KR20180077905A - Hybrid Content Caching Method and System - Google Patents

Abstract

According to an embodiment of the present invention, provided are a hybrid content caching method and system, capable of effectively using a limited caching capacity resource in order to reduce end-to-end service time delay in a hierarchical cellular network including a plurality of user terminals, a plurality of base stations, a plurality of cloud units, and a plurality of content servers. The hybrid content caching method includes a content request receiving step, a virtual que generating step, a caching target calculating step, and a caching target deciding step.

Description

[0001] Hybrid Content Caching Method and System [

One embodiment of the present invention relates to a hybrid content caching method and system.

The following description merely provides background information related to an embodiment of the present invention and does not constitute the prior art.

Recently, mobile traffic is increasing due to the spread of various smart devices, explosive growth of the Internet of Things, augmented reality and hologram. Accordingly, a problem of overload in a backhaul connected to a core network and a fronthaul connected to an edge network is becoming a serious problem.

Caching is a technology that stores contents that users often find at a physical distance from the user. It reduces the time delay for content transmission and reduces the load on the backhaul and front holes.

In order to solve the overload problem, edge-centric network such as edge computing caching and distributed self-organizing network, cloud computing caching, cloud-RAN (cloud-RAN) two types of networks have been proposed: a radio access network (RAN) and a cloud-centric network (SDN).

1 is a conceptual diagram of a content caching system in a hierarchical network;

The caching is effective only for a small number of users because the caching is applied only to the traffic passing through the base station. If the caching is performed in a cloud unit relatively far from the user, the service time delay of many users can be reduced, but the decrease in the service time delay is relatively small because the location of the caching is farther from the base station. That is, there is a conflict between the service time delay and the number of beneficiaries.

Due to the trade-off of the location of caching, a hybrid network operating method that flexibly applies these two methods has recently been proposed.

A cellular network having a hierarchical structure, that is, a hierarchical cellular network, has an advantage in that a caching position can be adaptively determined according to a network environment. Although all nodes in different layers in a hierarchical cellular network can perform a caching function, the contents are physically distributed in different places, and there is a limit to the amount of content that nodes can cache.

Conventional caching systems acquire popularity information of a content from a content provider and determine a content to be cached based on the popularity information. However, the state of each of the wireless links between the base station and the user, the backhaul connecting the cloud unit and the core network, and the base station and the user are dynamically changed according to time and space.

In addition, since the caching of the cloud unit and the caching of the base station affect each other's caching conditions, it is more difficult to determine the caching conditions based on the popularity information.

Accordingly, there is a need for a content caching method and system that can identify packets that are actually transmitted over a network, and determine optimal caching conditions in consideration of both time and space constraints of the network environment.

An embodiment of the present invention provides a hybrid content caching method and system capable of reducing service delay time and alleviating traffic overload in a backhaul and a front hall.

According to an embodiment of the present invention, there is provided a method for providing end-to-end communication in a hierarchical cellular network including a plurality of user terminals, a plurality of base stations, a plurality of cloud units and a plurality of content servers, A caching method using a system for reducing an end-to-end time delay, the method comprising: receiving a request signal from at least one user terminal of the plurality of user terminals; A virtual queue generation process of analyzing a header of a data packet transmitted and received in the hierarchical cellular network to identify a content corresponding to the request signal and generating a virtual queue for each content; A caching object calculation step of calculating a first set of contents to be cached in the plurality of cloud units based on the virtual queue; And a caching subject determination step of calculating a second set of contents to be cached to the plurality of base stations based on the first set of contents.

According to an embodiment of the present invention there is provided a hierarchical mobile communication system comprising a plurality of user terminals, each of which is a data node, a plurality of base stations, a plurality of cloud units and a plurality of content servers, A system for reducing an end-to-end time delay in a mobile network, the system comprising: a data receiver for receiving part or all of content data transmitted and received in the hierarchical mobile communication network; A data identification unit for analyzing a header of the content data and identifying a data amount, a transmission rate, and a transmission distance at a data node involved in transmission / reception of the content data; A virtual queue generation unit for generating a virtual queue at a data node involved in transmission / reception of the data based on the data amount, the transmission rate, and the transmission distance; And a caching object calculating unit for calculating a content to be cached based on the virtual queue.

According to an embodiment of the present invention, there is provided an effect of providing a hybrid content caching method and system capable of adaptively determining content caching according to a network environment.

