WO2011075899A1

WO2011075899A1 - Method, apparatus and system for implementing multiple web application requests scheduling

Info

Publication number: WO2011075899A1
Application number: PCT/CN2009/075977
Authority: WO
Inventors: 陈普
Original assignee: 华为技术有限公司
Priority date: 2009-12-24
Filing date: 2009-12-24
Publication date: 2011-06-30
Also published as: CN102217247A; CN102217247B

Abstract

A method, an apparatus and a system for implementing multiple Web application requests scheduling are disclosed by the embodiments of the present invention, which relate to the field of communication technology and are invented for processing application requests for plenty of Web sites simultaneously. The method of the present invention comprises: receiving a domain name resolution request sent from a domain name server; assigning at least one load balance server for the domain name server according to the domain name resolution request; receiving a request for assigning Web server processes sent from the load balance server which serves the domain name server; assigning at least one Web server process for the load balance server according to the request for assigning Web server processes, so that the load balance server sends an HTTP request of a user terminal to the Web server process. The embodiments of the present invention are mainly applied in a distribution network.

Description

Method, device and system for realizing multi-web application request scheduling

The present invention relates to the field of communications technologies, and in particular, to a method, device, and system for implementing multi-web application request scheduling.

Background technique

In the prior art, the technology for building a large-scale Web site based on the HTTP (Hypertext Transfer Protocol) protocol is very mature.

In the logical framework of a large-scale Web site constructed in the prior art, the user terminal side mainly refers to a user browser such as IE, Fi refox, and the like. On the server side, it includes: Doma in Name Server (DNS), Load Balance Server (LBS), Web Server (Web Server). The 匪 S is configured to receive a DNS request of the user browser, select an LBS server according to a specific algorithm, and return an IP address of the selected LBS server to the user browser. The LBS server is configured to receive an HTTP request of the user browser, select a web server for the user browser according to a specific algorithm, and forward the HTTP request to the web server. The web server is configured to process an HTTP request of the user browser, and return an HTTP processing response to the user browser after the processing is completed.

The specific algorithm includes: a Round Robin Scheduling algorithm, a Weighing Round-Robin Scheduling algorithm, and a source address dispersion. Source Hashing Scheduling algorithm, dynamic feedback load balancing algorithm, etc.

However, in the process of implementing the present invention, the inventors discovered that:

Since Web sites run on top of Web servers, leveraging existing logical architectures can only serve a limited number of Web sites. When a large number of (10,000, 100,000, or even more) Web sites run together on a Web server platform that includes several Web servers, it is difficult to deploy a large number of applications together with existing architectures. Summary of the invention

The embodiment of the invention provides a method, a device and a system for implementing multi-web application request scheduling, which can simultaneously process application requests of a large number of web sites.

The embodiment of the invention uses the following technical solutions:

A method for implementing multi-web application request scheduling, including:

Receiving a domain name resolution request sent by the domain name server;

Allocating at least one load balancing server to the domain name server according to the domain name resolution request; receiving an allocation web server process request sent by a load balancing server serving the domain name server;

And assigning, according to the allocating the web server process request, the load balancing server to the at least one web server process, so that the load balancing server sends the HTTP request of the user terminal to the web server process.

An apparatus for implementing multi-web application request scheduling includes:

a first request receiving unit, configured to receive a domain name resolution request sent by the domain name server, and a first allocation unit, configured to allocate at least one load balancing server to the domain name server according to the domain name resolution request;

a second request receiving unit, configured to receive an allocation web server process request sent by a load balancing server served by the domain name server;

a second allocation unit, configured to allocate, according to the allocating a web server process request, at least one web server process to the load balancing server, so that the load balancing server can send an HTTP request of the user terminal to the web server process.

