KR100376019B1 - Internet Web Server Management System and Method for Integrated Web Services - Google Patents

Abstract

The present invention relates to a method of operating an internet web server for an integrated web service that provides an effective web service through an efficient web server operation. In the method of operating an internet web server for integrated web services according to the present invention, the method of operating a web server for web services includes: receiving web service request information from a user, determining a type of the received web service, and determining, Perform the service according to the priority of the set service. Therefore, according to the present invention, the operation of a web server providing a web service on the Internet can be solved software or algorithmically by applying a traffic management function based on real-time traffic monitoring information. In response to user requirements for differentiated services, the system can provide enhanced quality services to users.

Description

Internet Web Server Management System and Method for Integrated Web Services

The present invention relates to a system and method for operating a web server on the Internet for an integrated web service that effectively operates a web server according to service provision in a web server providing a multimedia service.

Conventionally, access to a server of a client through the Internet is currently made innumerably, but a network traffic management technique is not currently used as a method for providing a high performance web service of a single system.

Here, the conventional traffic management scheme refers to a token ring scheme. Token ring is a LAN (local area network) technology developed by IBM used in the IBM cabling system. It is one of the access methods for transmission lines, LAN based on Institute of Electrical and Electronics Engineers (IEEE) 802.5, or 'Fiber Distributed Data Interface (FDDI)'.

In the token ring method, nodes such as workstations are connected in a ring in a transmission line. On the ring one special data called a token is traversed. The node that received the token can then send data, and the node that sent the data removes it from the ring when the data it sent rounds the ring, and the node with no data to send sends the token next. Send to node

This token ring method is one of the access methods of each node in the network. In the token ring method, the network is configured in a loop, and a signal representing an access right called a token goes around the network and transmits. The node acquires a token and gains control and forwards a message packet.

On the other hand, high speed web server, high capacity, and high performance have been progressed at a high speed.

However, there is a problem that the speed of the transmission link on the network side is not a fundamental solution to the traffic overload in the network.

In addition, since web services are provided only with high performance of hardware components, there is a problem in that it is not a fundamental solution for supporting multimedia Internet traffic.

The present invention has been proposed to solve the above problems, the Internet web server operating system for integrated web services to provide an efficient Internet service by applying the traffic management techniques performed on the network to the Web server operation for the Internet service and Way.

Accordingly, an object of the present invention is to provide a web server operating system for an integrated web service that can provide an efficient internet service by applying a web server operating method based on real-time traffic monitoring information in a web server.

Another object of the present invention is to provide a variety of services such as voice information, still image information, video information, etc. provided in a web server according to user demand and traffic priority to ensure the quality of service to the user and at the same time load on the web server. It is to provide web server operation system for integrated web service that can be minimized.

1 is a view showing a state change in accordance with the user access of a web server running on the Internet.

2 is a diagram illustrating a service provided corresponding to a state of a web server according to another exemplary embodiment of the present invention.

3 is a block diagram of an Internet web server operating system for integrated web services according to an embodiment of the present invention.

4 is a block diagram of an Internet web server operating system for an integrated web service according to another embodiment of the present invention.

FIG. 5A is a flowchart schematically illustrating data processing of a web server operating system to which service priority is applied when a service is requested from a user according to an exemplary embodiment of the present invention. FIG.

5B is a diagram illustrating in detail data processing of a web server operating system to which service priority is applied when a service is requested from a user according to an exemplary embodiment of the present invention.

FIG. 6A is a diagram illustrating data processing in order of a user's service request based on a web server state according to another exemplary embodiment of the present invention. FIG.

6B is a diagram illustrating in detail data processing in order of a service request of a user based on a state of a web server according to another exemplary embodiment of the present invention.

FIG. 7A is a diagram briefly showing data processing applying priority to a requested service of a user based on a state of a web server according to another preferred embodiment of the present invention; FIG.

7B and 7C are detailed views illustrating data processing in which priority is applied to a request service of a user based on a state of a web server according to another exemplary embodiment of the present invention.

<Description of the symbols for the main parts of the drawings>

100: traffic management unit 105: load monitoring unit of the web server

110: token generation unit for each traffic class 115: service scheduler

In order to achieve the above objects, according to an aspect of the present invention, in the method of operating a web server for a web service, receiving the web service request information from the user, and determines the type of the received web service, the determination result, in advance A method for operating a web server for an integrated web service that performs a corresponding service according to a set priority of a service, and an apparatus and system corresponding to the method are provided.

According to a preferred embodiment provided with a method for operating a web server for an integrated web service, the web service may be classified into a guaranteed service, a load controlled service, and an internet service.

According to another preferred embodiment provided with a method for operating a web server for an integrated web service, the service may be given a token for each requested service and performed according to the set priority for each service to which the token is granted.

In order to achieve the above objects, according to another aspect of the present invention, in the method of operating a web server for a web service, the priority is set according to the type of the web service, and if there is a web service request from the user, If there is a request for information of another kind of web service while the web service is being executed, the type of requested web service is determined to compare and determine the priority of the web service in progress and the currently requested web service. Provided are a method for operating a web server for an integrated web service that performs a service, and an apparatus and system corresponding to the method.

According to a preferred embodiment provided with a method for operating a web server for an integrated web service, the web service may be classified into a guaranteed service, a load control service, and an existing internet service.

According to another preferred embodiment provided with a method for operating a web server for an integrated web service, the service may be given a token for each requested service and performed according to the set priority for each service to which the token is granted.

In order to achieve the above objects, according to another aspect of the present invention, after setting a priority for each of a plurality of services, receiving the service request information from the user, granting tokens in accordance with the received service request information and It provides a service corresponding to the token provided, and determines whether there is a request of a service that a token of a higher priority exists during the provision of the service, and as a result, the token of the currently requested service is higher than the token of the service being provided. Higher priority, web server for integrated web services that preferentially provides users with higher priority services, continues to provide services afterwards, provides services, and then removes tokens from that service information. An operating method, apparatus and system corresponding to the method are provided.

