CN114625474A

CN114625474A - Container migration method and device, electronic equipment and storage medium

Info

Publication number: CN114625474A
Application number: CN202011443791.7A
Authority: CN
Inventors: 丁伟; 龙翼; 张胜举
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Suzhou Software Technology Co Ltd
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Suzhou Software Technology Co Ltd
Priority date: 2020-12-08
Filing date: 2020-12-08
Publication date: 2022-06-14

Abstract

The invention provides a container migration method, a container migration device, electronic equipment and a storage medium. The method comprises the following steps: determining resource consumption ratios of the alternative containers deployed within the first server; wherein the resource consumption ratio is: a ratio between an amount of resources consumed by a corresponding candidate container and an amount of resources owned within the first server; determining a resource growth rate of each alternative container deployed in the first server; determining a target container to be migrated from the first server from each alternative container according to each resource consumption rate and each resource growth rate; migrating the target container to a second server; wherein the load rate of the second server is lower than the load rate of the first server.

Description

Container migration method and device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of container migration technologies, and in particular, to a container migration method and apparatus, an electronic device, and a storage medium.

Background

With the rapid development of the cloud computing technology, most resources are currently deployed on a cloud server, and before the container technology appears, a virtual machine is always an important carrier of a cloud computing platform. Due to the fact that operation, maintenance and management costs of the virtual machines are high, configuration is messy. The container technology effectively solves the problems of troublesome configuration and difficult maintenance of the traditional virtual machine, and is gradually used by people due to the characteristics of quick container starting, less occupied resources and the like.

However, in a server cluster, since the number of containers deployed on different servers is usually different, and server resources consumed by different containers may also be different, a problem of load imbalance among different servers in the server cluster often occurs, for example, one or some servers are overloaded and other servers are underloaded.

Disclosure of Invention

The embodiment of the invention provides a container migration method and device, electronic equipment and a storage medium.

The technical scheme of the embodiment of the invention is realized as follows:

in a first aspect, an embodiment of the present invention provides a container transfer method, including:

determining resource consumption ratios of the alternative containers deployed within the first server; wherein the resource consumption ratio is: a ratio between an amount of resources consumed by a corresponding candidate container and an amount of resources owned within the first server;

determining a resource growth rate of each alternative container deployed in the first server;

determining a target container to be migrated from the first server from each alternative container according to each resource consumption rate and each resource growth rate;

migrating the target container to a second server; wherein the load rate of the second server is lower than the load rate of the first server.

In a second aspect, an embodiment of the present invention provides a container transfer apparatus, including:

a determination module to determine a resource consumption ratio of an alternative container deployed within a first server; determining a resource growth rate of an alternative container deployed within a first server; determining a target container to be migrated from the first server according to the resource consumption rate and the resource growth rate;

and the migration module is used for migrating the target container to a second server.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

a memory for storing executable instructions;

and the processor is used for implementing the container migration method provided by one or more of the above technical solutions when executing the executable instructions stored in the memory.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions; after being executed by a processor, the computer-executable instructions can implement the container migration method provided by one or more of the above technical solutions.

The embodiment of the invention provides a container migration method and device, electronic equipment and a storage medium. By determining a resource consumption ratio of an alternative container deployed within a first server; determining a resource growth rate of an alternative container deployed within a first server; determining a target container to be migrated from the first server according to the resource consumption rate and the resource growth rate; migrating the target container to a second server.

Therefore, the target container with high resource consumption rate and high resource growth rate is selected from the alternative containers according to the resource consumption rate and the resource growth rate, and the target container is transferred from the first server to the second server, so that the load pressure of the first server can be relieved, and the load balance among the servers can be realized; more containers can be operated on the server with low load rate, so that higher response rate and container service quality can be obtained.

Drawings

FIG. 1 is a schematic diagram of a user space based Checkpoint/Restore In Userspace (CRIU) tool for live migration;

figure 2 is a schematic diagram of another CRIU-based tool for performing a thermomigration;

FIG. 3 is a schematic flow diagram of a pre-copy based container migration method;

FIG. 4 is a schematic flow chart of a container migration method according to an embodiment of the present invention;

FIG. 5 is a schematic flow chart showing details of the target container determination step in the corresponding embodiment of FIG. 4;

FIG. 6 is a flowchart illustrating the details of the step of migrating the target container to the second server in the corresponding embodiment of FIG. 4;

FIG. 7 is a schematic structural diagram of a container transfer device according to an embodiment of the present invention;

FIG. 8 is a flow chart illustrating a method for determining a target container according to an embodiment of the present invention;

fig. 9 is a schematic flowchart of a target container migration method according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail with reference to the accompanying drawings, the described embodiments should not be construed as limiting the present invention, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.

In the following description, references to the terms "first \ second \ third" are only to distinguish similar objects and do not denote a particular order, but rather the terms "first \ second \ third" are used to interchange specific orders or sequences, where appropriate, to enable embodiments of the invention described herein to be practiced in other than the order shown or described herein.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to be limiting of the invention.