According to another aspect of an embodiment of the present invention, there is an effect of reducing the end-to-end delay of content transmission in a hierarchical cellular network.

According to another aspect of the present invention, there is an effect of mitigating traffic overload in backhaul and front holes included in a hierarchical cellular network.

1 is a conceptual diagram of a content caching system in a hierarchical network;
2 illustrates a hierarchical cellular network environment for illustrating a hybrid content caching system in accordance with an embodiment of the present invention.
FIG. 3 illustrates a virtual queue model applied to a hybrid content caching system according to an embodiment of the present invention.
4 is a conceptual diagram of a hybrid content caching system according to an embodiment of the present invention.
5 is a flowchart of a hybrid content caching method according to an embodiment of the present invention.
FIG. 6A is a graph illustrating an average virtual queue backlog for a caching capacity calculated by applying the hybrid content caching method according to an exemplary embodiment of the present invention. Referring to FIG.
6B is a graph illustrating a backhaul load reduction ratio of a caching capacity calculated by applying the hybrid content caching method according to an embodiment of the present invention.
6C is a graph illustrating a front hole load reduction ratio with respect to a caching capacity calculated by applying the hybrid content caching method according to an embodiment of the present invention.
FIG. 6D is a graph illustrating an average virtual queue backlog according to a variation of an average request arrival calculated by applying the hybrid content caching method according to an exemplary embodiment of the present invention. Referring to FIG.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that, in the drawings, like reference numerals are used to denote like elements in the drawings, even if they are shown in different drawings. In the following description of the present invention, detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

The first, second, i), ii), a), b), etc. may be used in describing the components of the embodiment according to the present invention. Such a code is intended to distinguish the constituent element from other constituent elements, and the nature of the constituent element, the order or the order of the constituent element is not limited by the code. It is also to be understood that when an element is referred to as being "comprising" or "comprising", it should be understood that it does not exclude other elements unless explicitly stated to the contrary, do.

Hereinafter, a method and system for hybrid content caching according to an embodiment of the present invention will be described with reference to the accompanying drawings.

2 illustrates a hierarchical cellular network environment for illustrating a hybrid content caching system in accordance with an embodiment of the present invention. We use the term 'cellular' here, but it means 'mobile communication'.

The hierarchical cellular network environment includes a plurality of content servers 210, a plurality of cloud units 220, a plurality of base stations 230, and a plurality of end user terminals 240.

The plurality of content servers 210 form a core network and are servers that store content data (hereinafter referred to as "content data" and "content" mixedly) before distributing at various places and applying caching .

The plurality of cloud units 220 manages a plurality of base stations 230 forming an edge network on a group basis and can temporarily store contents according to the situation of the network.

A plurality of base stations 230 receives a request from a plurality of end user terminals 240 and transmits a response corresponding to the request.

Typically, a data transmission path connecting a plurality of content servers 210 and a plurality of cloud units 220 is defined as a backhaul, and data connecting between a plurality of cloud units 220 and a plurality of base stations 230 Defines the transmission path as the front hole.

The plurality of cloud units 220 or the plurality of base stations 230 may store contents according to the network conditions and deliver the contents to the plurality of end user terminals 240 without going through the backhaul.

Assume the content set F = {1, ..., F}. Here, the size of each content file f is s? S. Each content file is divided into several chunks. Also, the content file can be recovered at the destination even if all chunks corresponding to that file are not successfully transmitted.

FIG. 3 illustrates a virtual queue model applied to a hybrid content caching system according to an embodiment of the present invention.

Individual contents f belonging to the content set F are stored in a plurality of content servers 210, and are physically distributed in various places.

및

만큼의 캐싱용량을 갖는다. 일반적으로 클라우드 유닛이 기지국보다 더 큰 캐싱용량을 갖는다 (

<

).The cellular network includes a set E of cloud units and a set N of base stations that group and manage a plurality of base stations 230. We define N _e as the set of base stations to be connected in the cloud unit e ∈ε. The base station n _e and the cloud unit e are

And

Of caching capacity. Generally, a cloud unit has a larger caching capacity than a base station (

<

).

만큼에 해당하는 콘텐츠 f의 서비스 요청이 들어온다.Hybrid content caching system in accordance with one embodiment of the present invention in a unit time of a time slot (time slot) of each base station and each n _e (t = 0, 1, ...... ), every time slot t based on From connected end-user terminals

The service request of the content f corresponding to the number of the content f comes in.

