CN107547661B - Container load balancing implementation method - Google Patents
Container load balancing implementation method Download PDFInfo
- Publication number
- CN107547661B CN107547661B CN201710947703.9A CN201710947703A CN107547661B CN 107547661 B CN107547661 B CN 107547661B CN 201710947703 A CN201710947703 A CN 201710947703A CN 107547661 B CN107547661 B CN 107547661B
- Authority
- CN
- China
- Prior art keywords
- tengine
- lvs
- application
- load
- service
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention provides a container load balancing implementation method which comprises an application layer, a Tengine load layer, an LVS load layer and a registration center, wherein the Tengine load layer and the LVS load layer are arranged into a double-layer shunting architecture, so that the throughput of the whole cluster system is greatly improved, and simultaneously, rich load balancing strategies are supported; the client is provided with a unique and stable entrance through the virtual IP, and the high availability of the load entrance is ensured. The application layer allows a plurality of applications to be deployed, and each application is matched with an application agent program; the Tengine load layer allows a plurality of Tengines to be deployed, and each Tengine is matched with a Tengine agent program; the LVS load layer allows deployment of multiple LVSs, each of which is configured with one LVS agent. The agent programs of each layer are matched with the registration center to realize the service registration and discovery capability, so that the whole architecture has the capability of horizontal extension.
Description
Technical Field
The invention relates to a container load balancing implementation method of a double-layer shunt architecture.
Background
With the fire heat of the Docker container technology, the Docker technology is used by both the telecommunication industry and the Internet industry to operate the applications in a containerization mode. There are three main solutions in the industry for how large-scale containers are managed on-line systems, namely Docker Swarm, Google Kubernets and Apache messes. The invention provides a container load balancing implementation method with higher performance and high availability by aiming at the improvement of Kubernets on the aspect of implementing the load balancing of containers.
Kubernets provides Service to provide reverse proxy and load balancing capabilities for multiple application containers at the backend, but there are some disadvantages to this approach provided by kubernets:
the access provided to external applications is not unique and unstable. The Service scheme of Kubernetes provides services to the outside in a host IP and host Port mode, and if a host providing the IP is down, the client needs to select another available host IP.
Limited load balancing strategies result in few application scenarios. The Service scheme of Kubernetes only supports polling and IP _ HASH (source address HASH scheduling), and cannot forward policies based on URL/HEADER/COOKIE, etc.
The load forwarding efficiency is low, and the method is not suitable for large-scale application clustering. The Service of Kubernetes realizes request forwarding through NAT conversion of Iptables, and the forwarding mode becomes a performance bottleneck under the condition of large-scale application request load.
Aiming at the defects, the invention provides a unique and stable entrance for external application, is suitable for load balancing strategies of most application scenes and is suitable for high-performance load forwarding capacity of large-scale application clusters.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a high-performance and high-availability container load balancing implementation method, which provides a unique and stable entry for a client through a virtual IP, and implements a dual-layer load offloading request through an LVS load balancer and a Tengine load balancer.
The invention is realized by the following steps: a high-performance and high-availability container load balancing implementation method comprises an application layer, a Tengine load layer, an LVS load layer and a registration center, wherein the Tengine load layer and the LVS load layer are arranged into a double-layer shunting architecture; the application layer allows a plurality of applications to be deployed, and each application is matched with an application agent program; the Tengine load layer allows a plurality of Tengines to be deployed, and each Tengine is matched with a Tengine agent program; the LVS load layer allows a plurality of LVSs to be deployed, wherein only one LVS is an Active activated state, the other LVSs are Standby states, and each LVS is matched with one LVS agent; the agent programs of each layer are matched with the registration center to realize the service registration and discovery capability, so that the whole architecture has the capability of horizontal extension.