In the related art, the container migration process can be divided into two steps, namely, the determination of the server and the target container and the migration of the target container.

For the process of determining the server and the target container, in the related art, generally, a top instruction is executed by traversing all the alternative servers, the server with the largest load and the server with the smallest load are continuously subjected to record updating, the servers to be migrated (with the largest load) and the servers to be migrated (with the smallest load) are selected, and then the container with the largest resource consumption is found out from the servers to be migrated as the target container. However, since the resources consumed by the server are dynamic, the method for determining the target container only according to the resource consumption condition of the container cannot effectively relieve the load pressure of the server.

For the migration process of the target container, in order to migrate the state information such as the memory, the file system, the network link and the like of the target container in the operation process on the source server to the target server, it is ensured that the target container can normally operate after being migrated to the target server. The migration method of the target container in the related art can be divided into:

1. the cold migration mode is to stop the service running on the container and migrate the container from the source server to the target server at the time of migration. But this cold migration approach requires service disruption.

2. The mode of hot migration is that the running state of the container service is saved at the source server and then transmitted to the target server, and the service is recovered at the target server. The hot migration mode has very small influence on the service of the user.

The hot migration of the container is usually realized by calling a process hot migration tool CRIU, the CRIU can freeze the process of the container, transmit the frozen process files to a target server, restore the frozen process by using the process files, and recover the service on the target server. As shown in fig. 1 and 2, fig. 1 is a schematic diagram of a thermal migration based on a CRIU tool, and fig. 2 is a schematic diagram of another thermal migration based on a CRIU tool.

In the related art, many live migration schemes are applied to the virtual machine more mature, and no very mature solution is provided for live migration of the container.

The academia refers to a mode of virtual machine live migration, proposes a 'stop-copy' and 'pre-copy' migration strategy of a container, and realizes the live migration of the container. For example, as shown in FIG. 3, FIG. 3 is a flow diagram of a pre-copy based container migration method. The container migration is performed in an iterative copy manner, but the memory pages in the container are changed, which causes the memory pages with very small changes to be repeatedly transmitted, and reduces the container migration efficiency.

An embodiment of the present invention provides a container transfer method, and fig. 4 is a schematic flow chart of the container transfer method provided in the embodiment of the present invention, and as shown in fig. 4, the method includes the following steps:

step 401, determining resource consumption ratios of the alternative containers deployed in the first server; wherein the resource consumption ratio is: a ratio between an amount of resources consumed by a corresponding candidate container and an amount of resources owned within the first server;

step 402, determining the resource growth rate of each alternative container deployed in the first server;

step 403, determining a target container to be migrated from the first server from each candidate container according to each resource consumption rate and each resource growth rate;

step 404, migrating the target container to a second server; wherein the load rate of the second server is lower than the load rate of the first server.

In embodiments of the present invention, the alternative container may be a virtualized container, such as a Docker container, a container, an cri-o container, or a rkt container.

It should be noted that the container provides an isolated operating space for the application program; each container contains an exclusive complete user environment space, and the change in one container does not affect the operating environment of other containers. Thus, the migration of the target container does not affect the operation of other alternative containers in the first server.

In view of dynamic resource consumption of services corresponding to each container, in order to determine the target container more accurately, the embodiment of the present invention determines the target container comprehensively according to the resource consumption rate and the resource growth rate of the candidate container, thereby ensuring that the load pressure of the first server is reduced and that the first server is not overloaded again due to the resource growth consumed by the container service within a period of time.

In practical applications, the resource consumption ratio is: a ratio of an amount of resources consumed by a corresponding candidate container within the first server to an amount of resources contained within the first server. For example, container 1, container 2, and up to container N are run in the first server. N may be any positive integer greater than 2.

The resource consumption ratio of container n is: the amount of resources consumed by container n is compared to the total resources contained within the first server. N is any positive integer less than or equal to N.

In some embodiments, the resource consumption of the alternative container may be directly viewed through a visualization management tool (e.g., docker ui, Portainer, or shipyard) of the container, for example, the total resource consumption of the alternative container is viewed.

In the embodiment of the present invention, the alternative container may be any container deployed in the first server, and the term "alternative container" herein means: is selected as a potential container for migration to the second server.

In some embodiments, the resource consumption of the alternative container and the resource amount of the first server may be obtained; and determining the resource consumption ratio corresponding to the alternative container according to the resource consumption condition of the alternative container and the resource quantity of the first server.

In the embodiment of the present invention, the resource growth rate is a data growth rate of the candidate container in a unit time; the data growth rate of the alternative container in unit time can be determined according to the growth condition of the resource consumption ratio of the alternative container.

In step 403, the candidate containers may be sorted according to each resource consumption rate and each resource growth rate, and a target container to be migrated from the first server may be determined from each candidate container according to two sorting results.

