WO2018121210A1 - Elastic scaling method and device for containerized application on paas platform - Google Patents

Abstract

Disclosed are an elastic scaling method and device for a containerized application on a PaaS platform. The method comprises: collecting, from a containerized application corresponding to a micro-service, corresponding user data; and according to a change situation of a pre-set index in the user data, scaling up or scaling down at least one containerized application for the micro-service.

Description

Method and device for shrinking containerized application of PaaS platform Technical field

The present disclosure relates to the field of communications technologies, and in particular, to a method and apparatus for shrinking a containerized application of a PaaS platform.

Background technique

With the rapid development of container technology represented by Docker, container clouds are becoming more and more concerned by the cloud computing field. PaaS (Platform-as-a-Service) is a cloud platform that rapidly develops and deploys customized applications in the software industry. It is also compatible with the development and deployment of containerized applications. The PaaS platform not only needs to complete the rapid deployment of containerized applications, but also supports the elastic scaling of the applications it carries to meet the needs of business expansion and capacity reduction. For example, in the telecommunications field, more and more network functions are beginning to be hosted on the PaaS platform in the form of containerized applications.

A traditional communication network element is split into functional energy sets composed of multiple services. A complete network function is generally composed of multiple services with different characteristics, each of which contains multiple microservices (Microservice). The entity of the microservice is generally a containerized application, which is also a functional entity that the PaaS platform can deploy independently. There may be several serving cells in the microservice to provide services to network users. When the service cell business volume changes suddenly, the PaaS platform needs to expand or shrink a number of micro service instances according to the current business volume elasticity, that is, containerized applications to meet business needs.

Summary of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

The present disclosure provides a method and apparatus for shrinking a containerized application of a PaaS platform.

In one aspect, the disclosure provides a method for scaling a containerized application of a PaaS platform, including: collecting corresponding user data from a containerized application corresponding to a micro service; and according to a change of a preset indicator in the user data, The microservice expands or contracts at least one containerized application.

In an exemplary embodiment, the collecting, by the micro-service corresponding containerized application, the corresponding user data comprises: when the containerized application outputs user data to the corresponding container, forwarding the user data to a network port; The ownership of the containerized application adds a label to the user data; and stores the user data to which the label is added.

In an exemplary embodiment, the expanding or shrinking the at least one containerized application for the microservice according to a change of a preset indicator in the user data comprises: comparing a preset indicator in the user data with a preset a relationship between a threshold and a second threshold; if the preset index is greater than the first threshold, expanding at least one containerized application for the microservice; and the preset indicator is smaller than the second threshold In the case of the microservice, the at least one containerized application is shrunk, wherein the first threshold is greater than the second threshold.

In an exemplary embodiment, the preset indicator includes an average service indicator of each containerized application corresponding to the micro service, or a highest service indicator in each containerized application corresponding to the micro service.

In an exemplary embodiment, the expanding or shrinking the at least one containerized application for the microservice according to the change of the preset indicator in the user data comprises: combining a plurality of preset indicators in the user data A change in the situation, for the microservice to expand or shrink at least one containerized application.

Before the expanding or shrinking the at least one containerized application for the microservice according to the change of the preset indicator in the user data, the method further includes: filtering the required key from the user data Data; performing data processing on the key data to obtain the preset indicator.

Before collecting the corresponding user data in the containerized application corresponding to the micro service, the method further includes: configuring the collection parameter of the user data, where the collection parameter includes at least one of the following: a sampling interval, a number of samples, The threshold of the contraction and the time of the contraction.

In another aspect, the present disclosure further provides a retracting device for a containerized application of a PaaS platform, comprising: an acquisition unit configured to collect corresponding user data from a containerized application corresponding to the micro service; and a contraction unit configured to be Deriving a change in a preset indicator in the user data, expanding or shrinking at least one containerized application for the microservice.

In an exemplary embodiment, the collection unit includes: a forwarding module configured to be the container When the application outputs the user data to the corresponding container, the user data is forwarded to the network port; the adding module is configured to add a label to the user data according to the attribution of the containerized application; the storage module is set to be added The user data of the tag is stored.