According to a preferred embodiment provided with a method for operating a web server for an integrated web service, if the token of the service being provided is a token of the priority higher than the token of the currently requested service, the service being provided may continue to be performed. Can be.

According to another preferred embodiment provided with a web server operating method for an integrated web service, the service may be divided into a guaranteed service, a load control service and an existing Internet service.

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

1 is a view showing a state change in accordance with the user access of a web server running on the Internet.

Referring to FIG. 1, in operation 100, when the web server operates in the normal state, the web server may maintain the normal state unless an external or internal sudden change occurs. In addition, when the number of users of the web server is small in the normal state, the web server transitions to an optimal state (step 105) by internally performing traffic management and packet scheduling by using an existing Internet traffic measuring apparatus.

However, if the number of users using the web server surges in step 100, the web server transitions to a virtual congestion state 110 where user traffic exceeds the server's hardware capacity.

In step 105, the web server remains in an optimal state unless the user traffic using the web server in the optimal state does not fluctuate and the user access to the web server is no longer increased. Then, the number of users using the web server is maintained. In the event of a spike, the web server transitions from an optimal state to a steady state (step 100).

In step 110, the virtual congestion state corresponds to a state in which the web server is already unable to provide normal web services. In this case, if the user access to the web server is continuously increased, the web server is in a normal congestion state. There is no transition to (step 100) and the virtual congestion state is continued.

However, the web server internally performs a traffic management function or applies a packet scheduling method for each traffic class to transition to a normal state (step 100).

Further, in step 110, the web server transitions to the actual congestion state (step 115) when an increase in the number of web server users continues in a virtual congestion state and an unexpected failure of a hardware component occurs.

On the other hand, the web server is no longer a user due to unpredictable explosion of user traffic or failure of the web server hardware device from the normal state (step 100), the optimal state (step 105), and the virtual congestion state (step 110) described above. When a situation is reached in which a desired quality of service cannot be provided for a service request of a congestion state, a congestion state (step 115) is reached.

As illustrated in FIG. 1, the web server is in a normal state (step 100), an optimal state (step 105), a virtual congestion state (step 110), and congestion according to the increase and decrease of the number of accessing users and the traffic characteristics required by the user. There are four states that can be defined as states (step 115). Therefore, the present invention is intended to operate the web server to maintain the normal state and the optimal state as possible out of the four states defined in FIG.

2 is a diagram illustrating a service provided corresponding to a state of a web server according to another exemplary embodiment of the present invention.

Referring to FIG. 2, the web server may display a state according to a service request of a user as shown in FIG. 1, in a normal state (see 100-1), an optimal state (105-FIG. 1), and a virtual congestion state 110. 1) and the congestion state (see 115-FIG. 1), and sets the types of services that can be provided according to the divided states in advance.

In detail, the web server may be configured to include service A (guaranteed service) and service B (load service) shown in FIG. 2 when a service request is made from a user in a normal state (see 100-1) and in an optimal state (105-1). Controlled Services) and Service C (formerly Internet Services).

Here, service A is a guaranteed service, and the guaranteed service is a service defined in Internet Engineering Task Force (IETF) request for comments (RFC) 2212, which guarantees a strict quality of service (QoS). For example, it guarantees a fixed data rate, has a bounded delay, no queuing loss, etc., so it can be used for hard real-time applications or As applied to a circuit emulation, there may be an audio service or a video service.

Service B is a load controlled service, and a load controlled service is a service defined in RFC 2212. For example, there is no absolute guarantee of bandwidth, latency and packet loss as a service similar to the existing Internet service under light load, but by providing sufficient resources in the network, it provides a high transmission probability. Therefore, this service is applied to adaptive real-time applications and may be an online service such as an aircraft reservation service.

In addition, the service C may be an existing Internet service, and the existing Internet service may include a file download and an email service as a general user service request.

The web server provides the above-described service A (guaranteed service) and service B (load controlled service) when the user requests a service in a virtual congestion state 110-FIG. 1.

In a congested state, the web server provides only the service A (guaranteed service) described above when a user requests a service.

As the service that can be provided in advance according to the state of the web server is set as shown in FIG. 2, the web server provides only the service that can be provided according to a user's request, which will be described in detail with reference to FIGS. 6A and 6B. Shall be.

3 is a diagram illustrating a configuration of an Internet web server operating system for an integrated web service according to an exemplary embodiment of the present invention.

Referring to FIG. 3, an Internet web server operating system for integrated web services may include a user requesting a current service from a web server as illustrated in FIG. 3 in addition to a system configuration (for example, a receiver and an input unit) in a general server. Traffic classification unit 300 for classifying the traffic of the traffic, load monitoring device unit 305 of the web server to measure the load of the web server in real time, token corresponding to the traffic class required by the current user according to the traffic classification method (token Token generation unit 310 for each traffic class and three token buffers (one token buffer for each traffic class) are generated.Guaranteed service, controlled load service, and existing Internet (Best-) effort) A service that always prioritizes tokens with higher priority by adding priority in order of service. The scheduler unit 215 is provided.

The guaranteed service is defined in the Internet Engineering Task Force (IETF) request for comments (RFC) 2212, which guarantees a strict quality of service (QoS). For example, it guarantees a fixed data rate, has a bounded delay, no queuing loss, etc., so it can be used for hard real-time applications or As applied to a circuit emulation, there may be an audio service or a video service.

Load-controlled services are those defined in RFC 2212. For example, there is no absolute guarantee of bandwidth, latency and packet loss as a service similar to the existing Internet service under light load, but by providing sufficient resources in the network, it provides a high transmission probability. Therefore, this service is applied to adaptive real-time applications and may be an online service such as an aircraft reservation service.

In addition, the existing Internet service may be a file download, an email service, etc. as a general user service request.

According to the above principle, it is possible to provide a service that is always stable and guarantees a service quality for a service request requiring real time or having a high priority.

In this case, the real-time property refers to traffic requiring real-time processing, and the above-described service classification uses a TOS (type of service) field in an IP (Internet Protocol) packet header.