For example, the first server contains 4 candidate containers, which are container a, container B, container C, and container D; respectively obtaining the resource consumption rate and the resource growth rate of the container A, the container B, the container C and the container D, and sequencing according to the resource consumption rate from large to small to obtain sequencing results which are sequentially as follows: container A, container D, container C, container B; sorting according to the resource consumption increase rate from large to small to obtain a sorting result, namely a container B, a container A, a container C and a container D in sequence; respectively acquiring a first ordering result and a second ordering result; and intersecting the first and second ordering results, and determining the target container according to the obtained intersection (namely the container A).

In step 404, the target container contains static data and dynamic data, wherein the static data may include: a base image file of the target container; the dynamic data may include: the operation data of the target container is usually stored in the memory.

The static data is data with lower modification frequency than the dynamic data. For example, the frequency of modification of the static data may be data below a frequency threshold, and the dynamic data may be data above the frequency threshold.

Here, the static data refers to data used as control or reference in the operation process, and generally does not change with the operation; such as config configuration files, images files, etc. And the dynamic data refers to data that changes with time in the system application. The dynamic data is used for reflecting the data in the operation process, and the modification frequency is higher, so that the dynamic data is stored in the memory; for example, process data corresponding to the container, read-write data of the file, and the like. The basic image file of the target container contains the content in the file system of the target container, and is the basis for creating the target container.

In some embodiments, prior to migrating the target container to a second server, migrating static data of the target container to the second server preferentially based on a network connection between a first server and the second server; and after the static data is successfully migrated, migrating the memory page contained in the target container to a second server.

In practical application, the static data of the target container may include configuration information of the target container; the configuration information of the target container can be used for reconstructing a creation command of the target container, so that the configuration of the target container is reproduced in the container reconstructed by the second server according to the configuration information of the target container.

In some embodiments, the method further comprises:

deleting the target container in the first server after successfully migrating the target container to the second server.

Migrating the target container to a second server, comprising:

according to the state information of a target container in a first server, establishing the target container with the same state as that in the first server;

migrating the memory page of the target container in the first server to the target container of the second server.

Wherein the status information comprises at least one of:

operation state information indicating an operation state of the target container;

attribute status information indicating attributes of the target container, for example, the attributes of the target container include: storage attributes, calculation attributes, and/or mixed calculation and storage attributes.

Alternatively, as shown in fig. 5, fig. 5 is a detailed flowchart of the target container determining step in the corresponding embodiment of fig. 4. Step 403 comprises:

4031, determining a migration value according to each resource consumption ratio, a first weight corresponding to each resource consumption ratio, each resource growth rate and a second weight corresponding to each resource growth rate; the migration value is used for representing the migration probability of the corresponding alternative container;

step 4032, according to the migration value, the target container is selected from the candidate containers.

The migration value may be: selecting a reference parameter value for a target container from the candidate containers.

In an embodiment, whether a target container meeting a migration condition exists in the candidate containers is determined according to the migration value, and if the candidate container meeting the migration condition exists, the candidate container meeting the migration condition is used as the target container to be migrated to the second server.

For example, alternative containers that satisfy the migration condition include:

and the candidate container with the migration value within the preset range set by the migration condition is the candidate container meeting the migration condition.

In the embodiment of the invention, the migration value is used for representing the migration probability of the corresponding alternative container; if the migration value is positively correlated with the migration probability, that is, the larger the migration value is, the larger the migration probability that the corresponding candidate container is taken as the target container is, at this time, the migration value is greater than a preset range and a corresponding migration lower limit threshold is set, which may be considered as: the migration condition is satisfied. If the migration value is negatively correlated with the migration probability, that is, the smaller the migration value is, the greater the migration probability that the corresponding candidate container is used as the target container is, at this time, the migration value is smaller than a migration lower limit threshold value set in a preset range, which may be considered as: the migration condition is satisfied.

In other embodiments, at least one target container is selected from the candidate containers based on the migration value. Sorting the migration values, and selecting one or more alternative containers corresponding to the migration values in the top sorting order as the target containers when the migration values are positively correlated with the migration probability according to the sorting result; or when the migration value is negatively correlated with the migration probability, selecting the alternative container corresponding to one or more migration values in the top ranking as the target container.

In some embodiments, the migration value of the target container may be determined by:

a migration value is a resource consumption ratio, namely a first weight value + a resource increase rate, namely a second weight value;

the first weight and the second weight may be set according to user requirements, for example, if the user considers that the resource consumption ratio of the container is more important, the first weight may be set to be larger.

Determining a migration value of the alternative container according to the sum of the resource consumption rate and the resource increase rate of the alternative container, determining the alternative container with high resource increase rate in the first server as a target container according to the migration value, ensuring that the load pressure of the first server is reduced through the migration of the target container, and simultaneously ensuring that the first server is not overloaded again due to the increase of the container service resources within a period of time.

In some embodiments, the migration value may be further determined according to a load rate of the first server, a resource consumption rate of the alternative container, a first weight corresponding to the resource consumption rate, the resource growth rate, and a second weight corresponding to the resource growth rate.

