According to the present invention, each of a plurality of MEC (Multi-Access Edge Computing) servers belonging to individual edge networks calculates an expected first content request frequency through request information on the first content of a request node; determining, by each of the plurality of MEC servers, to cache the first content when the calculated first content request frequency is greater than a preset first threshold; calculating, by an SDN controller, a second content request frequency expected through request information on the first content within the individual edge network; and when the calculated second content request frequency is greater than a preset second threshold, the SDN controller determines to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network. There is provided a content caching method comprising the step of doing.

Description

Method and apparatus for caching content in 5G MEC-based information-centric network

The present invention relates to a method and apparatus for caching content in a 5G MEC-based information-centric network.

In order to effectively provide ultra-low latency services, 5G utilizes MEC (Multi-Access Edge Computing), which can provide fast services by moving various network resources from the core cloud to the edge adjacent to users.

ICN (Information-Centric Networking), an information-centric network, is a network that performs name-based routing for effective content transmission and can be usefully used in 5G because of its advantages in terms of mobility management and content security.

Recently, research on reducing content delivery delay and distribution cost by caching content at the edge node in the 5G MEC network environment is being actively conducted.

Among information-centric network content caching techniques proposed in the 5G MEC network environment, in the study in Prior Document 1, freshness calculated by content request and creation time, calculated by the number of hops from the caching node to the requesting node, in order to optimize the service provision time in the 5G network We proposed a content caching method based on the content popularity calculated using the calculated distance factor and the access frequency rate calculated by the frequency of requests to the caching node.

In the study of Prior Document 2, the Caching Resource Allocation (CRA) optimization problem in the 5G environment is based on the characteristics of the edge node, and the Base Station Caching Capacity (BSCC) problem of allocating appropriate caching capacity to the base station connected to the edge and the allocated contents. We classified it as a Request MATching (RMAT) problem connecting the base station and the request node, and proposed a caching technique that minimizes energy consumption and content service time consumed for caching.

However, since the above research analyzes and processes data through a centralized cloud infrastructure, it may generate a load on a central controller such as an SDN controller, which may cause a service delay problem.

[Prior Document 1] O. Serhane, K. Yahyaoui, B. Nour, and H. Moungla, “CnS: A Cache and Split Scheme for 5G-enabled ICN Networks,” IEEE International Conference on Communications (ICC), Dublin, Ireland , June 2020. [Prior Document 2] Y. Tang, “Minimizing Energy for Caching Resource Allocation in Information-Centric Networking with Mobile Edge Computing,” DASC, PiCom, CBDCom, and CyberSciTech, Fukuoka, Japan, Aug. 2019

In order to solve the above problems of the prior art, the present invention is to propose a content caching method and apparatus in a 5G MEC-based information-centric network that can prevent service delay while reducing the load of the central controller.

In order to achieve the above object, according to an embodiment of the present invention, as a content caching method in a 5G MEC-based information-centric network, each of a plurality of Multi-Access Edge Computing (MEC) servers belonging to individual edge networks, calculating an expected first content request frequency through request information on the first content of a request node; determining, by each of the plurality of MEC servers, to cache the first content when the calculated first content request frequency is greater than a preset first threshold; calculating, by an SDN controller, a second content request frequency expected through request information on the first content within the individual edge network; and when the calculated second content request frequency is greater than a preset second threshold, the SDN controller determines to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network. There is provided a content caching method comprising the step of doing.

The method may further include determining, by the SDN controller, to delete the first content cached in a plurality of MEC servers belonging to the individual edge network after determining to cache the first content in the ICN node.

The first content request frequency may be calculated using an exponentially weighted moving average of the content request frequency for the first content calculated in the previous period and the number of requests for the first content transmitted to each of the plurality of MEC servers during the current period. can

The second content request frequency is the content request frequency for the first content from the plurality of MEC servers present in the individual edge network in the previous cycle and the second content request frequency transmitted to the plurality of MEC servers present in the individual edge network during the current cycle. 1 may be calculated using an exponentially weighted moving average of the number of requests for content.

If the contents requested by the requesting node are not cached in the MEC server receiving the interest packet of the requesting node and the ICN node connected to the edge network including the MEC server, the ICN node in the core network through information-centric network routing may transmit the content requested by the requesting node.