The variables to control are the content to be cached in the cloud unit and base station. The caching indicator is 1 if the content f is cached in base station n _e or cloud unit e in timeslot t, 0 otherwise. The caching indicator is defined by Equation (1).

The restrictions on the caching capacity of the base station and the cloud unit are expressed by Equations (2) and (3), respectively.

For convenience of explanation, the caching indicator I will also be expressed as a vector P in Equation (4).

는 클라우드 유닛 e에 캐싱되는 콘텐츠의 집합을 나타내고,

는 기지국 n_e에 캐싱되는 콘텐츠의 집합을 나타내며 수학식 5를 이용하여 P를 구할 수 있다.here,

Represents a set of content cached in the cloud unit e,

Represents the set of contents cached at the base station n _e and P can be obtained using Equation (5).

When the content f is requested by the user, the transmission starts from the closest one of the positions of the corresponding base station, the cloud unit, and the content server holding the content f, and is determined based on I.

For example, if the content is not cached anywhere, the transfer of content starts in the cloud unit, if the transfer starts from that content server, is not cached elsewhere, and is only cached in the cloud unit. Here, the cloud unit becomes the source node.

Each of a plurality of content servers, a cloud unit, a base station, and an end user terminal may be a source node or a destination node. The source node is the node where the data transfer starts, and the destination node is the node where the data transfer ends.

Since the wireless channels of the backhaul, the front hall, and the base station vary with time, the amount of content that can be transmitted per time slot also differs.

만큼의 데이터를 전송할 수 있고, 소스 노드가 클라우드 유닛인 경우, 타임슬롯 t 동안

만큼의 데이터를 전송할 수 있다. 콘텐츠가 캐싱되어 있지 않거나, 콘텐츠를 요청한 사용자의 위치가 콘텐츠 서버와 가장 가까운 경우, 즉, 소스 노드가 콘텐츠 서버인 경우, 타임슬롯 t 동안

만큼의 데이터를 전송할 수 있다.If the source node is a base station,

, And if the source node is a cloud unit,

Of data can be transmitted. If the content is not cached, or if the location of the user requesting the content is closest to the content server, i.e., the source node is a content server,

Of data can be transmitted.

In order to reflect the content request and the service amount for the content f and the base station n _e during the time slot t, the virtual queue of | f |

는 수학식 7로 나타낼 수 있다.here,

Can be expressed by Equation (7).

This value is determined by the caching indicator I and means the amount at which the content f is served by the base station n _e . Each base station and cloud unit can identify the content being served through a header analysis of the packets being processed through it. Here, the average virtual queue backlog reflects the average end-to-end delay of serving the corresponding content in the corresponding base station.

4 is a conceptual diagram of a hybrid content caching system according to an embodiment of the present invention.

The hybrid content caching system 400 includes a data receiving unit 410, a data identifying unit 420, a virtual queue generating unit 430, a caching target calculating unit 440, and a caching target determining unit 450). Here, each of the components may be configured on the path between the content and the plurality of end user terminals 240, and may be implemented in the form of software or hardware in the plurality of cloud units 220 or the plurality of base stations 230 It is possible.

The data receiving unit 410 receives the content data from the content server located at various places via the Internet. The data receiving unit 410 receives data transmitted and received from a plurality of cloud units 220 and a plurality of base stations 230 directly or indirectly connected to a plurality of end user terminals through a network.

The data receiving unit 410 is installed in each of the plurality of cloud units 220 and the plurality of base stations 230 and can receive data transmitted and received from individual cloud units and individual base stations.

The data receiving unit 410 may be configured to receive information on data transmitted and received by a plurality of cloud units 220 and a plurality of base stations 230 installed in a location connected via a separate network.

The data identifying unit 420 analyzes the header of the data packet received from the data receiving unit 410, identifies the contents of the data, and estimates the provided traffic. Here, the data identification unit 420 assumes the following. First, the content data transmitted from the same content server to the end user terminal has a similar data rate. Second, the physical distance between the content server and the corresponding base station has a similar relationship to the data rate between the content server and the end user terminal.