In step 4032, the candidate containers may be sorted according to the migration value, and at least one target container may be selected from the candidate containers according to the sorting result.

In some embodiments, the first weight may be a preset value or a default value. The default value of the first weight value can be 1; in other embodiments, the first weight may preset a preset value of the first weight to be any value between 0 and 1 according to a type of the first server and/or a resource scheduling policy of the first server, and the like.

Optionally, step 4031 includes:

and selecting N candidate containers with the highest resource consumption rate from the candidate containers to participate in the migration value calculation.

Here, N is an integer less than 1.

In practical application, in order to reduce the calculation amount, the various alternative containers may be sorted according to the resource consumption ratio of the various alternative containers; and selecting N candidate containers with the highest resource consumption ratio from the candidate containers according to the sorting result. Acquiring the resource growth rate of the N alternative containers; and calculating the migration values of the N candidate containers according to the resource growth rate and the resource consumption ratio of the N candidate containers. Sorting the N candidate containers according to the migration value; and determining at least one target container from the N candidate containers according to the sequencing result.

Optionally, the method further comprises:

and determining the second weight according to the service type of the alternative container.

In practical applications, the second weight corresponding to the alternative container may be determined according to importance of a service type corresponding to the alternative container. For example, if the service type corresponding to the container a is storage data and the service type corresponding to the container B is cloud computing, the second weight corresponding to the container a is smaller than the second weight corresponding to the container B.

Optionally, as shown in fig. 6, fig. 6 is a detailed flowchart of the step of migrating the target container to the second server in the embodiment corresponding to fig. 4. Step 404 includes:

step 4041, determining a change rate of dirty pages in memory pages included in the target container;

step 4042, determining a migration sequence in which the plurality of memory pages of the target container are migrated to the second server according to the change rate of the dirty pages;

step 4043, sequentially migrating the memory pages of the target container to the second server according to the migration order.

In the embodiment of the present invention, a memory monitoring module may be disposed in the first server, and configured to monitor the memory page of the target container in the first server, mark the memory page when the memory page in the target container is changed, and record the memory page in the dirty page set.

In step 4041, the rate of change of the dirty page may be used to represent a ratio of changed data on the dirty page to stored data of the dirty page.

In practical application, the change rate of the dirty page may be determined by a ratio of the number of bytes corresponding to the changed data in the dirty pages acquired twice to the number of bytes corresponding to the dirty page data acquired for the first time.

In actual implementation, as long as the change rate of the dirty page can reflect the change condition of the data in the dirty page, the embodiment of the present invention does not limit the specific calculation formula of the change rate of the dirty page.

Optionally, step 4042 may include:

if the change rate of the dirty pages is greater than a first threshold value, determining that the migration sequence of the dirty pages is a third sequence;

if the change rate of the dirty pages is smaller than a first threshold and larger than a second threshold, determining that the migration sequence of the dirty pages is a second sequence;

if the change rate of the dirty pages is smaller than a second threshold value, determining that the migration sequence of the dirty pages is a first sequence;

wherein the first order precedes the second order, and the second order precedes the third order.

In the embodiment of the present invention, the first threshold and the second threshold may be set according to a user requirement, wherein the first threshold is greater than the second threshold. For example, the first threshold is set to 50% and the second threshold is set to 10%. The present invention is not limited thereto.

Here, the dirty page refers to a memory page in which data and a data page of a disk do not coincide.

In practical application, if the change rate of the dirty pages is greater than a first threshold, indicating that the change frequency of the dirty pages is high, the dirty pages with high change frequency are set in the last migration, so as to reduce the repeated transmission of the dirty pages. If the change rate of the dirty pages is smaller than a first threshold and larger than a second threshold, the dirty pages are changed, but the change frequency is not high, and the dirty pages are arranged in front of the dirty pages with high change frequency for migration; if the change rate of the dirty pages is smaller than a second threshold, it is indicated that the dirty pages are changed very rarely, and the dirty pages in the dirty page set can be migrated preferentially when being migrated.

In some embodiments, the dirty pages in the target container may be classified according to the change rate of the dirty pages, and a first migration order corresponding to each set is determined; and determining a second migration sequence of the dirty pages in each set according to the change rate of the dirty pages in each set.

In practical application, the dirty pages in the target container can be classified according to the comparison result of the change rate of the dirty pages with the first threshold and the second threshold.

If the change rate of the dirty pages is greater than a first threshold, which indicates that the change times of the dirty pages are more, classifying the dirty pages into a high-change memory page set; if the change rate of the dirty pages is smaller than a first threshold and larger than a second threshold, the dirty pages are changed, but the change times are not large, and the dirty pages can be classified into a medium-changed memory page set; if the change rate of the dirty pages is smaller than a second threshold, the dirty pages are basically not changed, and the dirty pages are classified into a set of memory pages without changes. And determining that the first migration sequence of the non-changed memory page set is prior to the first migration sequence of the medium-changed memory set, and the first migration sequence of the medium-changed memory set is prior to the first migration sequence of the high-changed memory set.