The application agent program carries out health monitoring on the corresponding application and updates the health state of the application agent program to a registration center in real time;
the Tengine agent program carries out health monitoring on corresponding Tengine, updates the health state of the Tengine to a registration center in real time, and simultaneously obtains the application state change of the registration center in real time so as to update the load configuration file of the Tengine, thereby ensuring that the newly added application can be added into a load list of the Tengine in time and unavailable application is removed from the load list of the Tengine in time;
the LVS agent program carries out health monitoring on the corresponding LVS and switches the Active role of the LVS in time, the healthy LVS is allowed to seize a virtual IP to provide service for the outside, and the virtual IP is used as the only entrance of the system; and the LVS agent program simultaneously acquires the change of the Tengine state of the registration center in real time so as to update the LVS load configuration file, so that the newly added Tengine can be timely added into a load list of the LVS, and unavailable Tengine can be timely removed from the LVS load list.
The application agent program comprises a health monitoring module and a service registration module:
the health monitoring module requests an application open service port in a polling mode in the application starting or running process, if the request response is successful, the application service is considered to be available, otherwise, the application service is considered to be unavailable, and the monitoring result is notified to the service registration module;
and after receiving the monitoring result, the service registration module acquires the relevant information of the application and writes the state information into the registration center.
The Tengine agent program comprises a health monitoring module, a service registration module, a service discovery module and a configuration updating module:
the health monitoring module requests a service port opened by Tengine in a polling mode in the starting or running process of the Tengine, if the request response is successful, the Tengine service is considered to be available, otherwise, the Tengine service is considered to be unavailable, and the service registration module is notified of the monitoring result;
after receiving the monitoring result, the service registration module acquires the relevant information of the Tengine and writes the state information into the registration center;
the service discovery module is connected with the registration center in a long connection mode, monitors an application state updating event pushed by the registration center in real time, and transmits application state information to the configuration updating module after receiving an application state updating event notification;
and the configuration updating module updates the load configuration file of the Tengine after receiving the application state information and calls a loading configuration command of the Tengine to trigger the configuration hot loading of the Tengine, when the configuration hot loading is reloaded, a rear-end application list of the Tengine is also updated, the newly-added application is added into the load list of the Tengine, and the unavailable application is removed from the load list of the Tengine in time.
The LVS agent program comprises a health monitoring module, a service switching module, a service discovery module and a configuration updating module:
the health monitoring module requests an LVS open service port in a polling mode in the starting or running process of the LVS, if the request response is successful, the LVS service is considered to be available, otherwise, the LVS service is considered to be unavailable, and the monitoring result is notified to the service switching module;
after receiving the monitoring result, the service switching module triggers the LVSStandby to seize the virtual IP and provides service for the LVS Active to the outside if the LVS Active becomes unavailable;
the service discovery module is connected with the registration center in a long connection mode, monitors a Tengine state updating event pushed by the registration center in real time, and transmits Tengine state information to the configuration updating module after receiving a Tengine state updating notification;
and the configuration updating module updates the load configuration file of the LVS after receiving the Tengine state information and calls a loading configuration command of the LVS to trigger the configuration hot loading of the LVS, when the configuration hot loading is reloaded, a rear-end Tengine instance list of the LVS is also updated, the newly-added Tengine is added into the load list of the LVS, and the unavailable Tengine is removed from the load list of the LVS in time.
The Tengine is a load balancer working in an application layer of a network seven-layer protocol, the LVS is a load balancer working in a transmission layer of the network seven-layer protocol, and the registration center is software for storing data, provides distributed read-write capability and supports active state pushing to a client.
The invention has the following advantages:
1. providing a unique and stable portal for external applications. The present invention uses virtual IP as the only entry to the whole system based on LVS. The service discovery module and the configuration updating module of the LVS agent program enable the LVS to have the capability of updating the available Tengine instance at the rear end in time, and the health detection module and the service switching module ensure the stability of an LVS inlet.
2. The rich load balancing strategy is suitable for most application scenarios. Based on the rich load balancing strategy provided by Tengine, the invention can determine whether to carry out load balancing and load balancing strategies according to the URL/HEADER/COOKIE/browser category and language of the seven layers of the network protocol layer, so that the whole network is more intelligent. The service discovery module and the service configuration updating module of the Tengine agent program enable the Tengine to have the capability of updating the available application instance at the back end in time, and the health monitoring and service registration module enables the Tengine to have the transverse expansion capability.