The virtual queue creation unit 430 creates virtual queue information based on the contents identified by the data identification unit 420. [ The virtual queue information includes a content transmission amount 320 that can be transmitted from the base station, a content transmission amount 330 that can be transmitted from the content server 330, a content transmission amount 340 that can be transmitted from the cloud unit, a caching capacity of the cloud unit, And at least one value.

The caching subject calculation unit 440 determines a set of contents to be cached in the cloud unit and the base station for each time slot according to a hybrid caching algorithm. The hybrid caching algorithm has a Max-Weight type as shown in Equation (8) for maximizing the sum of the products of the length of the virtual queues and the service speed for each time slot.

Equation (8) is valid under the conditions of Equations (2) and (3).

However, since solving Equation (8) is NP-hard (non-deterministic polynomial-time hard) in the system to which the hybrid caching algorithm is applied, there is no algorithm that can solve in polynomial time.

Therefore, we use a suboptimal algorithm that can obtain a solution within a polynomial time but still guarantee a certain level of optimal performance.

The caching subject calculating unit 440 calculates a caching subject on the assumption that the connected base stations will follow the hybrid content caching algorithm in accordance with the caching decision of the caching subject. Here, the caching target means a set of contents to be cached in the base station and the cloud unit.

Since the caching scheme of the base stations is changed according to the caching status of the cloud unit, the caching target calculator 440 must find the number of all cases of caching available in the cloud unit to find the caching condition that maximizes the Max-Weight problem.

However, in this method, since the caching subject calculation unit 440 can not find the number of all cases within the polynomial time, the number of cases of caching needs to be greatly reduced.

The caching subject calculating unit 440 calculates a first set of contents which is a set of contents to be cached in consideration of the number of cases in which the cloud unit caches two or less contents.

Thereafter, the caching subject calculating unit 440 calculates a second set of contents, which is a set of contents to be cached in consideration of the number of cases in which the cloud unit only caches three contents, Caching contents are extracted one by one using greedy algorithm.

The caching subject determiner 450 compares two results calculated by the caching subject calculating unit 440 to select better caching.

That is, the caching subject determiner 450 selects a set that is advantageous for caching among the first content set and the second content set calculated by the caching subject calculating unit 440.

The caching subject calculation unit 440 also calculates the content to be cached for the base station in a similar manner.

The caching subject calculating unit 440 finds (1) the largest result of calculation of Equation (9) until (2) maximizes the caching capacity of a particular base station, (2) adds the value to Equation (10) 3) Repeat the process of removing f from F to finally select the content to be cached.

5 is a flowchart of a hybrid content caching method according to an embodiment of the present invention.

The data receiving unit 410 receives at least one of a virtual queue, a content transmission amount 330 that can be transmitted from the content server, a content transmission amount 340 that can be transmitted from the cloud unit, a caching capacity of the cloud unit, (S510).

The data identifying unit 420 analyzes the header of the data packet received from the data receiving unit 410, identifies the contents of the data, and estimates the provided traffic. Here, the data identification unit 420 assumes the following. First, the content data transmitted from the same content server to the end user terminal has a similar data rate. Second, the physical distance between the content server and the corresponding base station has a similar relationship to the data rate between the content server and the end user terminal.

The data calculating unit 430 calculates a first set of contents which is a set of contents to be cached in consideration of the number of cases in which the cloud unit caches two or less contents (S520).

으로 지정한다(S530).When the base station follow the hybrid caching algorithms, data calculation unit 430 finds the cache (P ^e) of the cloud unit that maximizes equation (11) with respect to caching (P ^e) of each cloud unit, which

(S530).

로 지정한다(S540).The caching subject calculating unit 440 calculates a second set of contents which is a set of contents to be cached in consideration of the number of cases in which the cloud unit only caches three contents,

(S540).

를

로

를

로 초기화한다(S550).The caching subject calculation unit 440

To

in

To

(S550).

및

를 계산하고,

를 만족하는 f^*를 찾는다(S562).When the base station follows the hybrid caching algorithm,

And

Lt; / RTI &gt;

Look for f ^* (S562) satisfying.

에 f^*를 추가하고, 그렇지 않으면

에서 f^*를 제거한다(S564).If the expression (12) is satisfied,

F ^{* &lt}; / RTI &gt;

Removes the f ^* (S564) in the.

를 최대화하는

를 찾고, 찾은 값을

로 지정한다(S566).The caching subject calculation unit 440 calculates

To maximize

And find the value

(S566).