Sorting the dirty pages in each set according to the change rate of the dirty pages in each set, and determining a second migration sequence of the dirty pages in each set according to a sorting result; and sequentially migrating the dirty pages in the sets according to the first migration sequence corresponding to each set and the second migration sequence corresponding to the dirty pages in each set.

Therefore, the migration sequence of the dirty pages is determined according to the change rate of the dirty pages in the memory, the transmission of repeated dirty pages is effectively reduced, the transmission times are reduced, and the container migration time is shortened.

Optionally, step 4043 may include:

migrating memory pages except the dirty pages of the target container to the second server;

and after the memory pages except the dirty pages are successfully migrated, migrating the dirty pages of the target container to a second server according to the migration sequence.

In practical application, the memory pages include dirty pages and memory pages other than the dirty pages, where the memory pages other than the dirty pages refer to memory pages that have not been changed; its migration order takes precedence over the migration order of the dirty pages.

Preferentially migrating memory pages other than dirty pages in the memory pages contained in the target container to a second server through communication connection between the first server and the second server; and after the memory pages except the dirty pages are successfully migrated, sequentially migrating the dirty pages to the second server according to the migration sequence corresponding to the dirty pages.

Optionally, before the step 401, the method may further include:

step 405, determining load condition information of servers in the cluster; and determining the first server to be migrated out of at least one container and the second server capable of accepting container migration according to a load balancing strategy and the load condition information.

Wherein the load condition information may include: average load, CPU usage, memory usage, and/or network bandwidth usage.

The average load refers to the number of processes occupying the CPU time and the number of processes waiting for the CPU time within a period of time; the CPU utilization rate refers to CPU resources occupied by the currently running program; the memory utilization rate is the memory consumed by all current processes; the network bandwidth utilization rate refers to the efficiency of receiving and sending information per second by the bandwidth.

In practical application, due to the diversity of tasks processed by the server, a large amount of real-time online processing services and a large amount of asynchronous processing services exist. The real-time online service processing time is short, the demand fluctuation is large, the asynchronous service processing time is long, and the data volume is huge. The completion of tasks in each container varies with the execution of the task. Thus, the load of the containers contained by the servers in the cluster may change dynamically as tasks are performed.

Some servers are not in time due to the fact that tasks are completed earlier and new tasks do not arrive in time, so that containers in the servers are in an idle state, resource utilization rate is low, and server load is low; some servers distribute new tasks due to too long tasks, so that containers in the servers are in a busy state, the resource utilization rate is too high, and the server load is too heavy. Whether the server load is low or the server load is heavy, instability of the cluster server is caused. Therefore, the servers with heavy load and low load need to be determined according to the load condition information of each server in the cluster, and the load among the servers in the cluster is balanced in a container migration manner.

In practical application, the load rate of each server can be obtained by traversing each server in the cluster; and selecting a maximum load server and a minimum load server from the cluster according to the load rate and the load balancing strategy of each server, determining the maximum load server as a first server, and determining the minimum load server as a second server.

In some embodiments, the first server and the second server may be determined by obtaining CPU, memory, and network bandwidth utilization of each server in the cluster, and determining the comprehensive load condition information of each server.

In some embodiments, a load threshold value may be preset, and the candidate servers in the cluster are monitored, and according to the load condition information of each candidate server in the cluster, if the load condition information of the candidate server exceeds the preset load threshold value, the candidate server is determined as the first server to be migrated from the at least one container; and determining a minimum load server from each candidate server, and determining the minimum load server as a second server to be migrated.

In some embodiments, prior to determining the first server and the second server, the method further comprises:

acquiring migration configuration information of each server in a cluster; configuring each server in the cluster according to the migration configuration information; the method comprises the steps of obtaining network configuration information of all servers in a cluster, and establishing communication connection among all the servers in the cluster based on the network configuration information.

Wherein the migration configuration information may be configuration information of a container migration tool in each server.

In practical application, according to the migration configuration information, the container migration tools in the servers in the cluster are configured, so that the container migration tools in the servers are configured in the same manner, and migration of containers between different servers can be supported. And acquiring network configuration information of each server in the cluster, and establishing communication connection among the servers based on the network configuration information, so that the container can be migrated based on the network connection among the servers.

Next, an embodiment of the present invention provides a container transfer device 70, as shown in fig. 7, and fig. 7 is a schematic structural diagram of the container transfer device provided in the embodiment of the present invention. The device comprises:

a determining module 71, configured to determine resource consumption ratios of the alternative containers deployed in the first server; determining a resource growth rate of each alternative container deployed in the first server; determining a target container to be migrated from the first server from each candidate container according to each resource consumption rate and each resource growth rate; wherein the resource consumption ratio is: a ratio between an amount of resources consumed by a corresponding candidate container and an amount of resources owned within the first server;

a migration module 72, configured to migrate the target container to a second server, where a load rate of the second server is lower than a load rate of the first server.