3. High performance load forwarding capability is suitable for large scale application clusters. The DR forwarding mode based on LVS of the invention utilizes the asymmetric characteristic of Internet service and can support millions of concurrent connections.
Drawings
The invention will be further described with reference to the accompanying drawings.
FIG. 1 is a diagram of a dual layer shunt architecture for the process of the present invention.
FIG. 2 is a flow chart of the working principle of the Tengine load layer in the method of the present invention.
Fig. 3 is a flow chart of the working principle of the LVS load layer in the method of the present invention.
Detailed Description
As shown in fig. 1, the high-performance and high-availability container load balancing implementation method of the present invention includes an application layer, a Tengine load layer, an LVS load layer and a registration center, where the Tengine load layer and the LVS load layer are configured as a dual-layer shunting architecture; the application layer allows a plurality of applications to be deployed, and each application is matched with an application agent program; the Tengine load layer allows a plurality of Tengines to be deployed, and each Tengine is matched with a Tengine agent program; the LVS load layer allows a plurality of LVSs to be deployed, wherein only one LVS is an Active activated state, the other LVSs are Standby states, and each LVS is matched with one LVS agent; the agent programs of each layer are matched with the registration center to realize the service registration and discovery capability, so that the whole architecture has the capability of horizontal extension.
The application agent program carries out health monitoring on the corresponding application and updates the health state of the application agent program to a registration center in real time;
the Tengine agent program carries out health monitoring on corresponding Tengine, updates the health state of the Tengine to a registration center in real time, and simultaneously obtains the application state change of the registration center in real time so as to update the load configuration file of the Tengine, thereby ensuring that the newly added application can be added into a load list of the Tengine in time and unavailable application is removed from the load list of the Tengine in time;
the LVS agent program carries out health monitoring on the corresponding LVS and switches the Active role of the LVS in time, the healthy LVS is allowed to seize a virtual IP to provide service for the outside, and the virtual IP is used as the only entrance of the system; and the LVS agent program simultaneously acquires the change of the Tengine state of the registration center in real time so as to update the LVS load configuration file, so that the newly added Tengine can be timely added into a load list of the LVS, and unavailable Tengine can be timely removed from the LVS load list.
The application agent program comprises a health monitoring module and a service registration module:
the health monitoring module requests an application open service port in a polling mode in the application starting or running process, if the request response is successful, the application service is considered to be available, otherwise, the application service is considered to be unavailable, and the monitoring result is notified to the service registration module;
and after receiving the monitoring result, the service registration module acquires the relevant information of the application and writes the state information into the registration center. The two modules can directly use the scheme provided by Kubernetes, and the KEY point of the realization is that enough detailed information is provided for the service registration module when the application container is started, wherein the information comprises IP and port of the application container, starting parameters, IP of a server where the application container is located and the like, and the service registration module writes state information into a registration center in a KEY: VALUE format, so that a Tengine load layer can acquire the state of the application from the registration center in real time.
The Tengine agent program comprises a health monitoring module, a service registration module, a service discovery module and a configuration updating module:
the health monitoring module requests a service port opened by Tengine in a polling mode in the starting or running process of the Tengine, if the request response is successful, the Tengine service is considered to be available, otherwise, the Tengine service is considered to be unavailable, and the service registration module is notified of the monitoring result;
after receiving the monitoring result, the service registration module acquires the relevant information of the Tengine and writes the state information into the registration center;
the service discovery module is connected with the registration center in a long connection mode, monitors an application state updating event pushed by the registration center in real time, and transmits application state information to the configuration updating module after receiving an application state updating event notification;
and the configuration updating module updates the load configuration file of the Tengine after receiving the application state information and calls a loading configuration command of the Tengine to trigger the configuration hot loading of the Tengine, when the configuration hot loading is reloaded, a rear-end application list of the Tengine is also updated, the newly-added application is added into the load list of the Tengine, and the unavailable application is removed from the load list of the Tengine in time.