In an exemplary embodiment, the pinching unit includes: a comparison module configured to compare a size relationship between a preset indicator in the user data and a first threshold and a second threshold; and an expansion module configured to be in the pre- If the indicator is greater than the first threshold, expanding at least one containerized application for the microservice; and the shrinking module is configured to be the microservice if the preset indicator is smaller than the second threshold Shrinking at least one containerized application, wherein the first threshold is greater than the second threshold.

In an exemplary embodiment, the preset indicator includes an average service indicator of each containerized application corresponding to the micro service, or a highest service indicator in each containerized application corresponding to the micro service.

In an exemplary embodiment, the pinching unit is configured to expand or contract at least one containerized application for the microservice in conjunction with a change in a plurality of preset metrics in the user data.

The apparatus further includes a screening unit configured to filter out from the user data before the microservice expands or shrinks at least one containerized application according to a change of a preset indicator in the user data Key data required; a processing unit configured to perform data processing on the key data to obtain the preset indicator.

The device further includes a configuration unit configured to configure an acquisition parameter of the user data before collecting corresponding user data from a containerized application corresponding to the microservice, the collection parameter comprising at least one of the following: sampling interval, sample Number, shrink threshold, and contraction time.

Embodiments of the present disclosure also provide a computer readable storage medium storing computer executable instructions that, when executed by a processor, implement the methods described above.

The method and device for shrinking the PaaS platform containerized application provided by the embodiment of the present disclosure can collect corresponding user data from the containerized application corresponding to the micro service, according to the change of the preset indicator in the user data, The microservice expands or shrinks at least one containerized application, so that the service volume of the service object of the containerized application can be more comprehensively reflected, and the most suitable resource is provided for the change of the traffic volume, thereby avoiding waste of platform resources. Or insufficient problem, effective Improve the utilization efficiency of platform resources.

Other aspects will be apparent upon reading and understanding the drawings and detailed description.

DRAWINGS

1 is a flowchart of a method for scaling a containerized application of a PaaS platform provided by an embodiment of the present disclosure;

2 is a schematic diagram of data flow of a method for capturing a containerized application of the PaaS platform provided by the present disclosure;

3 is a flow chart of a method of shrinking a containerized application of the PaaS platform of the embodiment shown in FIG. 2;

FIG. 4 is a schematic structural diagram of a retracting device for a containerized application of a PaaS platform according to an embodiment of the present disclosure.

detailed description

The elastic scaling of containerized applications on the current PaaS platform is mostly based on system resources occupied by applications, such as CPU and memory. However, the CPU of the microservice to which the serving cell belongs may not reflect the traffic carried by the cell, and the microservice may be collapsed according to the CPU usage, and platform resources may be wasted (unnecessary expansion). Or insufficient business resources (unnecessary shrinkage).

The present disclosure will be described in detail below with reference to the accompanying drawings. It is to be understood that the exemplary embodiments described herein are merely illustrative of the disclosure,

As shown in FIG. 1 , an embodiment of the present disclosure provides a method for scaling a containerized application of a PaaS platform, including:

S11: Collect corresponding user data from a containerized application corresponding to the micro service;

S12. Expand or shrink at least one containerized application for the microservice according to a change of a preset indicator in the user data.

The shrinking method of the PaaS platform containerized application provided by the embodiment of the present disclosure can be from the micro service Collecting corresponding user data in the corresponding containerized application, expanding or contracting at least one containerized application for the microservice according to the change of the preset indicator in the user data, so that the containerization can be more comprehensively reflected The service volume of the application service object provides the most suitable resource for the change of the service volume, thereby avoiding the problem of waste or insufficient platform resources, and effectively improving the utilization efficiency of the platform resources.

For example, there is a one-to-one or one-to-many correspondence between a microservice and a containerized application, that is, each microservice may have one or more containerized applications that serve it. The final vehicle for deploying applications on the PaaS platform is the Docker container, and various user data generated by the application can be output to the corresponding Docker container. Each Docker container has a corresponding daemon thread, Docker deamon, which allows the user data of the application running in the container to be output to the network port by customizing the Docker deamon.