Optionally, the determining module 71 is specifically configured to:

determining a migration value according to each resource consumption ratio, a first weight corresponding to each resource consumption ratio, each resource growth rate and a second weight corresponding to each resource growth rate; the migration value is used for representing the migration probability of the corresponding alternative container;

and selecting the target container from the various alternative containers according to the migration value.

Optionally, the determining module 71 is further specifically configured to:

Optionally, the determining module 71 is further configured to:

Optionally, the migration module 72 is specifically configured to:

determining the change rate of dirty pages in memory pages contained in the target container;

determining a migration sequence of the memory pages of the target container to a second server according to the change rate of the dirty pages;

and sequentially migrating the memory pages of the target container to the second server according to the migration sequence.

Optionally, the migration module 72 is further configured to:

Optionally, the determining module 71 is further configured to:

determining load condition information of servers in a cluster;

and determining the first server to be migrated out of at least one container and the second server capable of accepting container migration according to a load balancing strategy and the load condition information.

Therefore, the target container with high resource consumption rate and high resource growth rate is selected from the alternative containers according to the resource consumption rate and the resource growth rate, and the target container is transferred from the first server to the second server, so that the load pressure of the first server can be relieved, and the load balance among the servers can be realized; it is also ensured that the first server is not overloaded again due to the resource growth of the container for a period of time.

With reference to the above embodiments of the present invention, an exemplary application of the embodiments of the present invention in a practical application scenario will be described below.

The present example provides a container migration method, which includes the following specific steps:

step 801, building a required environment for container migration;

in practical application, based on network configuration parameters of a server cluster, network connection between all server nodes in the cluster is established, and all server nodes in the cluster can be ensured to be communicated with each other; and synchronizing the time of each server node, and closing the firewall and the swap partition in each server node. Checking whether the software versions of the dockers of the server nodes are consistent or not, and if not, acquiring the highest version information of the software versions of the dockers of the server nodes; and upgrading the dockers of all the server nodes in the cluster to the highest version.

Step 802, determining a server to be migrated and a server to be migrated from the server cluster according to the load condition information of the servers in the cluster;

in practical application, the load condition information of the server can be referred to through a top instruction, the server in the cluster is traversed, the maximum load server and the minimum load server are determined from the cluster according to the load condition information of the server, the maximum load server is determined as a server to be migrated, and the minimum load server is determined as a server to be migrated.

In some embodiments, the pending migration server may be determined by setting a load threshold according to which the server is determined.

In practical application, acquiring load condition information of the servers in the cluster, and determining a minimum load server from the cluster according to the load condition information of the servers; determining the minimum load server as a server to be migrated; and comparing the load condition information of other servers in the cluster with a preset load threshold, and if the load condition information of the server is greater than the load threshold, determining the server as a server to be migrated.

Step 803, acquiring the resource consumption rate and the resource consumption growth rate of the alternative container deployed in the server to be migrated, and determining a target container according to the resource consumption rate and the resource consumption growth rate;

wherein the resource consumption ratio is the ratio of the resource quantity consumed by the alternative container to the resource quantity in the server to be migrated; the resource consumption increase rate is a change value of the increase of the resource data in the alternative container in unit time.

In practical application, a migration value is determined according to the resource consumption ratio of the candidate container, the first weight corresponding to the resource consumption ratio, the resource consumption growth rate, and the second weight corresponding to the resource consumption growth rate.

Here, the first weight and the second weight may be set according to user requirements.

In some embodiments, the second weight may be determined according to a service type of the alternative container.

In practical application, the second weight corresponding to the service type may be determined according to the importance of the service type provided by the alternative container. For example, if the service type provided by the alternative container is used for storing data, the second weight corresponding to the service type may be set to a smaller number; if the service type provided by the alternative container is used for cloud computing, the second weight corresponding to the service type may be set to a larger number.

Here, the importance corresponding to different service types may be determined according to user requirements, and then the second weights corresponding to different service types may be set, which is not limited in the present application.

In some embodiments, the resource consumption ratios of the alternative containers deployed in the server to be migrated may be obtained first, and according to the resource consumption ratios, the N alternative containers with the highest resource consumption ratios may be selected; acquiring the resource consumption growth rate of the N candidate containers, determining the migration values of the N candidate containers according to the resource consumption rate of the N candidate containers and the resource consumption growth rate, and determining a target container from the N candidate containers according to the migration values. Wherein N is an integer greater than 1.