The step of determining that each of the plurality of MEC servers caches the first content may include determining the first content as candidate content to be cached when the calculated first content request frequency is greater than a preset first threshold value. doing; and determining whether to cache the candidate content to be cached by optimizing a content delivery consumption cost for the candidate content to be cached.

The step of determining, by the SDN controller, to cache the first content in the ICN node may include, when the calculated second content request frequency is greater than a preset second threshold value, the second content to the ICN node. determining candidate content to be cached; and determining whether to cache the candidate content to be cached by optimizing a content delivery consumption cost for the candidate content to be cached.

According to another aspect of the present invention, a computer readable recording medium storing a program for performing the above method is provided.

According to another aspect of the present invention, a content caching device located in a base station in a 5G MEC-based information-centric network, comprising: a processor; and a memory connected to the processor, wherein an expected first content request frequency is calculated through request information on the first content of a request node, and the calculated first content request frequency is greater than a preset first threshold value. If the first content is large, it is determined that the first content is cached, and the SDN controller calculates a second content request frequency expected through the request information for the first content within the individual edge network to store the first content as the first content. Provided is a content caching device that stores program instructions executed in the processor to delete the cached first content when it is determined to cache in an Information-Centric Networking (ICN) node connected to an individual edge network.

According to another aspect of the present invention, a content caching device connected to a core network in a 5G MEC-based information-centric network, comprising: a processor; and a memory connected to the processor, wherein an expected content request frequency is calculated through request information for first content within an individual edge network, and when the calculated content request frequency is greater than a preset threshold, the After determining to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network and caching the first content in the ICN node, a plurality of MECs belonging to the individual edge network A content caching device that stores program instructions executed in the processor to determine to delete first content cached in a server is provided.

According to the present invention, by analyzing and processing data in the MEC and the central controller through a content caching method, the load concentrated on the central controller is reduced and potential content that is frequently requested in the edge network is cached, thereby enabling an effective information-centric network content service. do.

1 is a diagram illustrating an MEC network environment applied in this embodiment.
2 is a diagram illustrating a process of caching and providing content according to an embodiment of the present invention.
3 is a diagram illustrating a content caching process in the SDN controller according to the present embodiment.

Since the present invention can make various changes and have various embodiments, specific embodiments are illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

1 is a diagram illustrating an MEC network environment applied in this embodiment.

As shown in Figure 1, the MEC network environment applied in this embodiment is composed of a core network (Core network, 100) and an edge network (Edge network, 102).

The core network 100 performs multiple information-centric network routing and is connected to a Software Defined Networking Controller (SDN) 104.

The core network 100 is composed of an ICN node 110, which is a core router, and the edge network 102 is composed of a plurality of MEC servers 112.

MEC servers 112 included in individual edge networks 102 are connected to individual ICN nodes 110 .

Each MEC server 112 is located within a base station within the edge network 102 .

The request node (Consumer, 114) is a node that requests content from the MEC server (112).

, ICN 노드(110)는

, 복수의 MEC 서버(112)는

, MEC 서버(112)를 통해 서비스를 받는 요청 노드(114)는

로 표현된다. Here, the edge network 102 is

, the ICN node 110

, the plurality of MEC servers 112

, the requesting node 114 receiving service through the MEC server 112

is expressed as

Content caching according to the present embodiment is performed in the MEC server 112 , the ICN node 110 connected to the edge network 102 , and the ICN node 110 of the core network 100 .

Content to be cached and whether to cache the corresponding content are determined by the MEC server 112 and the SDN controller 104 based on content request information collected by the MEC server 112 .

Each MEC server 112 included in the edge network 102 calculates an expected content request frequency through the content request information of the request node 114 and determines whether to cache the content according to the calculated content request frequency.

In order to cache content with a high request frequency, each MEC server 112 calculates an expected content request frequency as shown in Equation 1 below using the collected content request information.

는 MEC 서버(112) m에서 계산된 컨텐츠 c에 대한 컨텐츠 요청 빈도를 의미한다. here,

Means the content request frequency for the content c calculated in the MEC server 112 m.

에서 계산된

와 현재 주기 동안 MEC 서버(112) m으로 전달된 컨텐츠 c에 대한 요청 횟수

의 지수 가중 이동 평균을 이용하여 계산된다. Content request frequency is the previous cycle

calculated from

and the number of requests for the content c delivered to the MEC server 112 m during the current period

is calculated using an exponentially weighted moving average of

보다 큰 컨텐츠를 캐싱할 컨텐츠로 결정한다. The MEC server 112 sets a predetermined threshold value for the content request frequency as shown in Equation 2.