Briefly describing a web server operating method operated by such a system, the server sets priority for each of a plurality of services, receives service request information from a user, and generates tokens corresponding to the received service request information. Grant.

In addition, the server provides a service corresponding to the assigned token, and determines whether there is a request for a service having a token having a higher priority while providing the service. After this determination, if the token of the currently requested service has a higher priority than the token of the providing service, the corresponding service having a higher priority is preferentially provided to the user, and then the service being provided is continued.

Thereafter, after providing the service, the server removes the token of the corresponding service information.

As described above, the operation of the web server has been outlined, but will be described in more detail later with reference to FIG. 5B.

4 is a block diagram of an Internet web server operating system for an integrated web service according to another embodiment of the present invention.

Referring to FIG. 4, an Internet web server operating system for integrated web services may include a user requesting a current service to a web server as illustrated in FIG. 4 in addition to a system configuration (for example, a receiver and an input unit) in a general server. Traffic classification unit 400 for classifying traffic, and token generating unit 405 for granting a token in accordance with the traffic of the user requesting the service, and removing the token after completion of the service, Web The load monitoring device unit 410 of the web server measuring the load of the server in real time (the web server searches for the state of the server according to the load measured in real time by the load monitoring device).

It has three token buffers (one token buffer is managed for each traffic class) and prioritized in order of guaranteed service, controlled load service, and existing best-effort service. If there is a token of the service scheduler unit 415 is always provided to process it first.

The definition of the type of service will be described with reference to FIG. 3.

According to the above principle, it is possible to provide a service that is always stable and guarantees a service quality for a service request requiring real time or having a high priority.

In this case, the real-time property refers to traffic requiring real-time processing, and the above-described service classification uses a TOS (type of service) field in an IP (Internet Protocol) packet header.

5A is a flowchart schematically illustrating data processing of a web server operating system to which service priority is applied when a service is requested by a user according to an exemplary embodiment of the present invention.

Referring to FIG. 5A, in step 500, the web server classifies a service type into a guaranteed service, a load control service, and an existing Internet service, and sets priorities in the order of the guaranteed service, the load control service, and the existing Internet service.

In step 505, the web server receives the service request of the user, and in step 510, the web server performs the corresponding service requested by the user. For example, it receives an online service request, such as an airline reservation service, and provides service information therefor.

In step 515, if the web server generates a new service request from the user (for example, an existing low-priority Internet service such as a file download) even while performing the corresponding service requested by the user, the new user-requested service is currently present. It is determined whether the service has a higher priority than the running service.

As a result of the determination in step 515, if the new service request is a service of higher priority, the web server stops the current service and proceeds to step 510 again. However, if the service is not a higher priority, that is, a service of the same kind as the service being performed or a service having a lower priority than the service being performed, the process proceeds to step 520.

For example, referring to step 515, assume that the service currently being performed is an online service such as an air reservation, and the new service request from the user is a service request such as a file download.

According to the above assumption, an online service such as an air reservation is a load control service and corresponds to a higher priority service than an existing internet service such as file download. Therefore, as a result of the determination in step 515, an online service such as an air reservation currently being performed has a higher priority, and therefore proceeds to step 520.

However, in the above example, if the new service request from the user is a guaranteed service such as an audio service, it is higher in priority than a load control service such as air reservation, so that the current service is stopped and the process proceeds to step 510 again.

In step 520, the service is performed in the order of a preset priority, that is, a guaranteed service, a load control service, and an existing Internet service. According to the example described above, if the new service request is a service having a higher priority, the service performs a new requested service and continues the service currently being provided only when there is no service request having a higher priority than the currently provided service according to the set priority. Perform.

FIG. 5A is a brief description of the case of providing a service according to the set priority and will be described in more detail with reference to FIG. 5B.

5B is a diagram illustrating in detail data processing of a web server operating system to which a service priority is applied when a service is requested by a user according to an exemplary embodiment of the present invention.

Referring to FIG. 5B, in step 525, the web server sets a priority in advance according to the service type. That is, all kinds of services are classified into guaranteed services, load control services, and existing Internet services, and services are provided according to the priority of guaranteed services, load control services, and existing Internet services when a user requests a service.

The guaranteed service is defined in the Internet Engineering Task Force (IETF) request for comments (RFC) 2212, which guarantees a strict quality of service (QoS). For example, it guarantees a fixed data rate, has a bounded delay, no queuing loss, etc., so it can be used for hard real-time applications or As applied to a circuit emulation, there may be an audio service or a video service.

Load-controlled services are those defined in RFC 2212. For example, there is no absolute guarantee of bandwidth, latency and packet loss as a service similar to the existing Internet service under light load, but by providing sufficient resources in the network, it provides a high transmission probability. Therefore, this service is applied to adaptive real-time applications and may be an online service such as an aircraft reservation service.

In addition, the existing Internet service may be a file download, an email service, etc. as a general user service request.

In step 530, the web server receives the service request information of the user, and transmits the received service request information to the traffic classification unit 300-FIG. 3 in the server.

In step 535, the traffic classifying unit 300-see FIG. 3 classifies the traffic according to the received request information of the user. At this time, the traffic classification unit 300-FIG. 3 divides the service provided according to the service request of the user into a guaranteed service, a load control service, an existing Internet service, and the like.

The classification of such traffic in the server can be implemented by interpretation of certain fields in the packet.

For traffic classified by the traffic classification unit 300-see FIG. 3 in step 540, the token generation unit 310-see FIG. 3 for each traffic class provides tokens to the traffic corresponding to the service request. Is increased (token number 1 is increased), and the token is passed to the service scheduler unit (315-3).

In step 545, the service scheduler unit (315-FIG. 3) that manages one token buffer for each traffic class (ie, the traffic corresponding to the guaranteed service, the load control service, and the existing Internet service) preferentially assigns the guaranteed service to the guaranteed service. The token buffer is checked to determine if there is a token corresponding to the guaranteed service. In other words, it is determined whether there is a user request for the guaranteed service. If it is determined that there is a user request for the guaranteed service, the process proceeds to step 550. However, in step 545, the service scheduler unit 315-FIG. 3 examines the token buffer for the guaranteed service and determines that the token of the guaranteed service does not exist, that is, if there is no user request for the guaranteed service. do.