For example, as shown in fig. 8, fig. 8 is a flowchart illustrating a target container determination method according to an embodiment of the present invention. Acquiring the load rate of the servers in the cluster, and determining a minimum load server from the cluster according to the load rate of the servers; determining the minimum load server as a server to be migrated; and comparing the load rates of other servers in the cluster with a preset load threshold, and if the load rate of the server is greater than the load threshold, determining the server as a server to be migrated. Acquiring the resource consumption ratio of each alternative container in the server to be migrated, sequencing the resource consumption ratios of the alternative containers from large to small, and determining the first 3 alternative containers with the highest resource consumption ratio; acquiring the resource consumption growth rate of the 3 alternative containers, determining the migration values of the 3 alternative containers according to the sum of the resource consumption growth rate and the resource consumption rate, sequencing the migration values, and determining the target container according to the sequencing result.

Step 804, migrating the target container to the server to be migrated.

In practical application, the target container contains static data and dynamic data, wherein the static data can be an image file of the target container; the dynamic data is data generated when the target container runs and is stored in a memory page contained in the target container.

In some embodiments, the step 804 may include:

acquiring the change rate of dirty pages in the memory pages contained in the target container;

determining a migration sequence of the memory pages in the target container to the server to be migrated according to the change rate of the dirty pages;

and sequentially migrating the memory pages of the target container to the server to be migrated according to the migration sequence.

And the dirty page is a memory page with inconsistent storage data and disk storage data.

In practical application, by monitoring the memory pages contained in the target container, if the memory pages are changed, recording the memory pages into a dirty page set; counting the change rate of each dirty page in the dirty page set, sorting the dirty pages in the dirty page set according to the change rate, and determining the migration sequence of the dirty pages in the dirty page set according to the sorting result; firstly, preferentially migrating memory pages except dirty pages in memory pages contained in a target container to a server to be migrated, and migrating the dirty pages in a dirty page set to the server to be migrated according to a migration sequence after the memory pages except the dirty pages are successfully migrated.

The dirty page change rate is sizeof (second-time memory page data collection — first-time memory page data collection) -sizeof (first-time memory page data collection amount).

Here, the sizeof () is used to calculate the number of bytes of the data. Determining the dirty page change rate according to the ratio of the number of bytes corresponding to the changed data in the dirty pages acquired twice to the number of bytes corresponding to the dirty page data acquired for the first time.

In some embodiments, 3 types of regions may be established in the dirty page set, respectively a high change region, a transitional change region, and a no change region; and determining the area corresponding to the dirty page according to the change rate of the dirty page.

In practical application, if the change rate of the dirty pages is greater than a first threshold, indicating that the change frequency of the dirty pages is high, storing the dirty pages into a high change area; if the change rate of the dirty pages is smaller than a first threshold and larger than a second threshold, the dirty pages are changed, but the change frequency is not high, and the dirty pages are stored in a transition change area; and if the change rate of the dirty pages is smaller than a second threshold, indicating that the change frequency of the dirty pages is very small, and storing the dirty pages into a low change area. And determining the migration sequence of the dirty pages according to the regions corresponding to the dirty pages, wherein the migration sequence of the low change region is earlier than that of the transition change region, and the migration sequence of the transition change region is earlier than that of the transition change region.

In practical implementation, the first threshold and the second threshold may be set according to user requirements, for example, the first threshold is 50%, and the second threshold is 10%. This is not limited in the embodiments of the present invention.

In some embodiments, the migration order of the dirty pages in each region may be determined according to the change rate of the dirty pages.

In practical application, the dirty pages in each region can be sorted according to the change rate of the dirty pages in each region, and the migration sequence corresponding to the dirty pages in each region is determined according to the sorting result. In this way, the dirty pages with the minimum change rate in the low change area are migrated, and the dirty pages with the maximum change rate in the high change area are sequentially migrated.

In some embodiments, after the target container is determined, static data in the target container may be migrated preferentially; after the static data is successfully migrated, preferentially migrating memory pages except dirty pages in the memory pages contained in the target container; and after the memory pages except the dirty pages are successfully migrated, sequentially migrating the dirty pages in the target container according to the migration sequence corresponding to the dirty pages, and after the migration of the dirty pages is successful, pulling up the service of the target container in the server to be migrated, and deleting the target container in the server to be migrated.

For example, as shown in fig. 9, fig. 9 is a schematic flowchart of a target container migration method according to an embodiment of the present invention. In the migration preparation stage, firstly, according to the determined server to be migrated and the server to be migrated, establishing socket communication connection between the server to be migrated and the server to be migrated; the method comprises the steps that a to-be-migrated server sends a container migration request to a to-be-migrated server, and after receiving a container migration response of the to-be-migrated server, static data (such as images, volumes, config files and the like) of a target container are transmitted to the to-be-migrated server through a socket; after the static data of the target container is successfully transmitted, transmitting the memory page of the target container; obtaining the change rate of the dirty pages in the memory pages, classifying the dirty pages according to the first threshold and the second threshold, and sequencing the dirty pages in each region according to the change rate to obtain the transmission sequence of the dirty pages; preferentially transmitting the memory pages except the dirty pages in the memory pages to a server to be migrated, after the memory pages except the dirty pages are successfully transmitted, transmitting the dirty pages in the memory pages to the server to be migrated according to the transmission sequence, after the dirty pages are successfully transmitted, operating the service of the target container in the server to be migrated, and deleting the information of the target container in the server to be migrated.