In this embodiment, in step S11, collecting corresponding user data from the containerized application corresponding to the micro service may include:

Transmitting the user data to a network port when the containerized application outputs user data to the corresponding container;

Adding a label to the user data according to the attribution of the containerized application;

The user data to which the tag is added is stored.

For example, the PaaS platform can listen to network ports, receive user data forwarded by the application through Docker deamon, parse and filter user data, and filter out irrelevant data. Then, the corresponding user data is tagged according to the attribution of the user data. In this embodiment, the attribution of the user data may refer to which application, which container, which micro service, etc. the user data is from, and the added tag may be, for example, a container ID, a micro service ID, or the like.

After the tag is added, the user data needs to be stored to obtain the preset metrics from the user data. In the process of storing user data, in order to avoid data loss caused by storms, the user may be cached first, for example, the distributed data is processed by the distributed publish and subscribe system kafka. User data-specific message queues can also be created in kafka for subscription by back-end consumers. Kafka can not only provide message queues, but also provide a message persistence mechanism by writing disks, thus effectively avoiding data loss caused by data storms. In addition, slow The stored user data can be stored persistently using the elasticsearch database. For example, user data can be subscribed to from kafka via logstash-indexer and stored in a fixed format in the elasticsearch database. The user data in elasticsearch can be used as the data source of PaaS to obtain the corresponding preset indicators. Depending on the application scenario, the preset indicators that the user cares about may also be different, and the corresponding preset indicators may be obtained from the user data as needed.

After the user data is collected, at least one containerized application may be expanded or contracted for the micro service according to the change of the preset indicator in the user data in step S12. For example, expanding or shrinking at least one containerized application for the microservice may include: according to a change of a preset indicator in the user data:

Comparing a size relationship between the preset indicator in the user data and the first threshold and the second threshold;

And expanding, in the case where the preset indicator is greater than the first threshold, at least one containerized application for the microservice;

And in a case that the preset indicator is smaller than the second threshold, shrinking, by the micro service, at least one containerized application, wherein the first threshold is greater than the second threshold. That is, the first threshold corresponds to the elastic expansion threshold, and the second threshold corresponds to the elastic contraction threshold.

In this embodiment, the preset indicator may include an average service indicator of each containerized application corresponding to the micro service. For example, in an embodiment of the present disclosure, if the user access amount of the serving cell in the communication is concerned, the related parameter of the user access amount may be extracted in the collected user data, and the user access amount is corresponding. Preset indicator. If there are three containerized applications corresponding to a certain serving cell, the user access amount related parameter indicates that the average normalized parameter of the user access amount of each containerized application is 80%, that is, the user access of the three applications. The number has reached 80% of the maximum allowable access. If the first threshold is 75% and the second threshold is 40%, you can expand a containerized application for the microservice. The application shares a portion of user access, thereby reducing the amount of user access per containerized application. If there are three containerized applications corresponding to the serving cell, the user access amount related parameter indicates that the average normalized parameter of the user access amount of each containerized application is 35%, that is, the user access of the three applications. If the number is only 35% of the maximum allowable access, the containerized application can be reduced for the micro-service, and the user access of the three containerized applications can be shared to two containerized applications, thereby effectively saving the platform. Application resources. Of course, it is also possible to expand or contract according to the relationship between the preset index and the first threshold and the second threshold. An example of a plurality of containerized applications is not limited by the embodiments of the present disclosure.

In this embodiment, the preset indicator may also include a highest service indicator in each containerized application corresponding to the micro service. For example, if there are 4 containerized applications corresponding to a certain serving cell, the user access amount related parameter indicates that the maximum of the normalized parameters of the user access amount in the four containerized applications is 85%, if specified at this time. The first threshold is 80%, which can expand a containerized application for the microservices, and enable the containerized application with the largest user access to actively transfer some of the user access services to the newly expanded containerized application, thereby reducing the task most. The load of heavy containerized applications.

In order to improve the accuracy of the retraction control of the containerized application, a plurality of preset indicators may be set for the retraction of the application, and the microservices may be expanded or combined with changes in the plurality of preset indicators in the user data. Shrink one or more containerized applications. For example, the condition for shrinking a containerized application may be set as follows: the cell user access amount is less than 40% and the cell access failure rate is less than 20%. That is, a containerized application is only shrunk when both conditions are met. In this embodiment, according to the relationship between different user data, a sticky relationship between multiple user data is arranged, and a joint scaling strategy based on multiple types of user data is configured, so that the containerized application can be contracted in combination with various factors. control.

It can be understood that different microservices, different application scenarios, and different control requirements may need to extract different parts of user data to form different preset indicators. In order to obtain the corresponding preset indicator, in an embodiment of the present disclosure, before the at least one containerized application is expanded or contracted for the micro service according to a change of the preset indicator in the user data, the method may further include :

Filtering the required key data from the user data; in this embodiment, which key data is specifically required to be pre-selected by the user;

Data processing is performed on the key data to obtain the preset indicator.

In order to obtain the required user data, in the embodiment, before the corresponding user data is collected from the containerized application corresponding to the micro service in step S11, the scaling method of the PaaS platform containerized application provided by the embodiment of the present disclosure may also be used. include:

The collection parameter of the user data is configured, and the collection parameter includes one or more of the following: a sampling interval, a sample number, a contraction threshold, and a contraction penalty time. The sampling interval indicates how often the user data is collected, and the number of samples indicates how many samples are collected to start the change of the preset index. Judging, the contraction threshold can be divided into a contraction expansion threshold and a contraction contraction threshold. The contraction expansion threshold indicates that the preset index exceeds the threshold to expand a containerized application, and the contraction contraction threshold indicates that the preset index exceeds the threshold and decreases. A containerized application. The contraction penalty time indicates that the expansion or contraction of the containerization application is not performed for a period of time after the change of the contraction of the containerization application, and the corresponding expansion of the containerization application is not performed. The regulation of the contraction prevents the expansion or contraction of the containerized application from being too frequent and affects the stability of the system.

The method of projecting the PaaS platform containerization application provided by the present disclosure will be described in detail below by way of exemplary embodiments.

2 is a schematic diagram of data flow of a method for capturing a containerized application of the PaaS platform provided by the present disclosure, and FIG. 3 is a flowchart of a method for scaling a containerized application of the PaaS platform of the embodiment shown in FIG. 2 and FIG. 3, the method for shrinking the container application of the PaaS platform of the present embodiment may include the following steps:

S201: The PaaS platform provides a docker base image for deploying containerized applications. The image provides a customized print module, and the application can print user data to the standard output of the container according to the agreed format.

S202: Modify the Docker configuration file and configure the driver mode as tcp output, so that the user data can be redirected to the userkpi-forwarder (user data agent module). Userkpi-forwarder is mainly used to increase the attribution information of user data, such as adding micro service name and ID information, and outputting to the back end.

S203: The user data proxy module outputs the user data to the kafka (distributed publish and subscribe system), and creates a message queue dedicated to the user data in the kafka for the backend consumer to subscribe. Kafka is not only used as a message queue, but also because of its own message persistence mechanism, which can avoid data loss caused by data storms.

S204: The logstash-indexer subscribes to the user data from kafka and stores it in the elasticsearch database in a fixed format. User data in elasticsearch is the data source for the PaaS bombing module.

S205: The scaler reads data from the elasticsearch database according to characteristics of the user data, such as user data name, micro service ID, etc., and calculates a current average value of the user data according to a pre-configured algorithm.

S206: The scaler compares the current user data actual value and the zoom threshold to initiate a contraction decision. If the actual value of the user data is higher than the elastic extension threshold, the deployer (PaaS deployment module) is notified to flexibly extend the microservice. If the actual value of the user data is lower than the elastic shrinkage threshold, notify the deployer (PaaS deployment module) to elastically shrink the microservice.

Correspondingly, as shown in FIG. 4, an embodiment of the present disclosure further provides a retracting device for a containerized application of a PaaS platform, including:

The collecting unit 41 is configured to collect corresponding user data from the containerized application corresponding to the micro service;

The pinch unit 42 is configured to expand or contract at least one containerized application for the microservice according to a change in a preset indicator in the user data.

The captive device of the containerized application of the PaaS platform provided by the embodiment of the present disclosure can collect corresponding user data from the containerized application corresponding to the micro service, and according to the change of the preset indicator in the user data, the micro The service expands or shrinks at least one containerized application, so that the service volume of the service object of the containerized application can be more comprehensively reflected, and the most suitable resource is provided for the change of the traffic volume, thereby avoiding waste or insufficient platform resources. The problem effectively improves the utilization efficiency of platform resources.

In this embodiment, the collecting unit 41 may include:

a forwarding module, configured to forward the user data to a network port when the containerized application outputs user data to the corresponding container;

Adding a module, configured to add a label to the user data according to the attribution of the containerized application;

A storage module configured to store the user data to which the tag is added.

In this embodiment, the pinch unit 42 can include:

a comparison module, configured to compare a size relationship between the preset indicator in the user data and the first threshold and the second threshold;

An expansion module, configured to expand at least one containerized application for the microservice if the preset indicator is greater than the first threshold;

a shrinking module, configured to be in a case where the preset index is less than the second threshold The microservice shrinks at least one containerized application, wherein the first threshold is greater than the second threshold.

In this embodiment, the preset indicator may include an average service indicator of each containerized application corresponding to the micro service, or a highest service indicator in each containerized application corresponding to the micro service.

In this embodiment, the pinching unit may be configured to expand or shrink at least one containerized application for the microservice in combination with a change of a plurality of preset indicators in the user data.

The retracting device of the container application of the PaaS platform provided by the embodiment of the present disclosure may further include:

a screening unit, configured to filter out required key data from the user data before expanding or shrinking the at least one containerized application according to the change of the preset indicator in the user data ;

And a processing unit configured to perform data processing on the key data to obtain the preset indicator.

The retracting device of the PaaS platform container application provided by the embodiment of the present disclosure may further include a configuration unit configured to configure the collection parameter of the user data before collecting corresponding user data from the containerized application corresponding to the micro service. The acquisition parameters include at least one of the following: a sampling interval, a number of samples, a contraction threshold, and a penalty time.

Embodiments of the present disclosure also provide a communication network element, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being implemented by the processor The following steps are: collecting corresponding user data from the containerized application corresponding to the micro service; expanding or shrinking at least one containerized application for the micro service according to the change of the preset indicator in the user data.

Embodiments of the present disclosure also provide a computer readable storage medium storing computer executable instructions that, when executed by a processor, implement the methods described above.

The term "comprises", "comprises" or any other variants thereof is intended to encompass a non-exclusive inclusion, such that a process, method, article, or device that comprises a plurality of elements includes not only those elements but also Other elements listed, or elements that are inherent to such a process, method, article, or device. An element that is defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device that comprises the element.

Those of ordinary skill in the art will appreciate that all or some of the steps, systems, and functional blocks/units of the methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. In a hardware implementation, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be composed of several physical The components work together. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer readable medium, which may include computer storage media (or non-transitory media) and communication media (or transitory media). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and nonvolatile, implemented in any method or technology for storing information, such as computer readable instructions, data structures, program modules or other data. Sex, removable and non-removable media. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridge, magnetic tape, magnetic disk storage or other magnetic storage device, or may Any other medium used to store the desired information and that can be accessed by the computer. Moreover, it is well known to those skilled in the art that communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and can include any information delivery media. .

The above is only an exemplary embodiment of the present disclosure, and is not intended to limit the scope of the disclosure of the present disclosure, and the equivalent structure or equivalent process transformations made by the present disclosure and the contents of the drawings, or directly or indirectly applied to other related technologies. The scope of the invention is included in the scope of patent protection of the present disclosure.