The LVS agent program comprises a health monitoring module, a service switching module, a service discovery module and a configuration updating module:
the health monitoring module requests an LVS open service port in a polling mode in the starting or running process of the LVS, if the request response is successful, the LVS service is considered to be available, otherwise, the LVS service is considered to be unavailable, and the monitoring result is notified to the service switching module;
after receiving the monitoring result, the service switching module triggers the LVSStandby to seize the virtual IP and provides service for the LVS Active to the outside if the LVS Active becomes unavailable;
the service discovery module is connected with the registration center in a long connection mode, monitors a Tengine state updating event pushed by the registration center in real time, and transmits Tengine state information to the configuration updating module after receiving a Tengine state updating notification;
and the configuration updating module updates the load configuration file of the LVS after receiving the Tengine state information and calls a loading configuration command of the LVS to trigger the configuration hot loading of the LVS, when the configuration hot loading is reloaded, a rear-end Tengine instance list of the LVS is also updated, the newly-added Tengine is added into the load list of the LVS, and the unavailable Tengine is removed from the load list of the LVS in time.
The Tengine is a load balancer working in an application layer of a network seven-layer protocol, the LVS is a load balancer working in a transmission layer of the network seven-layer protocol, and the registration center is software for storing data, provides distributed read-write capability and supports active state pushing to a client.
Besides Tengine, each Tengine needs to be matched with a Tengine agent program, and the agent program comprises a health monitoring module, a service registration module, a service discovery module and a configuration updating module.
As shown in fig. 2, in the workflow of Tengine load layer, the health monitoring module requests a service port opened by Tengine in a polling manner, the request is performed every 3 seconds, and if the request response is successful, Tengine service is considered to be available; if the request response is not responded and the times exceed 3, the Tengine service is considered to be unavailable, and the monitoring result is notified to the service registration module. The service registration module receives the result notice of the health monitoring module, and the module acquires the relevant state information of the Tengine container, such as: and the state information is written into the registration center in a KEY-VALUE format, such as a container IP and a port of the Tengine container, starting parameters, and a server IP where the Tengine container is located. The service discovery module is connected with the registration center in a long connection mode and monitors the application state updating event pushed by the registration center in real time, and because the applications on the registration center have different types, the service discovery module of Tengine needs to monitor the applications of the specified type so as to reduce unnecessary event pushing, and transmits the application state information to the configuration updating module after receiving the notification. The configuration updating module receives the application state notification of the service discovery module, updates a load configuration file of the Tengine, calls a loading configuration command of the Tengine to trigger configuration hot loading of the Tengine, and updates a rear-end application instance list of the Tengine after configuration reloading.
Besides the LVS, each LVS needs to be matched with an LVS agent, and the LVS agent includes a health monitoring module, a service switching module, a service discovery module, and a configuration update module.
As shown in fig. 3, in the working process of the LVS load layer, the health monitoring module requests the service port opened by the LVS in a polling manner, the request is performed once every 3 seconds, if the request response is successful, the LVS service is considered to be available, and if the request response is not responded and the number of times exceeds 3, the LVS service is considered to be unavailable, and the monitoring result is notified to the service switching module. When the service switching module receives the result notification of the health monitoring module, if the LVS Active becomes unavailable, all the LVS Standby can be triggered to preempt the virtual IP, the LVS Standby which successfully preempts the virtual IP can become a LVS Active role to provide services to the outside, the originally unavailable LVS Active can be set to be in an unavailable state until the LVS Active becomes available through some way, such as manual intervention, the role then becomes the LVS Standby, and the module can be realized by using KeepAlived. The service discovery module is connected with the registration center in a long connection mode, monitors the Tengine state updating event pushed by the registration center only in real time so as to reduce unnecessary event pushing, and transmits the Tengine state information to the configuration updating module after receiving the notification. The configuration updating module receives the Tengine state notification of the service discovery module, updates the load configuration file of the LVS, calls a loading configuration command of the LVS to trigger configuration hot loading of the LVS, and updates the rear-end Tengine instance list of the LVS after the configuration is reloaded, so that the newly added Tengine can be timely added into the load list of the LVS, and the unavailable Tengine can be timely removed from the load list of the LVS.