The caching subject determination unit 450 determines whether the expression (13) is satisfied (S568).

The caching subject determiner 450 determines whether the expression (14) is satisfied (S570) if the expression (13) is not satisfied as a result of S568.

가 되고, 그렇지 않으면 클라우드 유닛 e에 캐싱해야 할 콘텐츠는

가 된다.As a result of the step S570, when the expression (14) is satisfied, the content to be cached in the cloud unit e

, Otherwise the content to be cached in cloud unit e is

.

In order to verify the performance of the hybrid content caching method according to an embodiment of the present invention, a simulation based on a trace based on a real environment is performed.

We used the content data set provided by YouTube for the size of the content and the average transmission speed. The content data set provided by YouTube includes the IP address of the content server and the destination, the average transmission speed between the content server and the destination, and the type information of the requested content. 100 of 18750 contents were randomly selected and used for simulation.

The scenario includes a total of four cloud units and three base stations per cloud unit, and the end user terminals are 20 per base station.

The size of the timeslot is 30 minutes and each base station is set to follow round robin scheduling.

The content service request follows the Bernoulli process for each time slot and is optionally assigned to the end user terminal.

The minimum chunk unit of the content is 6 kB. The caching capacity of the cloud unit is set to 50% of the entire contents. The caching capacity of the base station is set to 10 to 20% of the entire contents.

Indicators for performance analysis are the average virtual queue backlog and the front hole / backhaul load reduction ratio (fronthaul / backhaul load reduction ratio). The time-averaged length of the virtual queue is an indicator of end-to-end time delay. The front hole / backhaul load reduction ratio represents the reduction rate of the amount of traffic flowing through the front hole / backhaul when no caching is used at all.

Table 1 shows the types and operation methods of the comparison algorithm.

algorithm How it works Exhaustive search An algorithm that compares the number of all cases to find the optimal value The IGA (independent greedy algorithm) Algorithm to find base station caching and cloud unit caching independently of each other according to greed algorithm The downward greedy algorithm (DGA) Algorithm that first determines cloud unit caching independently according to greedy algorithm, and then finds base station caching according to greedy algorithm Uownward greedy algorithm (UGA) An algorithm that first determines base station caching independently according to the greedy algorithm, and then finds the cloud unit according to the greedy algorithm RHCA (reduced HCA) Algorithm that reduces the complexity of HCA

FIG. 6A is a graph illustrating an average virtual queue backlog for a caching capacity calculated by applying the hybrid content caching method according to an exemplary embodiment of the present invention. Referring to FIG.

The time average length of the virtual queue obtained using the hybrid content caching method according to an exemplary embodiment of the present invention is approximately 99.8% of the time average length of the virtual queue obtained using the exhaustive search, Performance. The time average length of the virtual queue obtained using the Hybrid Content Caching Method (RHCA) with reduced complexity was 99.6% of the time average length value of the virtual queue obtained using the full search.

6B is a graph illustrating a backhaul load reduction ratio of a caching capacity calculated by applying the hybrid content caching method according to an embodiment of the present invention.

For all methods, the backhaul load reduction rate increases as the caching capacity increases. IGA has the lowest rate of backhaul load reduction and full search has the highest backhaul load reduction rate. When using the hybrid content caching method, the load reduction of the backhaul was at least 90%. The backhaul load reduction ratio of the hybrid content caching method with reduced complexity is the next best, and the backhole load reduction ratio using the downhill greedy algorithm is the lowest among the comparison objects.

6C is a graph illustrating a front hole load reduction ratio with respect to a caching capacity calculated by applying the hybrid content caching method according to an embodiment of the present invention.

In the case of using the hybrid content caching method according to an embodiment of the present invention, the load reduction of the front hole can be achieved at least 40%. Considering that the complexity of RHCA is much lower than that of hybrid content caching, it is noteworthy that the performance of RHCA is nearly similar to that of the hybrid content caching method.

FIG. 6D is a graph illustrating average virtual queue backlog according to a change in average contents request rate calculated by applying the hybrid content caching method according to an embodiment of the present invention. to be.

As a result of measuring the time-averaged length of the virtual queue while increasing the amount of content service requests per unit time, the hybrid content caching method has a lower value than the other comparison algorithm, It is proved to be effective in reducing the end-to-end content service time delay of the hybrid content caching method.