A system for implementing multi-web application request scheduling, comprising: a domain name server, a management control unit, at least one load balancing server, and at least one web server;

The domain name server is configured to send a domain name resolution request to the management control unit, and determine, from the load balancing server allocated by the management control unit, a load balancing server served thereto; a management control unit, configured to allocate at least one load balancing server to the domain name server according to the domain name resolution request of the domain name server; receive an allocation web server process request sent by the load balancing server serving the domain name server, for the load The equalization server allocates at least one web server process;

The load balancing server is configured to select a web server process from the web server process allocated by the management control unit, and send the HTTP request of the user terminal to the selected web server process;

The web server is configured to process an HTTP request of the user terminal.

The method, device and system for implementing multi-web application request scheduling according to embodiments of the present invention, the management control unit performs scheduling management on the load balancing server according to the domain name resolution request of the domain name server, that is, from multiple load balancing servers. The domain name server selects a load balancing server capable of serving the domain name server; at the same time, the management control unit performs scheduling management on the web server process according to the request of the web server process of the load balancing server selected by the domain name server, That is, a web server process capable of processing an HTTP request of the user terminal is selected for the load balancing server from a plurality of web server processes. Therefore, with the embodiment of the present invention, it is not necessary to manually configure the correspondence between the domain name requested by the user terminal and the Web server on the load balancing server, so that the application requests of a large number of Web sites can be simultaneously processed.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following drawings will be briefly described, and the drawings in the following description are only some embodiments of the present invention. Other drawings may also be obtained from those of ordinary skill in the art in view of the drawings.

FIG. 1 is a flowchart of a method for implementing multi-web application request scheduling according to an embodiment of the present invention;

2 is a schematic diagram of an apparatus for implementing multi-web application request scheduling according to an embodiment of the present invention;

FIG. 3 is a first structural diagram of an apparatus for implementing multi-web application request scheduling according to an embodiment of the present invention; FIG. 4 is a second structural diagram of an apparatus for implementing multi-web application request scheduling according to an embodiment of the present invention; FIG. 5 is a third structural diagram of an apparatus for implementing multi-web application request scheduling according to an embodiment of the present invention; FIG. 6 is a schematic diagram of a system for implementing multi-web application request scheduling according to an embodiment of the present invention.

detailed description

BRIEF DESCRIPTION OF THE DRAWINGS The technical solutions in the embodiments of the present invention will be described in detail with reference to the accompanying drawings. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative work are within the scope of the present invention.

It should be noted that in the embodiment of the present invention, the concept of "massive" refers to 10,000, 100,000 or even more. Of course, the embodiments of the present invention can also be applied to the prior art scenarios to achieve better results.

For the application request that can process a large number of Web sites at the same time, as shown in FIG. 1 , the method for implementing multi-Web application request scheduling in the embodiment of the present invention includes the following steps:

Step 11. The management control unit receives the domain name resolution request (DNS request) sent by the domain name server. The domain name requested by the user terminal may be carried in the DNS request to facilitate the domain name resolution by the management control unit. In addition, the IP address information of the active address may be further included in the 匪S request, so that the domain name server determines the user terminal to which the domain name resolution request is sent.

Step 12: The management control unit allocates at least one load balancing server to the domain name server according to the domain name resolution request.

In an actual system, the number of load balancing servers may be 100 or more. The management control unit can actively monitor the load data and the running status of each load balancing server in real time according to the information reported by each load balancing server, and then select a load balancing server capable of providing services for the domain name server. The load data may include CPU usage, memory usage, bandwidth usage, and the like of each load balancing server.

When the load balancing server is selected, the management control unit may select according to a load balancing algorithm. For example, the management control unit may select at least one load balancing server with the lowest load according to the load condition of each load balancing server in the system, and set the IP address of the load balancing server. And the load information is stored in the load balancing server list, and then the load balancing server list is sent to the domain name server, so that the domain name server selects the load balancing server according to the load information of each load balancing server.

The load information may include a weight of the selected load balancing server. At this time, the domain name server may directly select the load balancing server according to the weight. Or the load information may include a load condition of the selected load balancing server. When the load information includes the load condition of the selected load balancing server, the domain name server calculates the weight of each load balancing server according to the load condition, and then selects the load balancing server according to the weight. Or the load information may also include the weight and load of the load balancing server selected by the management control unit, and the domain name server may still directly select the load balancing server according to the weight.