The invention is a container load balancing framework with a double-layer shunting mechanism, an LVS load layer is used as an entrance of the whole framework, the uniqueness and the stability of the entrance are ensured through a virtual IP and a keepalive, a DR mode of the LVS can support millions of concurrent connections by utilizing the asymmetric characteristic of Internet service, the throughput of the whole cluster system is greatly improved, a Tengine load layer supports rich distribution strategies of seven layers of network protocol layers to meet almost all application scenes, and the transverse expansion capability of the whole framework is ensured through a service registration and service discovery mechanism among the application layer, the Tengine load layer and the LVS load layer. The invention realizes high-performance and high-availability container load balancing.
Although specific embodiments of the invention have been described above, it will be understood by those skilled in the art that the specific embodiments described are illustrative only and are not limiting upon the scope of the invention, and that equivalent modifications and variations can be made by those skilled in the art without departing from the spirit of the invention, which is to be limited only by the appended claims.
Claims (5)
1. A container load balancing implementation method is characterized in that: the system comprises an application layer, a Tengine load layer, an LVS load layer and a registration center, wherein the Tengine load layer and the LVS load layer are arranged into a double-layer shunt architecture; the application layer allows a plurality of applications to be deployed, and each application is matched with an application agent program; the Tengine load layer allows a plurality of Tengines to be deployed, and each Tengine is matched with a Tengine agent program; the LVS load layer allows a plurality of LVSs to be deployed, wherein only one LVS is an Active activated state, the other LVSs are Standby states, and each LVS is matched with one LVS agent; the agent program of each layer is matched with the registration center to realize the service registration and discovery capability, so that the whole architecture has the capability of transverse expansion;
the application agent program carries out health monitoring on the corresponding application and updates the health state of the corresponding application to a registration center in real time;
the Tengine agent program carries out health monitoring on the corresponding Tengine, updates the health state of the corresponding Tengine to the registration center in real time, and simultaneously obtains the application state change of the registration center in real time so as to update the load configuration file of the Tengine, thereby ensuring that the newly added application can be added into a load list of the Tengine in time and unavailable application is removed from the load list of the Tengine in time;
the LVS agent program carries out health monitoring on the corresponding LVS and switches the Active role of the LVS in time, the healthy LVS is allowed to seize a virtual IP to provide service for the outside, and the virtual IP is used as the only entrance of the system; and the LVS agent program simultaneously acquires the change of the Tengine state of the registration center in real time so as to update the LVS load configuration file, so that the newly added Tengine can be timely added into a load list of the LVS, and unavailable Tengine can be timely removed from the LVS load list.
2. The method for implementing load balancing of containers according to claim 1, wherein: the application agent program comprises a health monitoring module and a service registration module:
the health monitoring module requests an application open service port in a polling mode in the application starting or running process, if the request response is successful, the application service is considered to be available, otherwise, the application service is considered to be unavailable, and the monitoring result is notified to the service registration module;
and after receiving the monitoring result, the service registration module acquires the relevant information of the application and writes the state information into the registration center.
3. The method for implementing load balancing of containers according to claim 1, wherein: the Tengine agent program comprises a health monitoring module, a service registration module, a service discovery module and a configuration updating module:
the health monitoring module requests a service port opened by Tengine in a polling mode in the starting or running process of the Tengine, if the request response is successful, the Tengine service is considered to be available, otherwise, the Tengine service is considered to be unavailable, and the service registration module is notified of the monitoring result;
after receiving the monitoring result, the service registration module acquires the relevant information of the Tengine and writes the state information into the registration center;
the service discovery module is connected with the registration center in a long connection mode, monitors an application state updating event pushed by the registration center in real time, and transmits application state information to the configuration updating module after receiving an application state updating event notification;
and the configuration updating module updates the load configuration file of the Tengine after receiving the application state information and calls a loading configuration command of the Tengine to trigger the configuration hot loading of the Tengine, when the configuration hot loading is reloaded, a rear-end application list of the Tengine is also updated, the newly-added application is added into the load list of the Tengine, and the unavailable application is removed from the load list of the Tengine in time.