In operation 550, the web server provides corresponding service information about the guaranteed service according to the result determined in operation 545.

For example, a web server always preferentially provides service information corresponding to a user's request for a guaranteed service that needs to be executed in real time, such as information such as an audio service or a video service requested by the user.

In step 555, the web server determines whether the completion of the current guaranteed service (for example, an audio service or a video service) is completed, and if the guaranteed service is completed, the process proceeds to step 560. If not, however, go back to step 550.

In step 560, the service scheduler unit 315-FIG. 3 transmits a predetermined signal to the traffic class token generating unit 310-FIG. 3 as the guaranteed service is completed, and the token generating unit 310-FIG. 3), the process proceeds to step 545 to remove the token of the guaranteed service that has been performed according to the received predetermined signal, and then provide the service corresponding to the user request that is continuously received.

For example, with reference to steps 545 to 560, when a request for an audio service is received from a user, the web server determines which of the service types preset for the audio service corresponds. The determination here is made by searching the header of the packet according to the user's request.

According to the search result, the web server determines that the service request received from the user is an audio service, and the audio service corresponds to a guaranteed service. The token is allocated according to the received service request, and a token buffer that stores the token for the guaranteed service according to the preset service priority order is set using the service scheduler (see 315-FIG. 3). Search and perform the service according to the token of the existing service.

During the execution of such a guaranteed service, it is not necessary to determine whether a token of a higher priority exists, since the guaranteed service is a service of the highest priority.

In step 565, the service scheduler unit 315-see FIG. 3, examines the token buffer for the load control service (e.g., an online service such as aircraft reservation, etc.) to determine if a token of the load control service exists. As a result of the determination in step 565, if there is a token of the load control service, the flow proceeds to step 570, and if there is no token of the load control service, the flow proceeds to step 590.

In step 570, the web server provides a corresponding load control service (e.g., an online service such as an aircraft reservation) at the request of the user.

In step 575, the web server determines whether the token of the highest priority corresponding to the current user request has been granted by the token generating unit 310-see FIG. 3, and if the token of the higher priority is not granted, proceeds to step 570. If a higher priority token has been granted, go to Step 545.

For example, referring to step 575, if a request for a guaranteed service such as an audio service or a video service is received from a user while the web server is currently performing an online service such as a flight reservation (corresponding to a load control service), This is classified as a service having a higher priority than the load control service currently being executed, and stops the current load control service and performs the corresponding service according to the guaranteed service request.

In step 580, the web server determines whether the load control service (e.g., an online service such as aircraft reservation) is completed by the user's request, and proceeds to step 585 if the load control service is completed. If no, go to Step 570.

In step 585, the service scheduler unit 315-FIG. 3 transmits a predetermined signal to the traffic class token generating unit 310-FIG. 3 according to the completion of the load control service, and the token generating unit 310-traffic class by traffic class. 3), the token of the load control service is removed according to the received constant signal. Thereafter, the web server proceeds to step 545 and continues to provide the corresponding service information according to the received service request of the user.

For example, in steps 565 to 580, when a service request such as an aircraft reservation is received from a user, the web server determines which of the service types preset for the aircraft reservation service corresponds. The determination here is made by searching the header of the packet according to the user's request.

According to the search result, the web server determines that the service request received from the user is an aircraft reservation service, and the aircraft reservation service corresponds to a load control service. The token is allocated according to the received service request, and the token scheduler storing the token for the load control service according to the preset service priority order is set using the service scheduler (see 315-FIG. 3). Search for and execute the service according to the token of the existing service.

During the execution of the load control service, it is determined whether a token having a higher priority exists, since there may be a guaranteed service having a higher priority than the load control service.

Therefore, if there is a user demand for a high priority guaranteed service such as an audio service or a video service while providing an aircraft reservation service as a current load control service, the audio service or video is given priority over the aircraft reservation service currently being performed. Perform guaranteed services such as services. This guaranteed service is performed first, and then the load control service currently being executed is continued.

In step 590, the service scheduler unit (215-Fig. 2) examines the token buffer of the existing Internet service (e.g., file download or e-mail service), and a token corresponding to the existing Internet service exists according to the user's request. If yes, go to Step 595.

However, if there is no token of the existing Internet service, the flow proceeds to step 615.

After determining that the service request received from the user in step 615 does not correspond to any of the preset service types, and outputs an error message, the process proceeds to step 530 to continuously receive the service request from the user.

In step 595, the web server performs a service corresponding to an existing Internet service (eg, file download, etc.) request from the user.

In step 600, the web server determines whether a service request of a higher priority has occurred. In other words, the web server determines whether a token of a high priority corresponding to a new user request has been granted by the talk generator 310-FIG. If a higher priority token has been assigned, go to step 545; if a higher priority token has not been granted, go to step 605.

For example, the web server stops the existing Internet service and executes the corresponding service having the higher priority if there is a token having a higher priority at any time, even if the existing Internet service such as file download is in progress.

In step 605, the web server determines whether the existing Internet service is completed. If the existing Internet service is completed, the web server proceeds to step 610. If the existing Internet service is not completed, the web server proceeds to step 595 again to complete the execution of the existing Internet service. do.

In step 610, the service scheduler unit 315 to FIG. 3 transmits the token corresponding to the requested service to the traffic class token generating unit 310 to FIG. 3 as the requested service is completed among the existing Internet services. The token generation unit 310-see FIG. 3 for each traffic class removes the token corresponding to the existing Internet service. Thereafter, the web server proceeds to step 545 and continuously provides service information corresponding to the service request of the user.

For example, in steps 590 to 605, when a service request such as a file download or an e-mail is received from a user, the web server may determine which of the preset service types is set for the service such as a file download or an e-mail. To judge. The determination here is made by searching the header of the packet according to the user's request.

According to the search result, the web server determines that the service request received from the user is an email service, and this email service corresponds to an existing Internet service. The token is allocated according to the received service request, and the token scheduler storing the token for the existing Internet service is set in the service scheduler 315-FIG. Search for and execute the service according to the token of the existing service.

During the execution of the existing Internet service, it is determined whether a token having a higher priority exists, since there may be a guaranteed service or a load control service having a higher priority than the existing Internet service.

Thus, while presently providing e-mail services as an existing Internet service, when user demands for high-priority guaranteed services such as audio or video services exist (or online bookings such as aircraft reservations as high priority load control services). If there is a user demand for the service), a guaranteed service such as an audio service or a video service is given priority over an e-mail service currently being performed. This guaranteed service is performed first, and then the existing Internet service which is currently being performed is continued (even if the new requested service corresponds to the load control service).

The web server repeatedly performs the above process as long as the service request is continuously received by the user.

FIG. 6A is a diagram illustrating data processing in order of a user's service request based on a web server state according to another exemplary embodiment of the present invention.

Referring to FIG. 6A, in step 620, the web server sets a service that can be provided according to the state of the web server (see FIG. 1) as shown in FIG. 2.

In other words, it provides both guaranteed service, load control service, and existing Internet service in steady state (100-FIG. 1) and optimal state (105-FIG. 1), and in hypothetical congestion state (110-FIG. 1). It provides a guaranteed service and a load control service, and provides only a guaranteed service in a congested state (see 115-1).

In step 625, the web server receives the service request of the user, and proceeds to step 630.

In step 630, the web server measures the current load in real time to determine the steady state (100-1), optimal state (105-1), virtual congestion state (110-1) described in FIG. And then detect if it is congested (115-FIG. 1).

At this time, the above four states can be arbitrarily applied according to a constant load amount. Therefore, for convenience of explanation, it is assumed that each state according to the load, for example, loads from 0 percent to less than 30 percent is present in a steady state and optimum state, and loads of less than 70 percent are present in more than 30 sets. Congestion can be classified as congestion.

In step 635, if the web server is found in a specific web server state (e.g., virtual congestion state) in step 630, the requested service by the user is provided in the current web server state (e.g., virtual congestion state). Determine if the service is available.

As a result of the determination in step 635, if the service can be provided, the process proceeds to step 640; otherwise, the process ends.

For example, referring to step 635, if the user's requested service is an existing Internet service such as file download or e-mail, the service cannot be provided in a virtual congestion state. However, in another example, if the requested service of the user is a guaranteed service such as an audio service or a video service, it is determined that the service can be provided even if the current web server state is a virtual congestion state, and then the process proceeds to step 640.

If the service can be provided according to the status of each web server in step 640, the web server performs the services in the order in which the user's request exists (different priority is not applied to each service).

6B is a diagram illustrating data processing in detail in order of a service request of a user based on a state of a web server according to another exemplary embodiment of the present invention.

Referring to FIG. 6B, in step 645, the web server sets a service that can be provided according to the state of the web server (see FIG. 1) as shown in FIG.

In other words, it provides both guaranteed service, load control service, and existing Internet service in steady state (100-FIG. 1) and optimal state (105-FIG. 1), and in hypothetical congestion state (110-FIG. 1). It provides a guaranteed service and a load control service, and provides only a guaranteed service in a congested state (see 115-1).

The guaranteed service is defined in the Internet Engineering Task Force (IETF) request for comments (RFC) 2212, which guarantees a strict quality of service (QoS). For example, it guarantees a fixed data rate, has a bounded delay, no queuing loss, etc., so it can be used for hard real-time applications or As applied to a circuit emulation, there may be an audio service or a video service.

Load-controlled services are those defined in RFC 2212. For example, there is no absolute guarantee of bandwidth, latency and packet loss as a service similar to the existing Internet service under light load, but by providing sufficient resources in the network, it provides a high transmission probability. Therefore, this service is applied to adaptive real-time applications and may be an online service such as an aircraft reservation service.

In addition, the existing Internet service may be a file download, an email service, etc. as a general user service request.

In step 650, the web server receives the user's service request, forwards the traffic corresponding to the received service request to the traffic classifier 400-4 in the server, and the traffic classifier 400-4 After classifying the traffic corresponding to the service request of the user, the flow proceeds to step 655.

The classification of such traffic in the server can be implemented by interpretation of certain fields in the packet.

For the traffic classified by the traffic classifier 400 (refer to FIG. 4) (request service by the user) in step 655, the token generation unit for each traffic class (refer to 405-FIG. 4) is assigned to the traffic corresponding to the service request of the user. The token is granted (increasing the token number 1), and the token is delivered to the service scheduler unit (415-FIG. 4).

In operation 660, the load monitoring apparatus unit 410 may refer to the load of the current web server in real time. This is to check where the current web server state corresponds to among the four states set in step 645.

In step 665, the web server determines whether the current state is a normal state (see 100-1) or an optimal state (see 105-1) according to the result retrieved by the load monitoring device unit 410-FIG. 4 in step 660. To judge.

As a result of the determination in step 665, if the current state of the web server is in a normal state (see 100-Fig. 1) or an optimal state (105-Fig. 1), the flow proceeds to step 670. If not, go to step 685.

If the determination result in step 665 is that the current state of the web server is in a normal state (see 100-1) or in an optimal state (see 105-1), the services that can be provided are a guaranteed service, a load control service and an existing Internet service. . And the services provided are made in the order of the user's request.

In step 670, the web server examines the token buffer managed by the service scheduler unit 415-see FIG. 4, and provides the corresponding service according to the tokens given in the service request order of the user.

For example, regardless of the order of the service request of the user, the guaranteed service, the load control service, and the existing Internet service, any service request performs the requested service in the priority of time.

In operation 675, the web server determines whether the service being performed is completed. As a result of the determination in step 675, if the service is completed, the process is terminated. If the service is not completed, the process proceeds to step 680.

In step 680, the web server sets a certain time according to the service, such as 1 hour in the case of downloading a file, for a certain time for performing the service requested by the user. If it is not performed, it is determined whether the service request of another user has been exceeded because it is impossible to provide the service. If the specified time (for example, 1 hour) is exceeded, the process is terminated. If no (eg, one hour) has been exceeded, proceed to step 670.

In operation 685, the web server determines whether the current state is a virtual congestion state 110-FIG. 1 according to the result of the search by the load monitoring apparatus unit 410-FIG. 4.

As a result of the determination in step 685, if the current state of the web server is a virtual congestion state 110-see FIG. 1, the flow proceeds to step 690. If not, go to step 700.

If the current state of the web server is a virtual congestion state (see 110-Fig. 1), the services that can be provided are a guaranteed service and a load control service. These services are then provided to the user in the order of their request.

In step 690, the web server determines whether the requested service of the user is an existing Internet service. As a result of the determination in step 690, if the existing Internet service, the process proceeds to step 695, and if not the existing Internet service, proceeds to step 670.

In step 695, the web server transmits a service unavailable message to the user because the existing Internet service cannot be provided in the virtual congestion state (see 110-FIG. 1) according to the condition set in step 645, and terminates the process.

In step 700 the web server determines whether the current web server state is congested (115-1). If it is determined in step 700 that the congestion state, the flow proceeds to step 705, otherwise proceeds to step 660.

In this case, the web server may provide a guaranteed service in a congested state (see 115-1).

In step 705, the web server determines whether the requested service of the user is a guaranteed service, and if yes, proceeds to step 670, otherwise proceeds to step 695.

Thus, according to the present invention, the web server pre-sets the service which can be provided according to the specific web server state, and provides the service in the order of the service request of the user when the requested service occurs.

The process is repeatedly performed as long as a service request is received from the user.

FIG. 7A is a diagram schematically illustrating data processing applying a priority to a request service of a user based on a state of a web server according to another exemplary embodiment of the present invention.

Referring to FIG. 7A, in step 710, the web server sets a priority provided to a user between a service that can be provided and a service that can be provided according to the state of the web server (see FIG. 1) as shown in FIG. 2.

In other words, it provides both guaranteed service, load control service, and existing Internet service in steady state (100-FIG. 1) and optimal state (105-FIG. 1), and in hypothetical congestion state (110-FIG. 1). It provides a guaranteed service and a load control service, and provides only a guaranteed service in a congested state (see 115-1).

In addition, the web server classifies priorities in order of guaranteed services, load control services, and existing Internet services, and provides services to the users according to the set priorities regardless of the user's request order.

In step 715 the web server receives the user's service request and proceeds to step 720.

In operation 720, the web server retrieves a current web server state, and proceeds to operation 725.

In step 725, the web server determines whether the user request service is a service that can be provided according to the current web server state.

For example, if the current web server state is a virtual congestion state (see 110-FIG. 1), the services that can be provided are a guaranteed service and a load control service, and among these services, the guaranteed service is given priority according to the priority set in step 710. Then provide load control services. Therefore, the web server determines whether the user request service corresponds to the above-mentioned service.

In step 730, the web server provides the requested service by the user according to the determination result in step 725.

The description with reference to FIG. 7A briefly describes the contents of the present invention and will be described in more detail with reference to FIGS. 7B and 7C.

7B and 7C are detailed views illustrating data processing in which priority is applied to a request service of a user based on a state of a web server according to another exemplary embodiment of the present invention.

For convenience of explanation, the following description will be made with reference to FIGS. 7B and 7C.

Referring to FIGS. 7B and 7C, in step 735, the web server sets a priority provided to a user between a service that can be provided and a service that can be provided according to the state of the web server (see FIG. 1) as shown in FIG. 2.

In other words, it provides both guaranteed service, load control service, and existing Internet service in steady state (100-FIG. 1) and optimal state (105-FIG. 1), and in hypothetical congestion state (110-FIG. 1). It provides a guaranteed service and a load control service, and provides only a guaranteed service in a congested state (see 115-1).

At this time, the above four states can be arbitrarily applied according to a constant load amount. Therefore, for convenience of explanation, it is assumed that each state according to the load, for example, loads from 0 percent to less than 30 percent is present in a steady state and optimum state, and loads of less than 70 percent are present in more than 30 sets. Congestion can be classified as congestion.

In addition, the web server classifies priorities in the order of guaranteed services, load control services, and existing Internet services among the available services, and provides services to the users according to the set priorities regardless of the user's request order.

In step 740, the web server receives the service request of the user, forwards the traffic corresponding to the received service request to the traffic classification unit 400-4 of the web server, and the traffic classification unit 400-4 of FIG. After classifying the traffic corresponding to the service request of the user, the flow proceeds to step 745.

The classification method in the server of such traffic (request service by user) can be implemented by interpretation of a specific field in a packet.

For the traffic classified by the traffic classifier 400 (see FIG. 4) (request service by the user) in step 745, the token generation unit (405-FIG. 4) for each traffic class is assigned to the traffic corresponding to the requested service of the user. The token is granted (increasing the token number 1), and the token is delivered to the service scheduler unit (415-FIG. 4).

In operation 750, the load monitoring apparatus unit 410 may refer to the load of the current web server in real time. This is to check whether the current web server state corresponds to one of the four states set in step 735.

In step 755, the web server determines whether the current state is a normal state (see 100-1) or an optimal state (105-1) according to the result retrieved by the load monitoring device unit 410-FIG. 4 in step 750. FIG. To judge.

As a result of the determination in step 755, if the current state of the web server is a normal state (see 100-1) or an optimal state (105-1), proceed to step 760. If not, however, go to Step 785.

If the determination result in step 755 is that the current state of the web server is in a normal state (see 100-1) or an optimal state (see 105-1), the services that can be provided are a guaranteed service, a load control service and an existing Internet service. . In addition, the service provided to the user is applied with the priority of the guaranteed service, the load control service, and the existing Internet service.

The guaranteed service is defined in the Internet Engineering Task Force (IETF) request for comments (RFC) 2212, which guarantees a strict quality of service (QoS). For example, it guarantees a fixed data rate, has a bounded delay, no queuing loss, etc., so it can be used for hard real-time applications or As applied to a circuit emulation, there may be an audio service or a video service.

The load controlled service is a service defined in 'RFC 2212'. For example, there is no absolute guarantee of bandwidth, latency and packet loss as a service similar to the existing Internet service under light load, but by providing sufficient resources in the network, it provides a high transmission probability. Therefore, this service is applied to adaptive real-time applications and may be an online service such as an aircraft reservation service.

In addition, the existing Internet service may be a file download, an email service, etc. as a general user service request.

In step 760, the service scheduler (415-FIG. 4) searches for a token buffer (a token buffer corresponding to the guaranteed service, the load control service, and the existing Internet service) in which the token corresponding to the user's requested service is stored. Proceed.

In step 765, the web server determines whether the service corresponding to the token retrieved in step 760 is a guaranteed service. If the service corresponding to the retrieved token is the guaranteed service, the process proceeds to step 770; otherwise, the process proceeds to step 795.

In step 770, the web server performs a service according to a user request, that is, an audio service or a video service.

In step 775, the web server determines whether the guaranteed service performed in step 770 is completed. If the execution is completed, the web server proceeds to step 780. If the execution is not completed, the web server proceeds to step 770.

In step 780, upon completion of the corresponding guaranteed service (eg, an audio service or a video service) according to the user's request, the token to which the service is assigned is removed, and the flow proceeds to step 740. FIG.

In step 785 the web server determines if the current state is a virtual congestion state 110-see FIG. 1. The hypothetical congestion state (110-FIG. 1) here is where there is less than 70 percent load on more than 30 sets, and the services that can be provided are guaranteed services and load control services.

In step 790, the service scheduler (415-FIG. 4) in the web server searches for a token buffer in which a token corresponding to the user's requested service is stored, that is, a token buffer corresponding to the guaranteed service and a token buffer corresponding to the load control service. And proceed to step 765.

In step 795, the web server determines whether the service corresponding to the user's request is a load control service, and if yes, proceeds to step 800 if it is a load control service.

In step 800, the web server determines whether the load control service according to the determination result of step 795 is a service that can be provided in a current web server state, that is, a virtual congestion state, and proceeds to step 805 if it is a service that can be provided; Proceed to 855.

In step 805, the web server performs a load control service such as aircraft reservation that can be provided in the current server state.

In step 810, the web server determines whether the load control service currently being performed is completed, and if so, proceeds to step 815; otherwise, proceeds to step 805.

In operation 815, the token generating unit 405-4 removes the token granted to the service according to the completion of the execution of the load control service (for example, a flight reservation, etc.) by the user's request. Proceed to 740.

In step 820, the web server determines whether the current state is a congested state (115-FIG. 1). Here, the congestion state 115-FIG. 1 is a case where there is a load of 70 sets or more, and the service that can be provided is a guaranteed service.

According to the determination at step 820, the web server proceeds to step 825 if the current state is a congested state (115-FIG. 1), otherwise proceeds to step 750.

In step 825, the web server searches for a token buffer corresponding to the guaranteed service by user request, and proceeds to step 765. FIG.

In operation 830, the web server determines whether the service corresponding to the user's request is an existing Internet service, such as a file download, and if so, proceeds to step 835. If not, the web server proceeds to step 855.

In step 835, the web server determines whether the service corresponding to the user's request is an existing Internet service, such as a file download, that can be provided in the current server state, and proceeds to step 840 if the service is available, otherwise proceeds to step 855. do.

In step 840, the web server performs a service based on a user's request, that is, an existing Internet service such as file download.

In step 845, the web server determines whether the existing Internet service performed in step 840 is completed. If the execution is completed, the web server proceeds to step 850. If the execution is not completed, the web server proceeds to step 840.

In operation 850, upon completion of the corresponding Internet service (eg, file download, etc.) requested by the user, the token assigned to the service is removed and the process proceeds to operation 740.

In step 855, the web server outputs an error message corresponding to a service that cannot be provided, and proceeds to step 740.

As long as the service request is received from the user, the web server repeatedly executes the above process.

The present invention is not limited to the above embodiments, and many variations are possible by those skilled in the art within the spirit of the present invention.

According to the present invention, a method of operating a web server providing a web service on the Internet by applying a traffic management function based on real-time traffic monitoring information to support a user's demand for a web server that is only increased by an existing hardware upgrade. Can be solved software or algorithmically.

In addition, the hardware upgrade method is able to overcome the above-mentioned level by integrating the existing Asynchronous Transfer Mode (ATM) and traffic control and management methods used in the packet switched network into the web server operation. It is possible to provide improved quality services to users in response to user requirements for differentiated services for web services.

Claims (27)

In the web server operation method for a web service, Receiving web service request information from a user; Determining a type of the received web service, wherein the web service is divided into a guaranteed service, a load controlled service, and an internet service; Performing the corresponding service according to the priority of the preset service as a result of the determination; Web server operating method for an integrated web service comprising a. delete The method of claim 1, The service grants a token for each service requested, and is performed according to the set priority for each service to which the token is granted. Web server operating method for an integrated web service characterized in that. In the web server operation method for a web service, Setting a priority according to a type of web service, wherein the web service is divided into a guaranteed service, a load controlled service, and an internet service; Performing a web service when there is a web service request from a user; If there is another kind of web service information request while performing the web service, determining the type of the requested web service and comparing and determining the priority of the web service in progress and the currently requested web service; As a result of the determination, performing a web service having a high priority Web server operating method for an integrated web service comprising a. delete The method of claim 4, wherein The service grants a token for each service requested, and is performed according to the set priority for each service to which the token is granted. Web server operating method for an integrated web service characterized in that. Setting priorities for a plurality of services, respectively; Receiving service request information from a user, wherein a type of service corresponding to the service request information is divided into a guaranteed service, a load controlled service, and an internet service; Granting a token corresponding to the received service request information; Providing a service corresponding to the given token, and determining whether there is a request of a service having a token having a higher priority while providing the service; As a result of the determination, when the token of the currently requested service has a higher priority than the token of the service being provided, providing the corresponding service with a higher priority to the user first, and then continuing to provide the service; After providing the service, removing the token of the corresponding service information Web server operating method for an integrated web service comprising a. The method of claim 7, wherein If the token of the service being provided is the token of the priority level than the token of the currently requested service, further comprising the step of continuing to provide the service being provided. Web server operating method for an integrated web service characterized in that. delete Setting priorities with respect to a plurality of services according to web server states; Receiving service request information from a user; Retrieving a current web server state based on the received service request information, wherein the web server state is at least one of a normal state, an optimal state, a virtual congestion state, and a congestion state; Determining whether the received service request information is a service that can be provided according to the searched web server state; As a result of the determination, if the received service request information is a service that can be provided, providing a corresponding service according to a set priority Web server operating method for an integrated web service comprising a. delete The method of claim 11, The web server state is set using a load Web server operating method for an integrated web service characterized in that. The method of claim 10, The plurality of services may be at least one of a guaranteed service, a load control service, and an existing Internet service. Web server operating method for an integrated web service characterized in that. The method of claim 10, The priority of the plurality of services is made in order of guaranteed service, load control service, and existing Internet service. Web server operating method for an integrated web service characterized in that. The method of claim 11, The service that can be provided in the normal state or the optimal state is at least one of a guaranteed service, a load control service, and an existing Internet service. Web server operating method for an integrated web service characterized in that. The method of claim 11, The service that can be provided in the virtual congestion state is at least one of a guaranteed service and a load control service. Web server operating method for an integrated web service characterized in that. The method of claim 11, The service that can be provided in the congested state is a guaranteed service Web server operating method for an integrated web service characterized in that. The method of claim 10, Granting a token to a service corresponding to the received service request information Web server operating method for an integrated web service, characterized in that it further comprises. The method of claim 10, Removing the token corresponding to the provided service after providing the corresponding service to the user; Web server operating method for an integrated web service, characterized in that it further comprises. A memory in which a program is stored; A processor stored in the memory to execute the program Including, The processor corresponds to the program, In operating a web server for web services, Receiving web service request information from a user; Determining a type of the received web service, wherein the web service is divided into a guaranteed service, a load controlled service, and an internet service; Performing the corresponding service according to the priority of the preset service as a result of the determination; Web server operating apparatus for an integrated web service, characterized in that for executing. In the web server operating system for a web service, A receiving unit for receiving web service request information from a user; A determination unit for determining a type of the received web service, wherein the web service is divided into a guaranteed service, a load control service, and an internet service; As a result of the determination, the execution unit for performing the service according to the priority of the preset service Web server operating system for an integrated web service comprising a. A memory in which a program is stored; A processor stored in the memory to execute the program Including, The processor corresponds to the program, In operating a web server for web services, Setting a priority according to a type of web service, wherein the web service is divided into a guaranteed service, a load controlled service, and an internet service; Performing a web service when there is a web service request from a user; If there is another kind of web service information request while performing the web service, determining the type of the requested web service and comparing and determining the priority of the web service in progress and the currently requested web service; As a result of the determination, performing a web service having a high priority Web server operating apparatus for an integrated web service, characterized in that for executing. In the web server operating system for a web service, A setting unit for setting a priority according to a type of web service, wherein the web service is divided into a guaranteed service, a load control service, and an internet service; An execution unit that performs a web service when a web service request is received from a user; And Determination unit for determining the type of the requested web service to further compare the priority of the web service in progress and the currently requested web service when there is a request for other types of web service information during the web service. Including, And the performing unit further performs a web service having a high priority according to a result of the comparison determination. A memory in which a program is stored; A processor stored in the memory to execute the program Including, The processor corresponds to the program, Setting priorities for a plurality of services, wherein the types of services are divided into guaranteed services, load controlled services, and Internet services; Receiving service request information from a user; Granting a token corresponding to the received service request information; Providing a service corresponding to the given token, and determining whether there is a request of a service having a token having a higher priority while providing the service; As a result of the determination, when the token of the currently requested service has a higher priority than the token of the service being provided, providing the corresponding service with a higher priority to the user first, and then continuing to provide the service; After providing the service, removing the token of the corresponding service information Web server operating apparatus for an integrated web service, characterized in that for executing. A setting unit for setting a priority for a plurality of services; A receiving unit for receiving service request information from a user, wherein the type of service corresponding to the service request information is divided into a guaranteed service, a load controlled service, and an Internet service; A token generator for granting a token corresponding to the received service request information; A service scheduler unit for providing a service corresponding to the given token and determining whether a service request for which a token of a higher priority exists exists is provided during the provision of the service; As a result of the determination, if the token of the currently requested service has a higher priority than the token of the service being provided, the execution unit for providing the corresponding service with a higher priority to the user first, and then continuing to provide the service. Provided with The token generator is to remove the token of the service information Web server operating system for an integrated web service, characterized in that it further comprises. A memory in which a program is stored; A processor stored in the memory to execute the program Including, The processor corresponds to the program, Setting a priority according to a web server state for a plurality of services, wherein the web server state is at least one of a normal state, an optimal state, a virtual congestion state, and a congestion state; Receiving service request information from a user; Retrieving a current web server status based on the received service request information; Determining whether the received service request information is a service that can be provided according to the searched web server state; As a result of the determination, if the received service request information is a service that can be provided, providing a corresponding service according to a set priority Web server operating apparatus for an integrated web service, characterized in that for executing. A setting unit for setting a priority of a plurality of services according to a web server state; A receiving unit for receiving service request information from a user; A search unit for searching for a current web server state based on the received service request information, wherein the web server state is at least one of a normal state, an optimal state, a virtual congestion state, and a congestion state; A determination unit determining whether the received service request information is a service that can be provided according to the searched web server state; The execution unit that provides the service according to the set priority, if the received service request information is a service that can be provided as a result of the determination Web server operating system for an integrated web service comprising a.