Industrial applicability

The method and device for shrinking the PaaS platform containerized application provided by the embodiment of the present disclosure can collect corresponding user data from the containerized application corresponding to the micro service, according to the change of the preset indicator in the user data, The microservice expands or shrinks at least one containerized application, so that the service volume of the service object of the containerized application can be more comprehensively reflected, and the most suitable resource is provided for the change of the traffic volume, thereby avoiding waste of platform resources. Or insufficient problems, effectively improving the utilization efficiency of platform resources. The present disclosure therefore has industrial applicability.

Claims (13)

A scaling method for a platform-as-a-service PaaS platform containerized application, comprising: Collecting corresponding user data from the containerized application corresponding to the micro service (S11); And expanding or shrinking at least one containerized application for the microservice according to a change of the preset indicator in the user data (S12). The method according to claim 1, wherein the collecting the corresponding user data from the containerized application corresponding to the micro service (S11) comprises: Transmitting the user data to a network port when the containerized application outputs user data to the corresponding container; Adding a label to the user data according to the attribution of the containerized application; The user data to which the tag is added is stored. The method according to claim 1 or 2, wherein the expanding or contracting the at least one containerized application (S12) for the microservice according to a change of a preset indicator in the user data comprises: Comparing a size relationship between the preset indicator in the user data and the first threshold and the second threshold; And expanding, in the case where the preset indicator is greater than the first threshold, at least one containerized application for the microservice; And in a case that the preset indicator is smaller than the second threshold, shrinking, by the micro service, at least one containerized application, wherein the first threshold is greater than the second threshold. The method according to claim 3, wherein the preset indicator comprises an average service indicator of a plurality of containerized applications corresponding to the micro service, or a highest service indicator of a plurality of containerized applications corresponding to the micro service . The method according to claim 1 or 2, wherein said said microservice expands or contracts at least one containerized application (S12) according to a change in a preset indicator in said user data The method also includes: Filtering the required key data from the user data; Data processing is performed on the key data to obtain the preset indicator. The method according to claim 1 or 2, wherein before the collecting the corresponding user data (S11) in the containerized application corresponding to the microservice, the method further comprises: The collection parameter of the user data is configured, and the collection parameter includes at least one of the following: a sampling interval, a sample number, a contraction threshold, and a contraction penalty time. A shrinking device for a containerized application of a PaaS platform, comprising: The collecting unit (41) is configured to collect corresponding user data from the containerized application corresponding to the micro service; The pinch unit (42) is configured to expand or contract at least one containerized application for the microservice based on a change in a preset indicator in the user data. The apparatus according to claim 7, wherein said acquisition unit (41) comprises: a forwarding module, configured to forward the user data to a network port when the containerized application outputs user data to the corresponding container; Adding a module, configured to add a label to the user data according to the attribution of the containerized application; A storage module configured to store the user data to which the tag is added. The device according to claim 7 or 8, wherein the pinching unit (42) comprises: a comparison module, configured to compare a size relationship between the preset indicator in the user data and the first threshold and the second threshold; An expansion module, configured to expand at least one containerized application for the microservice if the preset indicator is greater than the first threshold; And a shrinking module, configured to shrink the at least one containerized application for the microservice if the preset index is less than the second threshold, wherein the first threshold is greater than the second threshold. The device according to claim 9, wherein the preset indicator includes an average service indicator of the plurality of containerized applications corresponding to the micro service, or a highest service indicator of the plurality of containerized applications corresponding to the micro service . The device according to claim 7 or 8, further comprising: a screening unit, configured to change the preset indicator according to the user data, Filtering the required key data from the user data before expanding or shrinking the at least one containerized application for the microservice; And a processing unit configured to perform data processing on the key data to obtain the preset indicator. The apparatus according to claim 7 or 8, further comprising a configuration unit configured to configure an acquisition parameter of the user data, the acquisition parameter, before collecting corresponding user data from a containerized application corresponding to the microservice It includes at least one of the following: sampling interval, number of samples, shrink threshold, and cambering time. A computer readable storage medium storing computer executable instructions that, when executed by a processor, implement the method of any of claims 1-6.

Images