Determines the larger content as the content to cache.

The content caching of the SDN controller 104 includes (i) a potential content caching method for caching content with a high frequency of request in the edge network 102 and preventing duplicate content from being cached in a plurality of MEC servers 112. (ii) redundant content removal method for

The SDN controller 104 determines whether to cache the content delivered to the ICN node 110 connected to the edge network 102 based on the content request frequency expected from the edge network 102 calculated through the content request information.

In order to reduce the load on the MEC server 112 due to duplicatively cached content within the same cluster, the SDN controller 104 calculates an expected content request frequency within each edge network 102 as shown in Equation 3.

는 엣지 네트워크(102) e에 존재하는 모든 MEC 서버(112)

에서 컨텐츠 c에 대한 예상 요청 빈도를 의미하고 이전 주기

에서 계산된

와 현재 주기 동안 엣지 네트워크 e에 존재하는 모든 MEC 서버(112)

로 전달된 컨텐츠 c에 대한 요청 횟수 총 합

의 지수 가중 이동 평균을 이용하여 계산된다.

Is all MEC servers 112 present in the edge network 102 e

means the expected request frequency for content c in the previous cycle

calculated from

and all MEC servers 112 present in the edge network e during the current cycle.

Sum of the total number of requests for content c delivered to

is calculated using an exponentially weighted moving average of

As shown in Equation 4 below, the SDN controller 104 caches the content for which the expected content request frequency within the individual edge network 102 is greater than the threshold value in the ICN node 110 connected to the individual edge network 102. is determined by

The SDN controller 104 transfers the same content cached in a plurality of MEC servers 112 based on the content information cached in the MEC server 112 in the edge network 102 to the ICN node 102 connected to the edge network 102. cached in and deleted from the plurality of MEC servers (112).

As shown in Equation 4, when the number of identical contents cached in the plurality of MEC servers 112 in the edge network 102 is greater than the threshold value, the corresponding contents are connected to the edge network 102 including the plurality of MEC servers 112. Content to be cached in the ICN node 110 is determined.

The above process may be a process of determining candidate content to be cached by the plurality of MEC servers 112 and the SDN controller 104, and the final caching may be determined by solving the optimization problem of content delivery consumption cost as follows. .

The consumption cost when delivering content c to all request nodes 114 belonging to m of the MEC server 112 is calculated as in Equation 5.

,

는 각각 MEC 서버(112), ICN 노드(110)에서 요청 노드(114)로 컨텐츠를 전달하는데 소비되는 비용을 의미하고

,

는 각각 MEC 서버(112), ICN 노드(110)에서 컨텐츠 c의 캐싱 여부를 나타내는 값이다. in Equation 5

,

Means the cost consumed in delivering the content from the MEC server 112 and the ICN node 110 to the requesting node 114, respectively.

,

is a value indicating whether content c is cached in the MEC server 112 and the ICN node 110, respectively.

An objective function is defined as shown in Equation 6 in order to minimize consumption cost when delivering content. Equations 7 to 8 are constraints of the objective function Equation 6, and Equation 7 means that at least one content should be cached in the MEC server 112 or ICN node 110, and Equation 8 is the MEC server 112 ) means that the sum of the sizes of contents cached in ) must be smaller than the storage capacity of the MEC server 112 . Each MEC server 112 determines the contents to be cached in the MEC server 112 by solving the optimization problem of Equation 6.

는 엣지 네트워크(102)와 연결된 ICN 노드(110)

에서 컨텐츠 c의 캐싱 여부를 나타내는 값이다. The consumption cost when delivering content c to all request nodes 114 is calculated as in Equation 9

is the ICN node 110 connected to the edge network 102

It is a value indicating whether content c is cached in .

An objective function is defined as shown in Equation 10 in order to minimize consumption cost when delivering content. Equations 11 to 13 are constraints of the objective function Equation 10, and Equation 11 shows that the content is transmitted to the MEC server 112, the ICN node 110 connected to the edge network 102, or other ICN nodes 110 at least one Equation 12 indicates that when the content is cached in the ICN node 110 connected to the edge network 102, the MEC server 112 belonging to the edge network 102 cannot cache the corresponding content. Equation 13 indicates that the sum of the sizes of contents cached in the ICN node 110 connected to the MEC server 112 and the edge network 102 must be smaller than the storage capacity of the MEC server 112 and the ICN node 110, respectively. it means.

The SDN controller 104 may determine content to be cached in the MEC server 112 and the ICN node 110 connected to the edge network 102 by solving the optimization problem of Equation 10.

2 is a diagram illustrating a process of caching and providing content according to an embodiment of the present invention.

2 is a case where three requesting nodes (Consumer 1 to Consumer 3, 114-1 to 114-3) request content, the MEC server 112-3 located in the first and second base stations (Base Station 1/2). 1,112-2), a diagram illustrating a process of delivering content from the ICN node 110 connected to the edge network 102 including the same, the core network 100, and the SDN controller 104.

Referring to FIG. 2, the Interest packet generated by the request node 114 is transferred to the MEC server 112 located in the base station for information-centric network routing, and when the contents are cached in the MEC server 112, the MEC server ( 112) delivers the content to the requesting node.

In FIG. 2, the first request node 114-1 transmits an Interest packet to the first MEC server 112-1 (step 200), and the content cached by the first MEC server 112-1 receives the packet. to the second requesting node 114-2 (step 202).

Here, the first and second MEC servers 112-1 and 112-2 may pre-cache the content delivered from the core network 100 by calculating the expected content request frequency (step 204).

Meanwhile, when the contents are not cached in the MEC server 112, the Interest packet of the second request node 114-2 is delivered to the ICN node 110 connected to the edge network 102 through information-centric network routing. (step 206).

If the content is cached in the corresponding ICN node 110, the cached content is delivered to the second request node 114-2 (step 208).

If the content is not cached in the ICN node 110 connected to the edge network 102, the interest packet of the third request node 114-3 is delivered to the core network 102 through information-centric network routing (step 210), the ICN node in the core network transfers the content to the third request node 114-3 (step 212).

The MEC server 112 and the SDN controller 104 periodically perform content caching through caching optimization (step 214).

3 is a diagram illustrating a content caching process in the SDN controller according to the present embodiment.

Referring to FIG. 3 , the SDN controller 104 determines whether or not to cache content to the ICN node 110 connected to the edge network 102 based on the content request frequency expected from the edge network 102 calculated through content request information. It is determined whether or not (step 300).

In FIG. 3, it is considered that the content determined in step 300 is cached in the first and second MEC servers 112-1 and 112-2.

According to the decision in step 300, the corresponding content is cached in the ICN node 110 connected to the edge network 102 including the first and second MEC servers 112-1 and 112-2 (step 302), and the first and second MEC servers 112-1 and 112-2 are cached. In the second MEC servers 112-1 and 112-2, cached contents are deleted (steps 304 and 306).

As shown in FIG. 3 , determination of whether to cache content and a caching location may be made according to a preset cycle.

Content caching according to this embodiment may be performed in the processor of the MEC server 112 and the SDN controller 104 .

The processor may include a central processing unit (CPU) capable of executing a computer program or other virtual machines.

The processor is connected to a memory storing program instructions for calculating content request frequency, determining content caching through comparison with a threshold value, and the like.

The memory may include a non-volatile storage device such as a fixed hard drive or a removable storage device. The removable storage device may include a compact flash unit, a USB memory stick, and the like. The memory may also include volatile memory such as various random access memories.

Program instructions stored in the MEC server 112 according to this embodiment calculate the expected first content request frequency through the request information for the first content of the request node 114, and the calculated first content request frequency. When is greater than a preset first threshold, it is determined that the first content is cached, and the SDN controller 104 determines the second content expected through the request information for the first content within the individual edge network. When it is determined to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network by calculating the request frequency, the cached first content is deleted.

In addition, the program instructions stored in the SDN controller 104 according to the present embodiment calculate an expected content request frequency through request information for the first content within the individual edge network 102, and the calculated content request frequency. When is greater than a preset threshold, it is determined to cache the first content in an Information-Centric Networking (ICN) node 110 connected to the individual edge network 102, and the first content is stored in the ICN node. After determining to cache in , it is determined to delete the first content cached in the plurality of MEC servers belonging to the individual edge network.

The embodiments of the present invention described above have been disclosed for illustrative purposes, and those skilled in the art having ordinary knowledge of the present invention will be able to make various modifications, changes, and additions within the spirit and scope of the present invention, and such modifications, changes, and additions will be considered to fall within the scope of the following claims.

Claims (12)

As a content caching method in a 5G MEC-based information-centric network,
calculating, by each of a plurality of Multi-Access Edge Computing (MEC) servers belonging to an individual edge network, an expected first content request frequency through request information on the first content of a request node;
determining, by each of the plurality of MEC servers, to cache the first content when the calculated first content request frequency is greater than a preset first threshold;
calculating, by an SDN controller, a second content request frequency expected through request information on the first content within the individual edge network; and
The SDN controller determines to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network when the calculated second content request frequency is greater than a preset second threshold Including steps,
The step of determining that each of the plurality of MEC servers caches the first content,
determining the first content as candidate content to be cached when the calculated first content request frequency is greater than a preset first threshold; and
and determining whether to cache the candidate content to be cached by optimizing a content delivery consumption cost for the candidate content to be cached. According to claim 1,
The content caching method further comprising determining, by the SDN controller, to delete the first content cached in a plurality of MEC servers belonging to the individual edge network after determining to cache the first content in the ICN node. . According to claim 1,
The first content request frequency is calculated using an exponentially weighted moving average of the content request frequency for the first content calculated in the previous cycle and the number of requests for the first content delivered to each of the plurality of MEC servers during the current cycle. Content caching methods. According to claim 1,
The second content request frequency is the content request frequency for the first content from the plurality of MEC servers present in the individual edge network in the previous cycle and the second content request frequency transmitted to the plurality of MEC servers present in the individual edge network during the current cycle. 1 A content caching method that is calculated using an exponentially weighted moving average of the number of requests for content. According to claim 1,
If the contents requested by the requesting node are not cached in the MEC server receiving the interest packet of the requesting node and the ICN node connected to the edge network including the MEC server, the ICN node in the core network through information-centric network routing Content caching method for transmitting the content requested by the request node. delete According to claim 1,
The step of determining, by the SDN controller, to cache the first content in the ICN node,
determining the second content as candidate content to be cached in the ICN node when the calculated second content request frequency is greater than a preset second threshold; and
and determining whether to cache the candidate content to be cached by optimizing a content delivery consumption cost for the candidate content to be cached. A computer-readable recording medium storing a program for performing the method according to claim 1. As a content caching device located in a base station in a 5G MEC-based information-centric network,
processor; and
Including a memory coupled to the processor,
Calculate an expected first content request frequency through request information on the first content of the request node;
When the calculated first content request frequency is greater than a preset first threshold, determining to cache the first content;
It is determined to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network by calculating the expected second content request frequency through the request information on the first content within the individual edge network. If so, to delete the cached first content,
Store program instructions executed by the processor,
The program instructions are
To determine caching the first content to the ICN node,
When the calculated second content request frequency is greater than a preset second threshold, determining the second content as candidate content to be cached in the ICN node;
A content caching device configured to determine whether to cache the candidate content to be cached by optimizing a content delivery consumption cost for the candidate content to be cached. According to claim 9,
The first content request frequency is an exponentially weighted moving average of the content request frequency for the first content calculated in the previous cycle and the number of requests for the first content delivered to each of the plurality of MEC servers belonging to the individual edge network during the current cycle A content caching device calculated using As a content caching device connected to a core network in a 5G MEC-based information-centric network,
processor; and
Including a memory coupled to the processor,
Calculate an expected first content request frequency through request information for the first content within the individual edge network;
When the calculated first content request frequency is greater than a preset threshold, determining to cache the first content in an Information-Centric Networking (ICN) node connected to the individual edge network;
After determining to cache the first content in the ICN node, to determine to delete the first content cached in a plurality of MEC servers belonging to the individual edge network;
Store program instructions executed by the processor,
The program instructions are
When the calculated first content request frequency is greater than a preset first threshold, determining the first content as candidate content to be cached;
A content caching device configured to determine whether to cache the candidate content to be cached by optimizing a content delivery consumption cost for the candidate content to be cached. According to claim 11,
The content request frequency is the content request frequency for the first content from the plurality of MEC servers existing in the individual edge network in the previous cycle and the first content delivered to the plurality of MEC servers present in the individual edge network during the current cycle. A content caching device that is calculated using an exponentially weighted moving average of the number of requests for

Images