In addition, in order to avoid that after receiving the DNS request of the user terminal, the domain name server queries the management control unit for information about the load balancing server that can be provided to the service, and the domain name server may also save the load balancing server. List.

And, when the domain name resolution request carries the domain name requested by the user terminal, the management control unit may further parse the domain name requested by the user terminal after receiving the domain name resolution request. Since the mapping between the domain name requested by the user terminal and the web server needs to be cached on the load balancing server, after the management control unit parses a certain domain name A requested by the user terminal, when the user terminal requests the same domain name again, The mapping between the domain name A and the web server has been previously cached. Therefore, after the management control unit parses the domain name A again, the load balancing server can no longer cache the correspondence between the domain name A and the web server. relationship. Therefore, this eliminates the need to store a large number of domain names and Web server correspondences on the load balancing server, thereby saving storage space.

Step 1 3: The management control unit receives an allocation web server process request sent by a load balancing server served by the domain name server.

After receiving the HTTP request from the user terminal, the load balancing server selected by the domain name server analyzes the URL in the HTTP request (Un i form Res ource Loca tor , unified resource setting) The bit) information, and the actual domain name requested by the user terminal is obtained according to the HOST information in the URL (the Hos t header information of the H ttp protocol). Then, the load balancing server sends the actual domain name to the management control unit in the request of the distribution web server process.

Step 14: The management control unit allocates at least one web server process to the load balancing server according to the request for allocating a web server process, so that the load balancing server sends an HTTP request of the user terminal to the web server. process.

In a specific application, the domain name is an identifier of the Web site, and a domain name can only uniquely correspond to one Web site. Therefore, after receiving the request of the Web server process, the management control unit may first use the actual The domain name is an index, and the Web site information corresponding to the actual domain name requested by the user terminal in the request for allocating the Web server process is obtained according to the correspondence between the domain name and the Web site information stored in the management control unit. Then, based on the Web site information, it is determined whether there is a Web server process serving the Web site.

When there is a web server process serving the web site, the management control unit allocates at least one web server process to the domain name server according to the corresponding relationship between the stored web server process and the web site. When there is no web server process serving the web site, the management control unit re-allocates at least one web server process to the web site. At this time, the management control unit may select, by all the web server host systems in the system, a web server host system with the lowest load and the number of available web server processes is not zero, and request the web server host system to allocate the web site. Web server process.

It should be noted that in an actual system, there may be multiple web servers, that is, a web server host system, and each web server host system may have multiple (100 or more) web server processes running therein. Each web server process listens on a different port. Each load balancing server forwards the HTTP request of the user terminal for each web server process. Therefore, each web server process can be uniquely identified by simply obtaining the IP address of the web server host system and the port number of the internal web server process.

Therefore, the management control unit allocates a web server to the selected load balancing server. The process of providing the IP address of the web server host system capable of serving the selected load balancing server and the port number of the internal web server process to the load balancing server selected by the domain name server provides a service.

Since the management control unit may allocate one or more web server processes to the selected load balancing server, the load balancing server selected by the domain name server may select a web server process according to a load balancing algorithm. Then, the selected load balancing server may forward the HTTP request of the user terminal to the HTTP address, the CGI (Common Gateway Interface), or other manner according to the IP address of the selected Web server host system and the port number of the internal Web server process. Web server process.

It can be seen that, in the method for implementing multi-web application request scheduling according to the embodiment of the present invention, the management control unit performs scheduling management on the load balancing server according to the domain name resolution request of the domain name server, that is, from multiple load balancing servers. Selecting, for the domain name server, a load balancing server capable of serving the domain name server; and simultaneously scheduling, by the management control unit, the web server process according to the request of the web server process of the load balancing server selected by the domain name server Management, that is, selecting a web server process capable of processing an HTTP request of the user terminal for the load balancing server from a plurality of web server processes. Therefore, with the embodiment of the present invention, it is not necessary to manually configure the correspondence between the domain name requested by the user terminal and the Web server on the load balancing server, so that the application requests of a large number of Web sites can be simultaneously processed.

