KR20020094749A - Linked Group Partitioning Method and Virtual Space Service System using thereof - Google Patents

Abstract

PURPOSE: A method for LGP(Linked Group Partition) of a virtual space and a system for servicing a virtual space using the same are provided to secure a stabilization and realization by making each group has information with respect to a connection of each group divided by the virtual space, managing a group of a predetermined divided virtual space through a service host, reforming a virtual space group being managed in accordance with a load of each service host, and adjusting loads equally. CONSTITUTION: At least one LGP client(110-112) creates various kinds of command in accordance with a demand of a user, estimates a movement of an object in advance, and performs an interpolation in the case that a response of a connected network is delayed. The LGP client(110-112) divides virtual space data according to patterns, groups the data as many groups, and codes and decodes data being inputted or output. At least one LGP session server(140-142) receives/decodes various kinds of commands being output from the LGP client(110-112) through a network, manages a connection state of a user based on decoded data, classifies a virtual space data group received from the LGP client(110-112), and supplies the group. At least one LGP process server(151-152) is connected to all LGP session servers(140-142), processes the virtual space data group received from the LGP session server(140-142) independently, and performs a distributed processing of loads from the user which are pressed to each LGP session server(140-142).

Description

Linked group partitioning method and virtual space service system using

The present invention relates to a virtual space service system, and more particularly, to a linked group partition (LGP) method of a virtual space and a virtual space service system using the same.

Recently, with the continuous development of information and communication technology, an environment capable of delivering various types of information digitized, such as text, voice, image, graphics, and the like to a user by using an advanced network has been established.

Information and communication services are spreading to various service areas closely related to daily life such as information delivery, education, movies, shopping, entertainment, games, and medical care in response to the needs of users. In addition, as information and communication services have evolved into a living space that provides various information, they are changing into an extended communication environment called a virtual common living space using multimedia services. In other words, in a virtual three-dimensional space where physically remote users can coexist, they can meet and talk with each other online or share activities in daily life such as shopping, exhibition halls, meetings, travel, and game participation. A cyber society is being formed. This virtual space society has an environment of shops, exhibition halls, schools, cafes, banks, and so on, so that users can interact with their neighbors and live their daily lives using avatars that replace them. do.

This virtual space service does not provide a single host to build all the members that make up the virtual space, but rather allows multiple hosts to build some virtual space members and share them using a network to share information. do. In other words, one host will build a village bank and shop, the other host will build a museum, a library, etc., and the other hosts will build other members to work together to create a virtual village.

Representative types of information sharing models for such virtual space services include point-to-point distributed models, centralized shared server models, and multiple identical virtual space models.

FIG. 1 is a conceptual diagram illustrating a point-to-point distributed model for a virtual space service.

In FIG. 1, 11a, 11b, 11c, and 11d represent users who want to use the virtual space, and 12a, 12b, 12c, and 12d represent computers that the user uses. As shown in FIG. 1, the point-to-point variance model forms a connection with each user by using each computer 12a-12d in which each user 11a-11d holds the entire virtual space model. How to use the service.

However, in this point-to-point distributed model, there is no separate server to form the virtual space, so all users must store the virtual space model on their computer. In addition, each user should be displayed in the form of self-employment in the virtual space, so that his information should be transmitted to other users of the virtual space. To do this, one user must have a connection for transmitting information with all other users and must transmit the same information to all other users. Therefore, if n users use the virtual space service, n (n-1) / 2 connections should be formed and n-1 identical information should be transmitted to other users. As a result, this model is almost impossible to use in virtual space service that is accessed by many users at the same time because of connection complexity and redundant information transmission.

2 is a conceptual diagram illustrating a model of a centralized shared server for a virtual space service.

The centralized shared server model of FIG. 2 includes a user group 20 using a virtual space service and a virtual space service area 30 connected to the user group 20 through a network 25. The user group 20 includes a plurality of users 21a, 21b, and 21c who want to use the virtual space service, computers 23a, 23b, and 23c as interfaces for the users to use the virtual space service, and a virtual space service. Network 30 for connecting area 30 to users 21a, 21b, 21c.

Meanwhile, the virtual space service area 30 manages individual service hosts 31a, 31b, and 31c, which are in charge of their own virtual space services, and integrates these services into one virtual space service while managing these service hosts. Space sharing server 35, and a network 33 interconnecting them.