4. The method for implementing load balancing of containers according to claim 1, wherein: the LVS agent program comprises a health monitoring module, a service switching module, a service discovery module and a configuration updating module:
the health monitoring module requests an LVS open service port in a polling mode in the starting or running process of the LVS, if the request response is successful, the LVS service is considered to be available, otherwise, the LVS service is considered to be unavailable, and the monitoring result is notified to the service switching module;
after receiving the monitoring result, the service switching module triggers the LVSStandby to seize the virtual IP and provides service for the LVSActive to the outside if the LVSActive becomes unavailable;
the service discovery module is connected with the registration center in a long connection mode, monitors a Tengine state updating event pushed by the registration center in real time, and transmits Tengine state information to the configuration updating module after receiving a Tengine state updating notification;
and the configuration updating module updates the load configuration file of the LVS after receiving the Tengine state information and calls a loading configuration command of the LVS to trigger the configuration hot loading of the LVS, when the configuration hot loading is reloaded, a rear-end Tengine instance list of the LVS is also updated, the newly-added Tengine is added into the load list of the LVS, and the unavailable Tengine is removed from the load list of the LVS in time.
5. The method for implementing load balancing of containers according to claim 1, wherein: the Tengine is a load balancer working in an application layer of a network seven-layer protocol, the LVS is a load balancer working in a transmission layer of the network seven-layer protocol, and the registration center is software for storing data, provides distributed read-write capability and supports active state pushing to a client.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710947703.9A CN107547661B (en) | 2017-10-12 | 2017-10-12 | Container load balancing implementation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710947703.9A CN107547661B (en) | 2017-10-12 | 2017-10-12 | Container load balancing implementation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107547661A CN107547661A (en) | 2018-01-05 |
| CN107547661B true CN107547661B (en) | 2020-05-08 |
Family
ID=60967053
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710947703.9A Active CN107547661B (en) | 2017-10-12 | 2017-10-12 | Container load balancing implementation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107547661B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108616599B (en) * | 2018-05-11 | 2021-10-29 | 北京辰森世纪科技股份有限公司 | Method and device for registering and updating application service |
| CN109413209B (en) * | 2018-12-17 | 2021-07-09 | 上海金融期货信息技术有限公司 | Dynamic load balancing method and system based on double queues and registration center |
| CN109451065B (en) * | 2018-12-26 | 2021-06-01 | 中电福富信息科技有限公司 | Soft load balancing and shunting automation system and operation method thereof |
| CN110049089A (en) * | 2019-01-09 | 2019-07-23 | 阿里巴巴集团控股有限公司 | A kind of service register center |
| CN110851269A (en) * | 2019-10-17 | 2020-02-28 | 上海易点时空网络有限公司 | Optimization method, device, equipment and storage medium for load balancing of Linux virtual server |
| CN110851166A (en) * | 2019-11-07 | 2020-02-28 | 深圳市信联征信有限公司 | User-unaware application program updating method and device and computer equipment |
| CN111193720A (en) * | 2019-12-16 | 2020-05-22 | 中国电子科技集团公司第三十研究所 | Trust service adaptation method based on security agent |
| CN112099915B (en) * | 2020-09-07 | 2022-10-25 | 紫光云(南京)数字技术有限公司 | Soft load balancing dynamic issuing configuration method and system |
| CN112087516A (en) * | 2020-09-10 | 2020-12-15 | 星辰天合（北京）数据科技有限公司 | Storage upgrading method and device based on Docker virtualization technology |
| CN113157441B (en) * | 2021-03-24 | 2023-03-14 | 北京云上曲率科技有限公司 | Distributed decentralized load balancing method and system |