Considering the energy saving feature, the management control unit can also perform the following processing while processing an application request of a large number of Web sites:

(1) Perform energy-saving scheduling on the load balancing server in the system, that is, increase or decrease the number of service load balancing servers according to the total load of the servers in each load balancing system.

In this process, if the total load of all load balancing servers is high (such as the overall load exceeds 70%), the management scheduling unit will start more (a certain number) of load balancing servers. If the total load of all load balancing servers is low (eg, the overall load is less than 40%), the management scheduling unit will stop a certain number of load balancing servers. The specific implementation manner may be: when the load balancing server is insufficient, the management scheduling unit allocates the load balancing server as the load balancing server to be allocated to the domain name server when the load balancing server is allocated to the domain name server; At the time of the server, when the management scheduling unit allocates the load balancing server for the domain name server, the domain name server will not be allocated the load balancing that is to be stopped. When some load balancing servers do not process HTTP requests for a long time, to determine that these load balancing services require a stop, they can be stopped by shutting down, etc.

In addition, the IP address of the load balancing device that needs to be stopped may be allocated by the management control unit to other load balancing servers. When the management control unit allocates, it generally selects a load balancing server with a lower load for allocation. In this way, the user terminal can be prevented from continuously sending HTTP requests to the load balancing server that has stopped.

(2) Increase or decrease the number of Web server host systems for services according to the total load of each Web server host system in the system.

In a specific implementation process, the management scheduling unit calculates the overall load level after counting the load of each web server host system. If the overall load exceeds the set maximum value, the management dispatch unit will launch one or more new web server host systems. If the overall load is below the set minimum value, the management scheduling unit stops one or more web server host systems that control the low load, and transfers the web server processes running on these web server host systems to other web server host systems. . The mode of transfer may be: for all Web server processes serving the Web site on the Web server host system that need to be stopped, the management control unit will start a new Web server process on other Web server host systems, and start a new one. The number of web server processes is the same as the web server process that needs to be stopped. The new Web server process that is started will serve the Web site served by the Web server process that needs to be stopped. Then, the management control unit sets all the web server processes on the web server host system that need to be stopped to not receive the new request state. If the web server host systems that need to be stopped are not processed by the web request for a period of time, such as 1 minute, the web server host system can be shut down, such as shutdown, standby, and the like.

(3) The management control unit can also control the life cycle of each web server process. In a specific application, each web server process can serve at most one Web site (but does not necessarily have to serve a Web site). Each Web site does not have only one Web server process to serve it, but a Web site. When deployed, there can be no web server processes serving it. When a Web site just generates an access, the management scheduling unit controls a Web server process to serve the Web site. When the amount of access of the Web site increases to a certain amount, the management control unit controls a new Web server process to serve the Web site. When a Web site reaches a certain amount of data, it can be assigned the number of Web server processes by the amount of traffic per second / 50 + 1 of a Web site.

For example, when a Web site has 1 visit per second, there will be one Web server process serving the Web site; when the number of visits per second is increased to 50, two Web server processes are started. And when the traffic level of the Web site is reduced, the Management Control Unit will also reduce the number of Web server processes.

Based on the above, depending on the number of Web server processes required by the Web site, the life cycle of the Web server process itself may also be changed. When the web server process starts, it can not serve any web site. When a web site needs to increase the web server process to serve it, the management scheduling unit may control the web server process to load the web site, which leads to the web server. The process state is changed to "serving for the X site", and when the X site traffic is reduced, the management scheduling unit may control the web server process to stop the service, wherein the stop mode may be to kill the process or other means.

Through the above several processing methods, the energy saving effect is achieved by using the method described in the embodiment of the present invention.

In addition, in the method of the embodiment of the present invention, when the burden of a certain web server process is too heavy, for example, a large number of requests are processed every second or all web server processes are in operation, the management control unit may also adjust The processing status of each web server process improves processing efficiency. Specifically, the management control unit may first acquire current load information of each web server process, and then, according to the current load information, when the current load of the web server process is greater than or equal to a first preset threshold, Determining the web server process service when the utilization of the web server process is high to a certain extent Whether the web site still has other web server processes for the service. If not, assigning a web server process to the web site; if yes, reassigning the web server process when the current load of the other web server processes is greater than or equal to a second predetermined threshold, when the other web When the current load of the server process is less than the second preset threshold, the weight of the Web server process served by the Web site is adjusted as a reference for the load balancing server to select a Web server process that processes the HTTP request of the user terminal.

The first preset threshold value indicates a degree of utilization of a certain web server process, and the second preset threshold value indicates a degree of utilization of a certain web server process is low; and the first preset width is The value and the second preset threshold may be set by the system operation state monitoring personnel according to the actual operation of the system. For example, the first preset threshold may be 70%, and the second preset threshold may be 40%. .

A person skilled in the art can understand that all or part of the process of implementing the above embodiment method can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium. In execution, the flow of an embodiment of the methods as described above may be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

As shown in FIG. 2, an embodiment of the present invention further provides an apparatus for implementing multi-web application request scheduling, including:

a first request receiving unit 31, configured to receive a domain name resolution request sent by a domain name server; a first allocating unit 32, configured to allocate at least one load balancing server to the domain name server according to the domain name resolution request; 33. The method for receiving a web server process sent by the load balancing server serving the domain name server; the second allocating unit 34, configured to allocate at least one web to the load balancing server according to the request of the web server process The server process causes the load balancing server to send an HTTP request of the user terminal to the web server process.

As shown in FIG. 3, the first distribution receiving unit 32 includes: a selection module 321 configured to select at least one load balancing with the lowest load according to load conditions of each load balancing server in the system. a storage module 322, configured to store the selected IP address of the at least one load balancing server and the corresponding weight in the load balancing server list, and send a module 323, configured to send the load balancing server to the domain name server a list, such that the domain name server selects a load balancing server according to a weight of a load balancing server in the load balancing server list.

The second allocation unit 34 includes: a request parsing module 341, configured to acquire web site information corresponding to the actual domain name requested by the user terminal in the request of the web server process; and a determining module 342, configured to use the webpage according to the webpage The site information determines whether there is a web server process serving the web site; the distribution module 343 is configured to: when there is a web server process serving the web site, according to the corresponding relationship between the stored web server process and the web site The domain name server allocates at least one web server process; when there is no web server process serving the web site, selecting a web server host system for loading the web site, and requesting the web server host system to be The Web site assigns a web server process.

When the web server host system of the web site is selected, the allocation module may select a web server host system with the lowest load and the number of available web server processes is not zero from all the web server host systems in the system, and request the The web server host system allocates a web server process for the web site.

It can be seen that, by using the device of the embodiment of the present invention, the load balancing server can be scheduled and managed according to the domain name resolution request of the domain name server, that is, the domain name server can be selected for the domain name server from multiple load balancing servers. At the same time, according to the selected web server process request of the selected load balancing server, the web server process is scheduled and managed, that is, the plurality of web server processes are selected for the load balancing server to be able to process the user terminal. Web server process for HTTP requests. Therefore, with the embodiment of the present invention, it is not necessary to manually configure the correspondence between the domain name requested by the user terminal and the Web server on the load balancing server, so that the application requests of a large number of Web sites can be simultaneously processed.

In addition, the domain name resolution request may include a domain name requested by the user terminal, an IP address of the user terminal, and the like. As shown in Figure 4, in order to make it unnecessary to store a large number of domains on the load balancing server The device and the web server are configured to save the storage space. The device may further include: a domain name resolution unit 35, configured to parse the domain name requested by the user terminal carried in the domain name resolution request.

In order to achieve the energy saving effect, as shown in FIG. 4, the apparatus according to the embodiment of the present invention may further include: a control unit 36, configured to increase or decrease load balancing of services according to a total load degree of each load balancing server in the system. The number of servers; or the number of web hosting systems that increase or decrease the service based on the total load of each web host system in the system.

Or, as shown in FIG. 5, in order to improve the processing efficiency of each web server processing process, the apparatus in the embodiment of the present invention may further include: an information acquiring unit 37, configured to acquire current load information of each web server process; The determining unit 38 is configured to determine, according to the current load information, whether the web site served by the web server process has other web servers served when the current load of the web server process is greater than or equal to the first preset threshold a third allocation unit 39, configured to: when the determination result of the determining unit is negative, allocate a web server process to the web site; when the determining result of the determining unit 38 is yes, if the other web If the current load of the server process is greater than or equal to the second preset threshold, the web server process is re-allocated, and if the current load of the other web server process is less than the second preset threshold, the web service of the web site is adjusted. The weight of the server process.

The meanings of the first preset threshold and the second preset threshold are the same as those in the method embodiment, and may be set by the system operation state monitoring personnel according to actual operating conditions of the system, for example, the first pre- The set value can be 70%, and the second preset value can be 40%.

In addition, an embodiment of the present invention further provides a system for implementing multi-web application request scheduling.

As shown in FIG. 6, the system includes: a domain name server 41, a management control unit 42, at least one load balancing server 43, and at least one web server 44.

The domain name server 41 is configured to send a domain name resolution request to the management control unit, and determine a load balancing server served by the management control unit for the load balancing server allocated thereto. The management control unit 42 is configured to use, according to the domain name resolution request of the domain name server, the domain The name server allocates at least one load balancing server; receiving an allocation web server process request sent by the load balancing server serving the domain name server to allocate at least one for the load balancing server

The web server process; wherein the actual domain name requested by the user terminal is carried in the request for allocating a web server process. The load balancing server 43 is configured to select a web server process from the web server process allocated by the management control unit, and send the HTTP request of the user terminal to the selected web server process. The web server 44 is configured to process an HTTP request of the user terminal.

The interface between the management control unit and the domain name server is an F1 interface, and the interface between the management control unit and the load balancing server is an F2 interface, and the interface between the management control unit and each web server is F. 3 interface.

In addition, when the management control unit allocates at least one load balancing server to the domain name server, select at least one load balancing server with the lowest load according to the load condition of each load balancing server in the system, and select the at least one selected one. The IP address and load information of the load balancing server are stored in the load balancing server list, and the load balancing server list is sent to the domain name server, and the domain name management server is further configured to save the load balancing server list.

Moreover, the domain name resolution request may include a domain name requested by the user terminal, an IP address of the user terminal, and the like. Therefore, the management control unit is further configured to parse the domain name requested by the user terminal according to the domain name resolution request. This eliminates the need to store a large number of domain names and Web server correspondences on the load balancing server, thereby saving storage space.

In order to achieve the effect of energy saving, the management control unit is further configured to increase or decrease the number of load balancing servers of the service according to the total load degree of each load balancing server in the system; or according to the web server host system of each web server in the system. The total load level, the number of web server host systems that increase or decrease the service.

The web server host system may further include a management agent unit (not shown), which is mainly used to manage the web server process, such as starting a web server process, controlling the web server process to load, uninstalling the application, and killing the web. a server process or the like; and the management agent unit is further used as a communication intermediary between the management control unit and each web server process: to each web server The process forwards the scheduling task of the management control unit, and reports the load status of the web server process to the management control unit. The interface between the management agent unit and each web server process is a T1 interface.

Therefore, in order to improve the processing efficiency of each web server processing process, the management control unit may acquire or directly obtain current load information of the web server process in each web server host system through the management agent unit in the web server host system. And the management control unit may determine, according to the current load information, whether the web site served by the web server process has other services when the current load of the web server process is greater than or equal to a first preset threshold. Web server process; if not, assigning a web server process to the web site; if yes, reassigning the web server process when the current load of the other web server processes is greater than or equal to a second predetermined threshold When the current load of the other web server process is less than the second preset threshold, the weight of the web server process served by the web site is adjusted.

As can be seen from the above, in the system for implementing multi-web application request scheduling according to the embodiment of the present invention, the management control unit performs scheduling management on the load balancing server according to the domain name resolution request of the domain name server, that is, from multiple load balancing servers. Selecting, for the domain name server, a load balancing server capable of serving the domain name server; and simultaneously scheduling, by the management control unit, the web server process according to the request of the web server process of the load balancing server selected by the domain name server Management, that is, selecting a web server process capable of processing an HTTP request of the user terminal for the load balancing server from a plurality of web server processes. Therefore, with the embodiment of the present invention, it is not necessary to manually configure the correspondence between the domain name requested by the user terminal and the Web server on the load balancing server, so that the application requests of a large number of Web sites can be simultaneously processed.

In summary, the method, device, and system for implementing multi-web application request scheduling according to embodiments of the present invention do not need to manually configure a domain name and a web server requested by a user terminal on a load balancing server. Correspondence, so that it can process application requests of a large number of Web sites at the same time.

Moreover, since the state of the load balancing server and the web server host system can be managed, energy saving effects are achieved.

The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention. It should be covered by the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.

Claims

Claim

A method for implementing a multi-web application request scheduling, the method comprising: receiving a domain name resolution request sent by a domain name server;

The method according to claim 1, wherein after the receiving the domain name resolution request sent by the domain name server, the method further includes:

Parsing the domain name requested by the user terminal carried in the domain name resolution request.

3. The method according to claim 1, wherein allocating at least one load balancing server to the domain name server comprises:

Selecting at least one load balancing server with the lowest load according to the load condition of each load balancing server in the system, and storing the selected IP address and load information of the at least one load balancing server in the load balancing server list;

Sending the load balancing server list to the domain name server, so that the domain name server selects a load balancing server according to load information of the load balancing server in the load balancing server list.

The method according to claim 3, wherein the method further comprises: the domain name server saving the load balancing server list.

The method according to claim 1, wherein the assigning the at least one web server process to the domain name server according to the allocating a web server process request comprises:

Obtaining web site information corresponding to the actual domain name requested by the user terminal in the request for allocating the web server process; Determining, according to the web site information, whether there is a web server process serving the web site;

When there is a web server process serving the web site, assigning at least one web server process to the domain name server according to the corresponding relationship between the stored web server process and the web site; when there is no service for the web site When the web server process, reassign at least one web server process to the web site.

6. The method according to claim 5, wherein when there is no web server process serving the web site, re-assigning at least one web server process to the web site comprises:

A web server host system for loading the web site is selected, and the web server host system is requested to allocate a web server process for the web site.

7. The method according to claim 6, wherein the selecting a web server host system for loading the web site comprises:

Select the Web server host system with the lowest load and the number of available Web server processes is not zero from all Web server host systems.

The method according to any one of claims 1-7, wherein the method further comprises: increasing or decreasing the number of load balancing servers of the service according to the total load level of each load balancing server.

The method according to any one of claims 1-7, wherein the method further comprises: increasing or decreasing the number of served web server host systems according to the total load level of each web server host system.

The method according to any one of claims 1 to 7, wherein the method further comprises: acquiring current load information of a web server process in each web server host system; according to the current load information, When the current load of the web server process is greater than or equal to the first preset threshold, determining whether the web server served by the web server process has other web server processes served; If not, assign a web server process to the web site;

If yes, when the current load of the other web server processes is greater than or equal to the second preset threshold, the web server process is re-allocated, when the current load of the other web server processes is less than the second preset threshold, Adjust the weight of the web server process served for the web site.

1 1. A device for implementing multi-web application request scheduling, which is characterized in that:

a second allocation unit, configured to allocate, according to the allocating a web server process request, at least one web server process to the load balancing server, so that the load balancing server sends an HTTP request of the user terminal to the web server process .

The device according to claim 1 , wherein the first allocating unit comprises: a selecting module, configured to select at least one load balancing server with the lowest load according to load conditions of each load balancing server;

a storage module, configured to store the selected IP address and load information of the at least one load balancing server in a load balancing server list;

And a sending module, configured to send the load balancing server list to the domain name server, so that the domain name server selects a load balancing server according to load information of the load balancing server in the load balancing server list.

The device according to claim 11, wherein the second allocating unit comprises: a request parsing module, configured to acquire, corresponding to an actual domain name requested by the user terminal in the request of the allocating web server process Web site information;

a determining module, configured to determine, according to the web site information, whether there is a web server process serving the web site; An allocation module, configured to allocate at least one web server process to the domain name server according to a corresponding relationship between the stored web server process and the web site when there is a web server process serving the web site; when not present for the web server When the web server process of the site service is re-allocated at least one web server process for the web site.

The device according to claim 13, wherein when there is no web server process serving the web site, the allocating module is specifically configured to select the lowest load from all web server host systems. A web server host system with a non-zero number of web server processes is available, and the web server host system is requested to allocate a web server process to the web site.

The device according to any one of claims 1 to 14, wherein the device further comprises: a domain name resolution unit, configured to perform a domain name requested by the user terminal carried in the domain name resolution request Analysis.

The device according to any one of claims 1 to 14, wherein the device further comprises: a control unit, configured to increase or decrease a service load balancing server according to a total load degree of each load balancing server; Quantity; or increase or decrease the number of served web server host systems based on the total load of each web server host system.

The device according to any one of claims 1 to 14, wherein the device further comprises: an information acquiring unit, configured to acquire current load information of a web server process in each web server host system;

a determining unit, configured to determine, according to the current load information, when the current load of the web server process is greater than or equal to a first preset threshold, determining whether the web server served by the web server process has other web server processes served ;

a third allocation unit, configured to: when the determination result of the determining unit is negative, allocate a web server process to the web site; when the determining result of the judging unit is yes, if the other web server process is current If the load is greater than or equal to the second preset threshold, the web server process is re-allocated, and if the current load of the other web server process is less than the second preset threshold, the weight of the web server process served by the web site is adjusted. .

18. A system for implementing multi-web application request scheduling, the system comprising: a domain name server, a management control unit, at least one load balancing server, and at least one web server;

The domain name server is configured to send a domain name resolution request to the management control unit, and determine, from the load balancing server allocated by the management control unit, a load balancing server served thereby;

a management control unit, configured to allocate at least one load balancing server to the domain name server according to the domain name resolution request of the domain name server; receive an allocation web server process request sent by the load balancing server serving the domain name server, for the load The equalization server allocates at least one web server process;

The web server is configured to process an HTTP request of the user terminal.

The system according to claim 18, wherein the management control unit selects at least one load balancing server for the domain name server, and selects at least the lowest load according to the load condition of each load balancing server in the system. a load balancing server, and storing the selected IP address and load information of the at least one load balancing server in a load balancing server list, and sending the load balancing server list to the domain name management server;

The domain name management server is further configured to save the load balancing server list.

The system according to claim 18 or 19, wherein the management control unit is further configured to parse a domain name requested by the user terminal carried in the domain name resolution request.

The system according to claim 18 or 19, wherein the management control unit is further configured to: acquire current load information of a web server process in each web server host system; when the current web server process is current When the load is greater than or equal to the first preset threshold, determine whether the Web server served by the Web server process has other Web server processes that serve; if not, then The web site allocates a web server process; if yes, when the current load of the other web server processes is greater than or equal to a second preset threshold, the web server process is reassigned, when the current load of the other web server processes is less than the first When the threshold is preset, the weight of the web server process served by the web site is adjusted.