In the above configuration, the computer used by each user (21a, 21b, 21c) does not need to be provided with a virtual space model, dedicated to implement when receiving the virtual space model construction information transmitted from the virtual space sharing server 35 All you need is a browser. In addition, the user does not need to send his or her information to another user only once to send to the virtual space sharing server 35, the virtual space service automatically sends the modified information to other users. Therefore, the centralized shared server model can perform services for a large number of users, and has the advantage of minimizing the total cost of users who possess a virtual space model or transmit their information to other users.

However, this model has a problem in that a separate host is required to connect and manage individual virtual space service hosts 31a, 31b, and 31c and users, and to transfer information between them. In addition, as the number of subscribers increases, the service time for one subscriber decreases. Therefore, when one shared server exceeds the number of subscribers that can be processed in real time, it is impossible to update information in real time.

Another example of the virtual space service model to solve this problem is multiple identical virtual space models.

3 is a conceptual diagram illustrating such multiple identical virtual space models.

The multiple identical virtual space model is a concept in which multiple copies of the same virtual space service are performed and each copied virtual space service performs the same virtual space service while communicating with other services. In FIG. 3, service host A 43, service host B 44, service host C 45 and service host D 46 represent the same virtual space service that has been copied. Each of these service hosts has a plurality of users who want to use the service. These groups of users are shown in the figures as 41a, 41b, 41c, 41d.

As such, each copied virtual space service, that is, each service host 43 to 46 updates in real time information about a plurality of users belonging to it. However, the update of the information of the entire virtual space service user is performed by each shared server exchanging information, and the exchange of information of the shared server is usually not performed in real time. In this model, each copied service host (43 ~ 46) defines the contents of the service intensively, if the user wants to use his service, makes the user his own user, and if he wants to use other services. By being a user of the virtual space service, the virtual space service appears to be serviced in real time as a whole.

Among the models of the virtual space service described above, the same virtual space model is currently used the most for the traffic problem caused by the simultaneous use of multiple users or the processing of the large data. Various user interfaces are provided for spatial services to interact. In particular, in the virtual space service, when a user accesses the service, the user creates an avatar and travels through the constructed virtual space, so the user acquires necessary information while traveling through the virtual space using his own alternation. You can talk to other alter egos, shop, play games, and more. Therefore, the virtual space must be updated in real time on the monitor used by the user according to the movement of the alter ego. To best express this, it is ideal to receive all virtual space images and display them on a monitor. However, doing so would make it impossible to travel in virtual space in real time, because the amount of data to be processed is too large. To solve this problem, a browser dedicated to virtual space services has emerged. This dedicated browser has graphic data that can display virtual space such as buildings, roads, and alterations to display the virtual space, so it is possible to receive only simple information for realizing the graphic image itself from the virtual space service system. It is easy to build a virtual space.

The concept introduced to bring such a huge virtual space into a desired state is the concept of spatial partitioning. That is, since it is almost impossible to process a huge space at once when representing a virtual space, a method of dividing the virtual space into a smooth state for processing has been used. As such a space division method, a binary space partition (BSP) and quadtree method as a two-dimensional planar partitioning method and an octree method as a three-dimensional three-dimensional partitioning method have been used.

However, these conventional methods of partitioning have a loading point for connecting the partitioned spaces regardless of the method used. Therefore, it is impossible to move to another virtual space unless the loading point exists. There was a problem.

4 is a conceptual diagram to help explain an example of space division and the existence of a loading point in this conventional method. In FIG. 4, one huge virtual space 60 is divided equally into four. Each of the divided virtual spaces (divided spaces) 61, 62, 63, 64 is a loading point 61a, 61b; 61a, 61b; 62a, 62b; 63a, 63b, which is a passage that can move to its neighboring divided space. ; 64a, 64b) respectively. However, due to the existence of such a loading point in the conventional method, it is impossible to move to another space unless it reaches a specific point, that is, the loading point. This was an important factor causing inconveniences by making the user impossible to move immediately.

In addition, in the virtual space service, as the number of users increases, the load on the service host increases, so that the users are not in the same space but in different spaces. In order to overcome this problem, a separate service host is provided for each zone, but it becomes impossible to express continuous virtual space without equal load balancing of users. In addition, the loss of part of the virtual space occurs when the service host in charge of a zone goes down.

The present invention is to solve the above-described problems, by having each group has information about the connection state of each divided group when the virtual space is divided, it is possible to divide the group unnecessary loading point, arbitrary virtual space It is to provide a virtual space service system that guarantees stability and real-time realization by reorganizing a virtual space group managed by a service host and managing the load according to the load on each service host to uniformly adjust the load.

1 is a conceptual diagram illustrating a point-to-point distribution model for a conventional virtual space service.

2 is a conceptual diagram illustrating a model of a centralized shared server for a conventional virtual space service.

3 is a conceptual diagram illustrating multiple identical virtual space models for a conventional virtual space service.

4 is an exemplary view showing an example of partitioning a virtual space according to a conventional method.

5 is an exemplary view showing an example of partitioning a virtual space according to the present invention.

6 is a model diagram of a virtual space system operated based on a group divided according to the LGP of the present invention.

Explanation of symbols on the main parts of the drawing

101, 102, 103; User 110, 111, 112; LGP Client Engine

130; Load balancers 140, 141, 142; LGP Session Server

151, 152; LGP Process Server 160; controller

The linked group partition (LGP) method of the virtual space of the present invention for achieving the above object comprises the steps of: dividing the virtual space constituting the virtual society into a plurality of groups which are the minimum units of information processing; Recording location information of a group neighboring each of the divided plurality of groups; And forming a table by creating a map for the location information.

The virtual space service system of the present invention generates various commands according to a user's request, predicts and moves the object in advance when the connected network is slowed down, and executes the interpolation. At least one LGP client grouped into a group and capable of encrypting and decrypting input and output data; At least one LGP that receives and decrypts various commands output from the LGP client through a network, manages a connection state of the user based on the decrypted data, and classifies and supplies virtual space data groups received from the LGP client. A session server; At least one LGP process, each of which is connected to all the LGP session servers, independently processes the virtual space data group received from the LGP session server, and evenly distributes the load from the user on each LGP session server. It characterized in that it comprises a server.

In addition, in the above configuration, it is preferable to further install a load balancer for user distribution processing between the plurality of LGP session server and the network. If you configure the system in this way, you can automatically distribute the user without worrying about other parts.

In addition, in the above configuration, a plurality of LGP process servers monitor a state of each server, and if an abnormality occurs in any one server, a controller is additionally installed to delegate the processing task to another server. By constructing the system in this way, even if one server goes down, the server is handing over the process that the server was processing, and the processing authority is transferred to another server so that the system can be recovered smoothly.

The network used in the above-described configuration is preferably an internet network based on TCP / IP.

The above and other objects and features and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

5 is a conceptual diagram illustrating a state in which a virtual space is divided according to the present invention. As shown in FIG. 5, one huge virtual space 70 is divided into a minimum unit of information processing called Groups 71 to 79. This divided data group is referred to herein as a linked group partition (LGP). Each group has information about each other's connection status. That is, when one group 75 is taken as an example, the group 75 is located on the left side, the group 76 on the right side, the group 72 on the right side, and the group 78 on the bottom side. Is held in. Similarly, other groups 71-74 and 76-79 hold information about groups in their vicinity, so that any group can display its location. In this way, each group represents a connection state between each other, and it becomes possible to represent a huge virtual space according to the connection state, so that it is not necessary to load data for the entire virtual space at once. In other words, it is possible to select and load only the necessary area, so that it is not necessary to consider data on the virtual space that is not currently used. Therefore, the amount of data to be loaded is drastically reduced to facilitate real-time loading, and unlike the conventional loading point service, the virtual world can be experienced without interruption.

6 is a model diagram of a virtual space system operated based on a group divided according to the LGP of the present invention. In FIG. 6, reference numerals 101 to 103 denote users who want to use a virtual space service, 110 to 112 denote LGP client engines mounted on a computer used to receive a virtual space service from a user side, and 120 denotes a user and a virtual space. TCP / IP-based Internet network connecting service provider side, 130 are load balancers to properly distribute the workload required from user side, and 140 ~ 142 are servers between user and service provider server. An LGP session server in charge of management, 151 and 152 represent an LGP process server for processing data allocated by the LGP session servers 140 to 142, and 160 represent a controller for controlling the process servers 151 and 152, respectively. .

Each LGP client engine 110-112 is a browser installed to receive various services from a service provider. Basically, each LGP client engine 110-112 accesses a virtual space service according to a service request from each user 101-103. In addition, it has a function of terminating the connection with the virtual space service according to the service interruption request. Each of the LGP client engines 101 to 103 communicates with the LGP session servers 110 to 112 and performs encryption for maintaining security since the communication between them is performed through the Internet, and from the LGP session servers 110 to 112. Has the function of decrypting the received encrypted message. In addition, the LGP client engines 110 to 112 predict and execute the movement of the object in advance when the response is delayed due to network traffic, and interpolate when the predicted path is out of the actual path. It enables continuous work. In addition, each LGP client engine (110-112) divides the virtual space into a plurality of groups by the LGP to access the virtual space requested by the user.

The load balancer 120 distributes the contents of the virtual space service required by the users 101 to 103 through the respective LGP client engines 110 to 112 to the LGP session servers 140 to 142 as appropriate. In this case, the load applied from the user can be properly distributed without burdening the LGP process servers 151 and 152.

The LGP session servers 140 to 142 determine whether the user is in the connected state or the terminated state according to the access command and the end command from the user received from the LGP client engines 110 to 112 and manage the users. LGP session servers 140-142 also interpret encrypted packets and prevent packet replay. Basically, this LGP session server (140 ~ 142) is connected to all LGP process server (151, 152), according to the status of users connected to the LGP session server (140 ~ 142) to the user's data to any LGP process server You choose to send it. At this time, the amount of data sent to the LGP process server is collected and sent to minimize the burden on the internal network. In addition, since the LGP process server does not request unnecessary connection establishment request, accept or reject it, the process server can concentrate on processing only the tasks assigned to it.

Each LGP process server 151, 152 is responsible for processing any data group sent from the LGP session server 140-142 according to its type. These LGP process servers have shared properties so that each can share data with each other. In addition, the LGP process servers 151 and 152 are connected to a controller 160 for monitoring the status of these process servers.

The controller 160 monitors the statuses of the LGP process servers 151 and 152 to adjust the load between these servers to be equal, and if one server goes down, transfers the processes processed by the server to another server. Ensure there is no interruption.

Preferred embodiments of the present invention described above are disclosed for purposes of illustration, and those skilled in the art will be able to make various modifications, changes, substitutions and additions through the spirit and scope of the present invention as set forth in the appended claims.

According to the division of the virtual space by the LGP of the present invention as described above, only the location information of each divided group needs to be grasped, so that it is easy to manage a huge virtual space and express it in real time.

In addition, according to the virtual space service system of the present invention using the group of the divided virtual spaces, it becomes possible to uniformly distribute processing of the user, and monitor the state in which the load is applied, and if the load is biased on one server, it is transferred to another server. As it can be transferred and handled, it is possible to obtain a stable system operation.

Claims (5)

Dividing the virtual space constituting the virtual society into a plurality of groups which are the minimum units of information processing; Recording location information of a group neighboring each of the divided plurality of groups; And Creating a table by creating a map of the location information; Linked group division method of the virtual space comprising a. Generate various commands according to the user's request, predict the movement of the object in advance when the response of the connected network is delayed, interpolate and execute, divide the virtual space data by type, group them into a plurality of groups, and input / output data At least one LGP client capable of encrypting and decrypting; At least one LGP that receives and decrypts various commands output from the LGP client through a network, manages a connection state of the user based on the decrypted data, and classifies and supplies virtual space data groups received from the LGP client. Session server; And At least one LGP process, each of which is connected to all the LGP session servers, independently processes the virtual space data group received from the LGP session server, and evenly distributes the load from the user on each LGP session server. Virtual space service system comprising a server. The method of claim 1, And a load balancer for distributed processing of a user between the plurality of LGP session servers and the network. The method of claim 1, And a controller connected to the plurality of LGP process servers to monitor the status of each server and to delegate the processing to another server when an error occurs in any one server. The method according to any one of claims 2 to 4, The network is a virtual space service system, characterized in that the Internet based on TCP / IP.