In FIG. 5, it is described that each process is sequentially executed, but it is not limited thereto. In other words, the present invention is not limited to the time-series order, since it is possible to change the process described in FIG. 5 and execute one or more processes in parallel.

Meanwhile, each step of the flowchart shown in FIG. 5 can be implemented as a computer-readable code in a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. That is, a computer-readable recording medium includes a magnetic storage medium (e.g., ROM, floppy disk, hard disk, etc.), an optical reading medium (e.g., CD ROM, And the like). In addition, the computer-readable recording medium may be distributed over a network-connected computer system so that computer-readable code can be stored and executed in a distributed manner.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims. Modifications and variations will be possible. Therefore, the embodiments according to the present invention are not intended to limit the scope of the technical idea of the present embodiment, but are intended to be illustrative, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of an embodiment according to the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be interpreted as being included in the scope of the embodiment of the present invention.

110, 210: a content server,
120, 220: cloud unit (cloud unit)
130 and 230: a base station,
140, 240: end user terminal
310: Amount of data requested by the end user terminal
320: Amount of content transmitted from the base station
330: The amount of content that can be transferred from the content server
340: The amount of content that can be transferred from the cloud unit
400: Hybrid Content Caching System

Claims (13)

To-end time in a hierarchical mobile network including a plurality of user terminals, a plurality of base stations, a plurality of cloud units, and a plurality of content servers. A caching method using a system for reducing delay, the method comprising:
Receiving a content request signal from at least one of the plurality of user terminals;
A virtual queue generation step of analyzing a header of a data packet transmitted and received in the hierarchical mobile communication network to identify a content corresponding to the content request signal and generating a virtual queue for each content;
A caching object calculation step of calculating a first set of contents to be cached in the plurality of cloud units based on the virtual queue; And
A caching subject confirmation step of calculating a second set of contents to be cached to the plurality of base stations based on the first set of contents;
The method comprising the steps of: The method according to claim 1,
Further comprising a content caching determination process for determining whether a content corresponding to the content request signal is cached when the content request signal is received. 3. The method of claim 2,
And a direct transmission process of transmitting the content from the plurality of content servers to the at least one user terminal if the content corresponding to the content request signal is not cached anywhere as a result of the content caching determination process A method of caching hybrid content. The method according to claim 1,
The caching subject calculation process includes:
And decreasing the number of cases in the case of caching. 5. The method of claim 4,
The caching subject calculation process includes:
And a first object calculation step of considering a number of cases in which the number of contents is cached only to two or less. 6. The method of claim 5,
The caching object identification process includes:
And a second object calculation step of repeatedly extracting the contents one by one according to a greedy algorithm in the case of caching only the number of the contents. The method according to claim 6,
The caching target determining process includes:
And selecting a smaller value by comparing the average virtual queue backlog of the two virtual queues calculated through the first object calculation process and the second object calculation process . The method according to claim 1,
The caching subject calculation process includes:
And maximizing a sum of values obtained by multiplying the virtual queue length by a service rate. The method according to claim 1,
And determining the second set of content after determining the first set of content. 10. The method of claim 9,
The caching target determining process includes:
And calculating a set of the content using a greedy algorithm until a caching capacity of the at least one base station is maximized. An end-to-end service in a hierarchical mobile network comprising a plurality of user terminals, each of which is a data node, a plurality of base stations, a plurality of cloud units and a plurality of content servers, to-end time delay, the system comprising:
A data receiving unit for receiving a part or all of contents data transmitted and received in the hierarchical mobile communication network;
A data identification unit for analyzing a header of the content data and identifying a data amount, a transmission rate, and a transmission distance at a data node involved in transmission / reception of the content data;
A virtual queue generation unit for generating a virtual queue at a data node involved in transmission / reception of the data based on the data amount, the transmission rate, and the transmission distance; And
A caching target calculating unit for calculating a content to be cached based on the virtual queue,
Wherein the hybrid content caching system comprises: 12. The method of claim 11,
The caching-
Based on the number of cases in which the number of contents to be cached is considered to be two or less, the first caching subject content is calculated, and based on the number of cases in which the number of contents to be cached is three, 2 &lt; / RTI &gt; content to be cached. 13. The method of claim 12,
Wherein a smaller value is selected by comparing a time average length of two virtual queues calculated by applying the first caching subject content and the second caching subject content.

Images