An embodiment of the present invention further provides an electronic device, where the electronic device includes:

a memory for storing executable instructions;

and the processor is used for implementing the container migration method provided by the embodiment of the invention when executing the executable instructions stored in the memory.

The hardware structure of an electronic device provided by the embodiment of the present invention is described in detail below, where the electronic device includes, but is not limited to, a server or a terminal. The electronic device includes: the at least one processor, the memory, and optionally the electronic device, may further comprise at least one communication interface, the various components of the electronic device being coupled together by a bus system, it being understood that the bus system is used to enable connected communication between these components. The bus system includes a power bus, a control bus, and a status signal bus in addition to a data bus.

It will be appreciated that the memory can be either volatile memory or nonvolatile memory, and can include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical disk, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface storage may be disk storage or tape storage. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced DRAM), Synchronous Dynamic Random Access Memory (SLDRAM), Direct Memory (DRmb Access), and Random Access Memory (DRAM). The described memory for embodiments of the present invention is intended to comprise, without being limited to, these and any other suitable types of memory.

The memory in embodiments of the present invention is used to store various types of data to support the operation of the electronic device. Examples of such data include: any computer program for operating on an electronic device, the program implementing the method of an embodiment of the invention, may be embodied in a memory.

The method disclosed by the embodiment of the invention can be applied to a processor or realized by the processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software. The Processor may be a general purpose Processor, a Digital Signal Processor (DSP), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. The processor may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed by the embodiment of the invention can be directly implemented by a hardware decoding processor, or can be implemented by combining hardware and software modules in the decoding processor. The software modules may be located in a storage medium having a memory and a processor reading the information in the memory and combining the hardware to perform the steps of the method.

In an exemplary embodiment, the electronic Device may be implemented by one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), Field-Programmable Gate arrays (FPGAs), general purpose processors, controllers, Micro Controllers (MCUs), microprocessors (microprocessors), or other electronic components for performing the above-described methods.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms. The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment. In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

An embodiment of the present invention further provides a computer storage medium, where a computer program is stored, and after the computer program is executed by a processor, the computer program executes a container migration method provided in one or more of the foregoing technical solutions, for example, the method shown in fig. 4 may be executed.

The computer storage medium provided by the embodiment of the invention comprises: a mobile storage device, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. Alternatively, the computer storage medium may be a non-transitory storage medium. The non-transitory storage medium herein may also be referred to as a non-volatile storage medium.

In some embodiments, the computer-readable storage medium may be memory such as FRAM, ROM, PROM, EP ROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or may be various devices including one or any combination of the above memories. The computer may be a variety of computing devices including intelligent terminals and servers.

In some embodiments, executable instructions may be written in any form of programming language (including compiled or interpreted languages), in the form of programs, software modules, scripts or code, and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

By way of example, executable instructions may correspond, but do not necessarily have to correspond, to files in a file system, and may be stored in a portion of a file that holds other programs or data, such as in one or more scripts in a hypertext Markup Language (H TML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code).

By way of example, executable instructions may be deployed to be executed on one computing device or on multiple computing devices at one site or distributed across multiple sites and interconnected by a communication network.

The above description is only an example of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and scope of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method of container transfer, comprising:

2. The method of claim 1, wherein determining the target container to migrate from the candidate containers to the first server according to the resource consumption rates and the resource growth rates comprises:

3. The method of claim 2, wherein determining the target container to migrate from the candidate containers to the first server according to the resource consumption rates and the resource growth rates comprises:

4. The method of claim 2, further comprising:

5. The method of claim 1, wherein the migrating the target container to a second server comprises:

6. The method according to claim 5, wherein the determining, according to the change rate of the dirty pages, a migration order in which the multiple memory pages of the target container are migrated to the second server includes:

7. The method according to claim 6, wherein the migrating the memory pages of the target container to the second server in sequence according to the migration order includes:

8. The method of claim 1, wherein prior to said determining resource consumption ratios for each alternative container deployed within the first server, the method further comprises:

determining load condition information of servers in a cluster;

9. A container transfer apparatus, comprising:

a determining module for determining resource consumption ratios of the alternative containers deployed within the first server; determining a resource growth rate of each alternative container deployed in the first server; determining a target container to be migrated from the first server from each alternative container according to each resource consumption rate and each resource growth rate; wherein the resource consumption ratio is: a ratio between an amount of resources consumed by a corresponding candidate container and an amount of resources owned within the first server;

the migration module is used for migrating the target container to a second server; wherein the load rate of the second server is lower than the load rate of the first server.

10. An electronic device, comprising:

a memory for storing executable instructions;

a processor for implementing the container migration method of any one of claims 1-8 when executing executable instructions stored in the memory.

11. A computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, implement the container migration method of any one of claims 1-8.