| CN114553704B (en) * | 2022-02-11 | 2023-03-24 | 中国科学院软件研究所 | Method and system for supporting multiple devices to access server simultaneously to realize capacity expansion and contraction |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442493A (en) * | 2008-12-26 | 2009-05-27 | 华为技术有限公司 | Method for distributing IP message, cluster system and load equalizer |
| CN101561774A (en) * | 2009-03-31 | 2009-10-21 | 深圳创维数字技术股份有限公司 | Monitoring method and monitoring system of self-defined protocol service program |
| CN103618634A (en) * | 2013-12-11 | 2014-03-05 | 上海爱数软件有限公司 | Method for automatically finding nodes in cluster |
| CN105979009A (en) * | 2016-07-06 | 2016-09-28 | 乾云众创（北京）信息科技研究院有限公司 | Method for automatically balancing increased load of cloud application container |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130212642A1 (en) * | 2012-02-12 | 2013-08-15 | John J. Walsh | Resilient Device Authentication System |
-
2017
- 2017-10-12 CN CN201710947703.9A patent/CN107547661B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442493A (en) * | 2008-12-26 | 2009-05-27 | 华为技术有限公司 | Method for distributing IP message, cluster system and load equalizer |
| CN101561774A (en) * | 2009-03-31 | 2009-10-21 | 深圳创维数字技术股份有限公司 | Monitoring method and monitoring system of self-defined protocol service program |
| CN103618634A (en) * | 2013-12-11 | 2014-03-05 | 上海爱数软件有限公司 | Method for automatically finding nodes in cluster |
| CN105979009A (en) * | 2016-07-06 | 2016-09-28 | 乾云众创（北京）信息科技研究院有限公司 | Method for automatically balancing increased load of cloud application container |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107547661A (en) | 2018-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107547661B (en) | Container load balancing implementation method | |
| CN110113441B (en) | Computer equipment, system and method for realizing load balance | |
| US8321862B2 (en) | System for migrating a virtual machine and resource usage data to a chosen target host based on a migration policy | |
| EP4178134A1 (en) | Data transmission method, proxy server, storage medium, and electronic device | |
| CN111615066B (en) | Distributed micro-service registration and calling method based on broadcast | |
| JP5381998B2 (en) | Cluster control system, cluster control method, and program | |
| US20130007253A1 (en) | Method, system and corresponding device for load balancing | |
| US9071612B2 (en) | Service providing system | |
| US10250671B2 (en) | P2P-based file transmission control method and P2P communication control device therefor | |
| US9736235B2 (en) | Computer system, computer, and load balancing method | |
| US11665090B1 (en) | Using fast-path nodes of packet processing services as intermediaries for workload migration workflows | |
| US20140115028A1 (en) | Method and system for sharded resource brokering | |
| WO2021120633A1 (en) | Load balancing method and related device | |
| WO2017050036A1 (en) | Resource allocation information transmission and data distribution method and device | |
| CN110830574A (en) | Method for realizing intranet load balance based on docker container | |
| JP2012242994A (en) | Transaction processing system, transaction processing method and transaction processing program | |
| CN113301079A (en) | Data acquisition method, system, computing device and storage medium | |
| CN101534255A (en) | A method and device for realizing oriented processing of certain request | |
| WO2020056550A1 (en) | Service distribution device and method in software defined network, and storage medium | |
| KR101997602B1 (en) | Resource Dependency Service Method for M2M Resource Management | |
| CN110971872B (en) | Video image information acquisition method based on distributed cluster | |
| CN111711569A (en) | System and method for realizing request dynamic routing in enterprise distributed application | |
| US11477303B1 (en) | Broker cell for distributed message system | |
| WO2023109554A1 (en) | Data processing method for distributed system, and system, node and storage medium | |
| CN108259211B (en) | Equipment management method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 350000 12, 89 Avenue, Wufeng street, Gulou District, Fuzhou, Fujian. Applicant after: China Electric fufu Mdt InfoTech Ltd Address before: 350000 12, 89 Avenue, Wufeng street, Gulou District, Fuzhou, Fujian. Applicant before: Fujian Fushitong Information Software Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |