CONTAINER REARRANGEMENT SYSTEM, CONTAINER REARRANGEMENT METHOD, AND CONTAINER REARRANGEMENT PROGRAM
The system addresses the issue of excessive container network communication by relocating containers based on data traffic thresholds, reducing network load and enhancing application performance.
Patent Information
- Application Number
- JP2020179314
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-10-27
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2040-10-27
AI Technical Summary
Existing container relocation systems do not consider the amount of communication between containers, leading to increased network data communication.
A system that measures and relocates containers based on data communication thresholds to reduce network communication by placing excess traffic containers on the same server.
Reduces data communication between containers via the network, improving application performance by minimizing network load.
Smart Images

Figure 0007585709000001 
Figure 0007585709000002 
Figure 0007585709000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a container rearrangement system, a server, a container rearrangement method, and a container rearrangement program for rearranging a container. [Background technology]
[0002] In recent years, microservices architecture, which is composed of multiple services, has been attracting attention. Generally, in a microservices architecture, an application is composed of multiple containers. Various techniques have been proposed to relocate such containers.
[0003] As an example of a technology for relocating containers, Patent Document 1 discloses a migration management system in which a management server collects resource usage information from multiple operating servers on which containers are placed, and relocates containers so as to reduce the remaining CPU and memory capacity of the operating servers. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] JP 2017-146791 A Summary of the Invention [Problem to be solved by the invention]
[0005] However, the migration management system disclosed in Patent Document 1 has the problem that it is not possible to reduce the amount of communication between multiple containers over a network because it does not take into account the amount of communication between each container when relocating the container.
[0006] In view of the above-mentioned problems, an object of the present invention is to provide a container rearrangement system, a server, a container rearrangement method, and a container rearrangement program that are capable of reducing the amount of data communication between multiple containers via a network. [Means for solving the problem]
[0007] A container relocation system according to an exemplary embodiment of the present invention comprises: A first server and a plurality of second servers; Each of the plurality of second servers a container spawned by the first server; a communication traffic measurement unit that is generated in association with the container and measures data communication traffic between a plurality of containers owned by a plurality of second servers, The first server is a communication information acquisition unit that acquires, from a plurality of communication traffic measurement units associated with each of the plurality of containers, a data communication traffic between the container associated with the communication traffic measurement unit and another container; and a container rearrangement unit that, when a data communication volume between containers generated on different second servers exceeds a threshold, rearranges at least one of a pair of communication volume excess containers associated with the data communication volume exceeding the threshold so that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are placed on the same second server.
[0008] According to an exemplary embodiment of the present invention, the server a communication information acquiring unit that acquires, from a plurality of servers, data communication volumes between a plurality of containers generated in each of the plurality of servers; The system further includes a container rearrangement unit that, when a data communication volume between containers generated on different servers exceeds a threshold, rearranges at least one of a pair of communication volume excess containers associated with the data communication volume exceeding the threshold so that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are arranged on the same server.
[0009] A method for rearranging containers according to an exemplary embodiment of the present invention includes the steps of: Obtaining data traffic between a plurality of containers generated on each of the plurality of servers from the plurality of servers; Determine whether or not a data communication amount between containers generated on different servers exceeds a threshold value; When a data traffic amount exceeds the threshold, at least one of the pair of traffic excess containers related to the data traffic amount exceeding the threshold is relocated so that the pair of traffic excess containers related to the data traffic amount exceeding the threshold is located on the same server.
[0010] A container rearrangement program executed in an information processing device according to an exemplary embodiment of the present invention includes the steps of: acquiring, from the plurality of servers, data communication amounts between the plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between containers generated on different servers exceeds a threshold; If there is a data traffic volume exceeding the threshold, the step of relocating at least one of the pair of excess communication volume containers associated with the data traffic volume exceeding the threshold is executed so that the pair of excess communication volume containers associated with the data traffic volume exceeding the threshold are located on the same server. Effect of the Invention
[0011] The present invention makes it possible to provide a container rearrangement system, a server, a container rearrangement method, and a container rearrangement program that are capable of reducing the amount of data communication between multiple containers via a network. [Brief description of the drawings]
[0012] [Figure 1] FIG. 1 illustrates a container relocation system according to an exemplary embodiment of the present invention. [Diagram 2] FIG. 2 illustrates a configuration of a master node according to an exemplary embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram illustrating a configuration of a work node according to an exemplary embodiment of the present invention. [Figure 4] FIG. 13 is a diagram illustrating an example of an upper limit threshold and a lower limit threshold. [Diagram 5]FIG. 11 is a diagram illustrating an example of a maximum number of containers and a minimum number of containers. [Figure 6] FIG. 11 is a diagram illustrating an example of communication information. [Figure 7] FIG. 1 illustrates major components of a container relocation system according to an exemplary embodiment of the present invention. [Figure 8] FIG. 2 is a sequence diagram showing an example of a process executed in the container rearrangement system according to an exemplary embodiment of the present invention. [Figure 9] 11 is a flowchart illustrating a process performed by a master node according to an exemplary embodiment of the present invention. [Figure 10] 11 is a flowchart illustrating a process performed by a master node according to an exemplary embodiment of the present invention. [Figure 11] 11 is a flowchart illustrating a process performed by a master node according to an exemplary embodiment of the present invention. [Figure 12] 1A and 1B are conceptual diagrams illustrating an example of a cluster environment before and after a container is relocated; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0013] An exemplary embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing a container rearrangement system 1 according to an exemplary embodiment of the present invention. The container rearrangement system 1 includes a cluster environment 2 and a management device 30. The cluster environment 2 includes a master node (MN) 10, work nodes (WN) 20 to 22, and a network 40. Although three work nodes are shown in FIG. 1, any number of work nodes can be installed in the cluster environment 2.
[0014] The master node 10 is a server that manages the work nodes 20 to 22. The master node 10 can rearrange containers by having the work nodes 20 to 22 generate and delete containers. The master node 10 corresponds to a first server.
[0015] The work nodes 20 to 22 generate and delete containers based on a request from the master node 10. The work nodes 20 to 22 correspond to second servers. The master node 10 and the work nodes 20 to 22 can communicate data with each other via a network 40. The network 40 can be configured to include a WAN (Wide Area Network) and a LAN (Local Area Network).
[0016] The management device 30 is an information processing device that controls the master node 10. The management device 30 can communicate data with the master node 10 via a network 50. The network 50 can be configured to include a WAN or a LAN.
[0017] The management device 30 includes an instruction receiving unit 31 that receives an instruction from a user regarding the rearrangement of containers. The instruction receiving unit 31 can be realized by a program executed by a processor such as a central processing unit (CPU) or a micro processing unit (MPU) included in the management device 30.
[0018] The instruction receiving unit 31 receives from a user an instruction to register or update the upper and lower thresholds and the maximum and minimum numbers of containers. The upper and lower thresholds and the maximum and minimum numbers of containers are used when the master node 10 determines the containers to be rearranged and the work nodes that should generate the containers to be rearranged. Upon receiving these instructions, the instruction receiving unit 31 transmits a request to register and update this information to the master node 10.
[0019] 2 is a diagram showing the configuration of the master node 10. The master node 10 includes a computing device 100, a communication interface (I / F) 110, and a storage device 120.
[0020] The arithmetic device 100 is a processor such as a CPU or an MPU that controls devices and circuits included in the master node 10. The arithmetic device 100 executes a container rearrangement method by reading and executing a container rearrangement program from the storage device 120. The container rearrangement program includes a request reception unit 101, a communication information acquisition unit 102, and a container rearrangement unit 103. The functions of these container rearrangement programs may be realized by integrated circuits such as a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC).
[0021] The request receiving unit 101 is a program that receives a request for rearrangement of a container from the management device 30. Specifically, when the request receiving unit 101 receives a request to register an upper limit threshold and a lower limit threshold from the management device 30, the request receiving unit 101 registers the upper limit threshold and the lower limit threshold in the storage device 120. Furthermore, when the request receiving unit 101 receives a request to update the upper limit threshold and the lower limit threshold, the request receiving unit 101 updates the upper limit threshold and the lower limit threshold stored in the storage device 120.
[0022] 4 shows an example of the upper limit threshold and the lower limit threshold. The upper limit threshold and the lower limit threshold are thresholds that define the maximum and minimum values of the amount of data communication between multiple containers permitted in the cluster environment 2. The upper limit threshold and the lower limit threshold are used when the container rearrangement unit 103 determines the amount of data communication between multiple containers.
[0023] When the request receiving unit 101 receives a request to register the maximum number of containers and the minimum number of containers from the management device 30, the request receiving unit 101 registers the maximum number of containers and the minimum number of containers in the storage device 120. Furthermore, when the request receiving unit 101 receives a request to update the maximum number of containers and the minimum number of containers, the request receiving unit 101 updates the maximum number of containers and the minimum number of containers stored in the storage device 120.
[0024] 5 shows an example of the maximum number of containers and the minimum number of containers. The maximum number of containers and the minimum number of containers are the maximum number of containers and the minimum number of containers permitted in each of the work nodes 20 to 22. The maximum number of containers and the minimum number of containers are used when the container rearrangement unit 103 determines the number of containers in the work nodes 20 to 22. An individual maximum number of containers and minimum number of containers are assigned to each of the work nodes 20 to 22.
[0025] The communication information acquiring unit 102 is a program that acquires communication information including data communication volume between multiple containers generated in the work nodes 20 to 22 from the work nodes 20 to 22. Upon receiving communication information from the work nodes 20 to 22, the communication information acquiring unit 102 stores the communication information in the storage device 120.
[0026] An example of communication information is shown in Fig. 6. The communication information includes identification information of a source container, identification information of a source work node, identification information of a destination container, identification information of a destination work node, and the amount of data communication between these containers.
[0027] The container rearrangement unit 103 is a program that rearranges containers generated on the work nodes 20 to 22. The container rearrangement unit 103 determines a container to be rearranged and a work node in which the container to be rearranged should be generated, based on the data communication volume between a plurality of containers generated on different work nodes 20 to 22. Specifically, the container rearrangement unit 103 rearranges at least one of a pair of containers (hereinafter referred to as "communication volume excess containers") associated with a data communication volume exceeding an upper limit threshold, so that the pair of containers are arranged on the same work node. The container rearrangement unit 103 transmits container creation requests and deletion requests to the work nodes 20 to 22.
[0028] The communication interface 110 is a device that performs data communication between the master node 10 and the work nodes 20 to 22 and the management device 30.
[0029] The storage device 120 is a storage device in which various information such as the container rearrangement program executed by the calculation device 100, the upper and lower thresholds, the maximum and minimum numbers of containers, etc. In another embodiment, the upper and lower thresholds and the maximum and minimum numbers of containers may be stored in independent data servers. In this case, the master node 10 can access these data servers via the network 40.
[0030] 3 is a diagram showing the configuration of work node 20. Work node 20 includes a calculation device 200, a communication interface 210, and a storage device 220. Work nodes 21 and 22 have the same configuration as work node 20.
[0031] The arithmetic device 200 is a processor such as a CPU or an MPU that controls devices and circuits included in the work node 20. The arithmetic device 200 generates containers 201 and 202 and communication traffic measurement units 221 and 222 by reading and executing a program for generating a container and a communication traffic measurement unit from the storage device 220.
[0032] The containers 201 and 202 are processes that are generated based on a request from the master node 10 and are started on an OS (Operating System) of the work node 20. The containers 201 and 202 can be components of an application or middleware. When the work node 20 receives a request to generate a container from the master node 10, the work node 20 generates the container.
[0033] The communication traffic measurement units 221 and 222 are functional units that are generated in association with the containers 201 and 202, respectively, and measure the data traffic between the containers and other containers. The communication traffic measurement units 221 and 222 can be realized as processes started on the OS of the work node 20, similar to the containers. A specific example of the communication traffic measurement units 221 and 222 is a component that functions as a proxy for the associated container. This component can realize a service mesh in the cluster environment 2. The data traffic measured by the communication traffic measurement units 221 and 222 includes the data traffic between a pair of containers generated on the same work node and the data traffic between a pair of containers generated on two different work nodes.
[0034] When the communication traffic measurement units 221, 222 receive a request for communication information from the master node 10, they associate the data communication traffic measured by the communication traffic measurement units 221, 222 with the identification information of a pair of containers that communicated regarding the data communication traffic, and transmit this communication information to the master node 10.
[0035] The communication interface 210 is a device that performs data communication between the work nodes 20 and the master node 10. The storage device 220 is a storage device in which various information such as programs executed by the arithmetic device 200 is stored.
[0036] 7 is a diagram showing main components of a container rearrangement system 1 according to an exemplary embodiment of the present invention. The container rearrangement system 1 includes a master node 10 and a work node 20. The master node 10 includes a communication information acquisition unit 102 and a container rearrangement unit 103. The work node 20 includes containers 201, 202 and communication traffic measurement units 221, 222.
[0037] Fig. 8 is a sequence diagram showing an example of a process executed by the container rearrangement system 1 according to an exemplary embodiment of the present invention. In the example shown in Fig. 8, the container rearrangement system 1 has three work nodes 20 to 22.
[0038] In step S10, the communication traffic measurement unit of the work node 20 measures the amount of data communication between the container with which the communication traffic measurement unit is associated and other containers. In step S11, the communication traffic measurement unit of the work node 21 measures the amount of data communication between the container with which the communication traffic measurement unit is associated and other containers. In step S12, the communication traffic measurement unit of the work node 22 measures the amount of data communication between the container with which the communication traffic measurement unit is associated and other containers. The processes of steps S10 to S12 are periodically executed in each work node.
[0039] In step S13, the master node 10 transmits a request for communication information to the communication amount measurement units of the work nodes 20 to 22. Upon receiving the request for communication information, the communication amount measurement unit of the work node 20 transmits the communication information to the master node 10 in step S14. Upon receiving the request for communication information, the communication amount measurement unit of the work node 21 transmits the communication information to the master node 10 in step S15. Upon receiving the request for communication information, the communication amount measurement unit of the work node 22 transmits the communication information to the master node 10 in step S16.
[0040] Upon receiving communication information from the work nodes 20 to 22, the master node 10 stores the communication information in the storage device 120 in step S17. In step S18, the master node 10 determines a container to be relocated, a work node that will generate the container to be relocated, and a work node that will delete the container to be relocated. In the example shown in Fig. 8, it is assumed that the container to be relocated has been generated in the work node 22, and the work node 20 is the work node that should generate the container to be relocated.
[0041] In step S19, the master node 10 transmits a request to generate a container to be rearranged to the work node 20. Upon receiving the request to generate a container to be rearranged, the work node 20 generates the container to be rearranged in step S20 and generates a communication traffic meter in association with the container.
[0042] In step S21, the master node 10 transmits a deletion request for the container to be rearranged to the work node 22. Upon receiving the deletion request for the container to be rearranged, the work node 22 deletes the container to be rearranged and deletes the communication traffic meter associated with the container in step S22.
[0043] 9 to 11 are flowcharts showing processing executed by the master node 10 according to an exemplary embodiment of the present invention. In step S101, the communication information acquisition unit 102 of the master node 10 requests communication information from each communication amount measurement unit of the work nodes 20 to 22. In step S102, the communication information acquisition unit 102 determines whether or not communication information has been received from each communication amount measurement unit of the work nodes 20 to 22. If communication information has not been received (NO), step S102 is executed again. On the other hand, if communication information has been received (YES), in step S103, the communication information acquisition unit 102 stores the received communication information in the storage device 120.
[0044] In step S104, the container rearrangement unit 103 refers to the communication information stored in the storage device 120 and determines whether or not any of the data communication volumes between containers generated on different work nodes exceeds the upper threshold. If there is no data communication volume that exceeds the upper threshold (NO), the process returns to step S101. On the other hand, if there is any data communication volume that exceeds the upper threshold (YES), the process branches to step S105. Here, if there are multiple data communication volumes that exceed the upper threshold, the following process is executed for the one with the largest data communication volume.
[0045] In step S105, the container rearrangement unit 103 determines whether or not there is a work node having the number of containers equal to or less than the maximum number of containers assigned to each work node when one of the two work nodes having a pair of communication volume excess containers associated with a data communication volume exceeding the upper limit threshold is generated. If there is a work node having the number of containers equal to or less than the maximum number of containers assigned to each work node (YES), the process branches to step S106.
[0046] In step S106, when one of a pair of communication volume excess containers is generated, the container rearrangement unit 103 determines whether the number of containers in both of the two work nodes is equal to or less than the maximum number of containers assigned to each work node. If the number of containers in both of the two work nodes is not equal to or less than the maximum number of containers (NO), that is, if the number of containers in one of the work nodes is equal to or less than the maximum number of containers, the process branches to step S107.
[0047] In step S107, the container rearrangement unit 103 causes the one work node, whose container number is equal to or less than the maximum container number, to generate the communication volume excess container of the other work node of the pair of communication volume excess containers, i.e., the container to be rearranged. In step S108, the container rearrangement unit 103 causes the other work node to delete the container to be rearranged, and the process returns to step S101.
[0048] In step S106, if it is determined that the number of containers in both of the two work nodes is equal to or less than the maximum number of containers (YES), the process branches to step S109. In step S109, the container rearrangement unit 103 determines whether the difference between the current number of containers and the current maximum number of containers is the same for the two work nodes. If the difference is different (NO), the process branches to step S110.
[0049] In step S110, the container rearrangement unit 103 causes the work node with the larger difference between the current number of containers and the current maximum number of containers to generate the container to be rearranged. In step S111, the container rearrangement unit 103 causes the other work node, i.e., the work node with the smaller difference between the current number of containers and the current maximum number of containers, to delete the container to be rearranged, and the process returns to step S101.
[0050] In step S109, if it is determined that the difference between the current number of containers and the current maximum number of containers in the two work nodes is the same (YES), the process branches to step S112. In step S112, the container rearrangement unit 103 causes the work node with the smallest difference between the current number of containers and the current minimum number of containers to generate a container to be rearranged. In step S113, the container rearrangement unit 103 causes the other work node, i.e., the work node with the largeest difference between the current number of containers and the current minimum number of containers, to delete the container to be rearranged, and the process returns to step S101.
[0051] In the determination in step S105, if the number of containers for each of the two work nodes is not equal to or less than the maximum number of containers (NO), the process branches to step S114. In step S114, the container rearrangement unit 103 determines whether or not there is a work node, among the two work nodes having each of a pair of communication volume excess containers, in which the number of containers when both of the pair of communication volume excess containers to be rearranged are generated is equal to or less than the maximum number of containers assigned to the other work node. If there is no work node in which the number of containers is equal to or less than the maximum number of containers (NO), the process returns to step S101. On the other hand, if there is a work node in which the number of containers is equal to or less than the maximum number of containers (YES), the process branches to step S115.
[0052] In step S115, the container rearrangement unit 103 determines whether or not there are multiple work nodes in which the number of containers is equal to or less than the maximum number of containers. If there is only one work node (NO), the process branches to step S116. In step S116, the container rearrangement unit 103 causes the work node in which the number of containers is equal to or less than the maximum number of containers to generate both of the pair of communication volume excess containers to be rearranged. In step S117, the container rearrangement unit 103 causes both of the two work nodes that each have a pair of communication volume excess containers to delete the communication volume excess containers, and the process returns to step S101.
[0053] In step S115, if there are multiple work nodes where the number of containers is equal to or less than the maximum number of containers (YES), the process branches to step S118. In step S118, the container rearrangement unit 103 determines whether the difference between the current number of containers and the current maximum number of containers is the same for multiple work nodes where the number of containers is equal to or less than the maximum number of containers. If the difference is different (NO), the process branches to step S119.
[0054] In step S119, the container rearrangement unit 103 causes the work node having the largest difference between the current number of containers and the current maximum number of containers to generate both of the pair of communication volume excess containers to be rearranged. In step S120, the container rearrangement unit 103 causes both of the two work nodes having each of the pair of communication volume excess containers to delete the corresponding communication volume excess containers, and the process returns to step S101.
[0055] In step S118, if it is determined that the difference between the current number of containers and the current maximum number of containers is the same for multiple work nodes in which the number of containers is equal to or less than the maximum number of containers (YES), the process branches to step S121. In step S121, the container rearrangement unit 103 causes the work node in which the difference between the current number of containers and the current minimum number of containers is the smallest to generate both of the pair of communication volume excess containers to be rearranged. In step S122, the container rearrangement unit 103 causes both of the two work nodes that each have a pair of communication volume excess containers to delete the communication volume excess containers, and the process returns to step S101.
[0056] FIG. 12 is a conceptual diagram showing an example of the cluster environment 2 before and after the relocation of a container. In the example shown in FIG. 12, it is assumed that the data communication volume between the container 201 of the work node 20 and the container 206 of the work node 22 exceeds the upper limit threshold before the relocation of the container. That is, the container 201 and the container 206 correspond to the containers with excess communication volume. In this case, since the container 201 and the container 206 are generated in different work nodes 20 and 22, at least one of the container 201 and the container 206 is relocated to the same work node. In the example shown in FIG. 12, the container 206 is relocated to the work node 20. As a result, the containers 201 and 206, which are containers with excess communication volume, are relocated to the same work node 20.
[0057] In the above-described embodiment, the communication information acquisition unit 102 of the first server, which is the master node 10, acquires data communication volumes between a container associated with the communication volume measurement unit and other containers from a plurality of communication volume measurement units generated in the second servers, which are the work nodes 20 to 22. When a data communication volume exceeding a threshold exists among the data communication volumes between containers generated in different second servers, the container rearrangement unit 103 of the first server rearranges at least one of a pair of communication volume excess containers associated with the data communication volume exceeding the threshold so that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are arranged in the same second server.
[0058] As a result, data communication between a pair of containers with excess communication volume, which was previously performed via the network 40, is performed in the same second server, thereby reducing the amount of data communication between the containers via the network 40. As a result, the load on the network 40 is reduced, improving the performance of applications and middleware composed of multiple containers generated on multiple second servers.
[0059] Furthermore, when a communication volume excess container is generated in the other of the two second servers, if there is a second server in which the number of containers is equal to or less than the maximum number of containers assigned to each of the second servers, the container rearrangement unit 103 causes the second server in which the number of containers is equal to or less than the maximum number of containers to generate a container to be rearranged that is the same as the communication volume excess container generated in the other second server, and causes the other second server to delete the communication volume excess container. As a result, both communication volume excess containers are arranged in one of the two second servers in which they were generated, and it is possible to reduce the amount of data communication between the containers via the network 40.
[0060] Furthermore, when a communication volume excess container owned by the other second server is generated, and the number of containers in both of the two second servers becomes equal to or less than the maximum number of containers, the container rearrangement unit 103 causes the second server having the larger first difference between the current number of containers and the current maximum number of containers to generate the container to be rearranged, and the second server having the smaller first difference to delete the communication volume excess container. As a result, both of the communication volume excess containers are arranged in the second server having the larger first difference between the current number of containers and the current maximum number of containers, of the two second servers. In other words, both of the communication volume excess containers can be arranged in the second server having a margin above the current maximum number of containers.
[0061] Furthermore, when the first difference is the same for the two second servers, the container rearrangement unit 103 causes the second server having the smaller second difference between the current number of containers and the current minimum number of containers to generate the container to be rearranged, and causes the second server having the larger second difference to delete the container with the excess communication volume. This makes it possible to arrange both of the containers with the excess communication volume in the second server having the smaller second difference between the current number of containers and the current minimum number of containers, i.e., the second server having the fewer number of containers generated.
[0062] Furthermore, for each of the two second servers in which a communication volume excess container is generated, when the number of containers in the other second server exceeds the maximum number of containers when the communication volume excess container in the other second server is generated, if there is another second server other than the two second servers in which the number of containers when both of the pair of communication volume excess containers are generated is equal to or less than the maximum number of containers assigned to the other second server, the container rearrangement unit 103 causes the other second server with the number of containers equal to or less than the maximum number of containers to generate two containers to be rearranged that are the same as the pair of communication volume excess containers, and causes the two second servers to delete the communication volume excess container. This makes it possible to arrange both of the communication volume excess containers in the other second server even if the number of containers in the two second servers exceeds the predetermined maximum number of containers due to the rearrangement of the containers.
[0063] Furthermore, if there are multiple other second servers whose number of containers when both of the pair of communication volume excess containers are generated is equal to or less than the maximum number of containers assigned to the other second servers, the container rearrangement unit 103 causes the other second server with the largest difference between the current number of containers and the current maximum number of containers to generate the two containers to be rearranged, and causes the two second servers to delete the communication volume excess containers. This makes it possible to arrange both communication volume excess containers in the second server with the largest difference between the current number of containers and the current maximum number of containers, i.e., the other second server with the most room in the current maximum number of containers.
[0064] Furthermore, when there are multiple other second servers with the same difference, the container rearrangement unit 103 causes the other second server with the smallest difference between the current number of containers and the current minimum number of containers to generate the two containers to be rearranged, and causes the two second servers to delete the container with the excess communication volume. This makes it possible to arrange both containers with the excess communication volume in the second server with the smallest difference between the current number of containers and the current minimum number of containers, among the other second servers, i.e., the second server with the smallest number of containers generated.
[0065] In the above example, the program can be stored and provided to the computer using various types of non-transitory computer readable media. The non-transitory computer readable media includes various types of tangible storage media. Examples of the non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs, CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs). The program may also be provided to the computer by various types of transitory computer readable media. Examples of the transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer readable media can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or via a wireless communication path.
[0066] Furthermore, the container rearrangement program can be executed by an information processing device including a computer, a processor, an FPGA, an ASIC, or the like.
[0067] The present invention is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit of the present invention. [Explanation of symbols]
[0068] 1. Container relocation system 2 Cluster Environment 10 Masternodes 100 Computing equipment 101 Request Reception Department 102 Communication information acquisition unit 103 Container Relocation Department 110 Communication Interface 120 Storage device 20~22 Work Nodes 200 Computing equipment 201~206 Container 210 Communication Interface 220 Storage device 221,222 Communication Traffic Measurement Department 30 Management device 31 Instruction Reception Department 40,50 Network
Claims
1. A first server and a plurality of second servers; Each of the plurality of second servers a container generated by the first server; a communication traffic measurement unit that is generated in association with the container and that measures data communication traffic between the containers owned by the second servers; The first server comprises: a communication information acquisition unit that acquires, from a plurality of the communication traffic measurement units that are associated with each of the plurality of containers, a data communication traffic between a container associated with the communication traffic measurement unit and another container; a container relocation unit that, when a data communication volume between the containers generated in different second servers exceeds a threshold, relocates at least one of the pair of communication volume excess containers associated with the data communication volume exceeding the threshold such that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are located in the same second server; When the communication volume excess container is generated in the other of the two second servers, and there is a second server in which the number of containers is equal to or less than the maximum number of containers assigned to each of the second servers, the container rearrangement unit causes the second server in which the number of containers is equal to or less than the maximum number of containers to generate a container to be rearranged that is the same as the communication volume excess container generated in the other second server, and causes the other second server to delete the communication volume excess container. Container relocation system.
2. A system including a first server and a plurality of second servers, Each of the plurality of second servers a container generated by the first server; a communication traffic measurement unit that is generated in association with the container and that measures data communication traffic between the containers owned by the second servers; The first server comprises: a communication information acquisition unit that acquires, from a plurality of the communication traffic measurement units that are associated with each of the plurality of containers, a data communication traffic between a container associated with the communication traffic measurement unit and another container; a container relocation unit that, when a data communication volume between the containers generated in different second servers exceeds a threshold, relocates at least one of the pair of communication volume excess containers associated with the data communication volume exceeding the threshold such that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are located in the same second server; when the communication volume excess container owned by the other second server is generated, and the number of containers in both of the two second servers in which the communication volume excess container was generated becomes equal to or less than the maximum number of containers assigned to each of the second servers, the container rearrangement unit causes the second server of the two second servers, which has a larger difference between the current number of containers and the current maximum number of containers, to generate a container to be rearranged that is the same as the communication volume excess container generated in the other second server, and causes the second server with the smaller difference to delete the communication volume excess container.
3. A system including a first server and a plurality of second servers, Each of the plurality of second servers a container generated by the first server; a communication traffic measurement unit that is generated in association with the container and that measures data communication traffic between the containers owned by the second servers; The first server comprises: a communication information acquisition unit that acquires, from a plurality of the communication traffic measurement units that are associated with each of the plurality of containers, a data communication traffic between a container associated with the communication traffic measurement unit and another container; a container relocation unit that, when a data communication volume between the containers generated in different second servers exceeds a threshold, relocates at least one of the pair of communication volume excess containers associated with the data communication volume exceeding the threshold such that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are located in the same second server; when the communication volume excess container owned by the other second server is generated, and the number of containers in both of the two second servers in which the communication volume excess container was generated becomes equal to or less than the maximum number of containers assigned to each of the second servers, if a first difference between the current number of containers and the current maximum number of containers is the same for the two second servers, the container rearrangement unit causes the second server of the two second servers for which a second difference between the current number of containers and the current minimum number of containers is smaller to generate a container to be rearranged that is the same as the communication volume excess container generated in the other second server, and causes the second server for which the second difference is larger to delete the communication volume excess container.
4. A system comprising a first server and a plurality of second servers, Each of the plurality of second servers a container generated by the first server; a communication traffic measurement unit that is generated in association with the container and that measures data communication traffic between the containers owned by the second servers; The first server comprises: a communication information acquisition unit that acquires, from a plurality of the communication traffic measurement units that are associated with each of the plurality of containers, a data communication traffic between a container associated with the communication traffic measurement unit and another container; a container relocation unit that, when a data communication volume between the containers generated in different second servers exceeds a threshold, relocates at least one of the pair of communication volume excess containers associated with the data communication volume exceeding the threshold such that the pair of communication volume excess containers associated with the data communication volume exceeding the threshold are located in the same second server; a container relocation system in which, for each of two second servers in which the communication volume excess container is generated, when the number of containers owned by the other second server is generated and the number of containers owned by the other second server is generated exceeds a maximum number of containers assigned to each of the second servers, and when there is another second server among the two second servers in which the number of containers when both of the pair of communication volume excess containers are generated is equal to or less than the maximum number of containers assigned to the other second server, the container relocation unit causes the other second server in which the number of containers is equal to or less than the maximum number of containers to generate two containers to be relocated that are the same as each of the pair of communication volume excess containers, and causes the two second servers to delete the communication volume excess container.
5. Among the other second servers, if there are a plurality of other second servers in which the number of containers when both of the pair of communication volume excess containers are generated is equal to or less than the maximum number of containers, The container rearrangement system according to claim 4, wherein the container rearrangement unit causes the other second server having the largest difference between the current number of containers and the current maximum number of containers to generate the two containers to be rearranged, and causes the two second servers to delete the excess communication volume container.
6. If there are a plurality of other second servers among the other second servers, the number of containers when both of the pair of communication volume excess containers are generated is equal to or less than the maximum number of containers, 5. The container rearrangement system according to claim 4, wherein, when there are multiple other second servers having the same difference between the current number of containers and the current maximum number of containers, the container rearrangement unit causes the other second server having the smallest difference between the current number of containers and the current minimum number of containers to generate the two containers to be rearranged, and causes the two second servers to delete the excess communication volume container.
7. A computer comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated on different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, it is determined whether there is a server among the two servers in which a communication volume excess container having a data communication volume exceeding the threshold is generated, the number of containers of which is equal to or less than the maximum number of containers assigned to each of the servers when the communication volume excess container owned by the other server is generated; when it is determined that there is a server where the number of containers is equal to or less than the maximum number of containers, the server where the number of containers is equal to or less than the maximum number of containers is caused to generate a container to be relocated that is the same as the communication volume excess container generated in the other server, and the other server is caused to delete the communication volume excess container; How to rearrange containers.
8. A computer comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated on different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, when a communication volume excess container having a data communication volume exceeding the threshold owned by the other server is generated, it is determined whether the number of containers in both of the two servers in which the communication volume excess container is generated is equal to or less than the maximum number of containers assigned to each of the two servers; When it is determined that the numbers of both containers will be equal to or less than the maximum number of containers, the server of the two servers, which has a larger difference between the current number of containers and the current maximum number of containers, is caused to generate a container to be relocated that is the same as the communication volume excess container generated by the other server, and the server, which has a smaller difference, is caused to delete the communication volume excess container. How to rearrange containers.
9. A computer comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated on different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, when a communication volume excess container having a data communication volume exceeding the threshold owned by the other server is generated, it is determined whether the number of containers in both of the two servers in which the communication volume excess container is generated is equal to or less than the maximum number of containers assigned to each of the servers; When it is determined that the numbers of containers of both servers will be equal to or less than the maximum number of containers, if a first difference between the current number of containers and the current maximum number of containers is the same for the two servers, the server of the two servers having a smaller second difference between the current number of containers and the current minimum number of containers is caused to generate a container to be relocated that is the same as the communication volume excess container generated by the other server, and the server of the two servers having a larger second difference is caused to delete the communication volume excess container. How to rearrange containers.
10. A computer comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated on different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, it is determined whether or not there is a server in which a communication volume excess container having a data communication volume exceeding the threshold is generated, the number of containers in the other server is equal to or less than the maximum number of containers assigned to each of the servers when the communication volume excess container in the other server is generated; When it is determined that there is no server for which the number of containers is equal to or less than the maximum number of containers, it is determined whether there is any other server other than the two servers for which the number of containers when both of the communication volume excess containers of the two servers are generated is equal to or less than the maximum number of containers assigned to the other server; when it is determined that there is another server whose number of containers is equal to or less than the maximum number of containers, generate two containers to be relocated that are identical to the two servers' containers whose communication volume exceeds the limit, in the other server whose number of containers is equal to or less than the maximum number of containers, and cause the two servers to delete the containers whose communication volume exceeds the limit. How to rearrange containers.
11. A container rearrangement program executed in an information processing device, the program comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated in different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, determining whether there is a server among the two servers in which a data communication volume excess container having a data communication volume exceeding the threshold is generated, the number of containers in the other server is equal to or less than a maximum number of containers assigned to each of the servers when the data communication volume excess container in the other server is generated; when it is determined that there is a server in which the number of containers is equal to or less than the maximum number of containers, generating a container to be relocated that is the same as the communication volume excess container generated in the other server in the server in which the number of containers is equal to or less than the maximum number of containers, and causing the other server to delete the communication volume excess container; A container relocation program that executes the following:
12. A container rearrangement program executed in an information processing device, the program comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated in different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, determining whether or not the number of containers in both of the two servers in which the communication volume excess container is generated is equal to or less than the maximum number of containers assigned to each of the two servers when a communication volume excess container in the other server is generated with a data communication volume exceeding the threshold; when it is determined that the numbers of both containers will be equal to or less than the maximum number of containers, having the server of the two servers, which has a larger difference between the current number of containers and the current maximum number of containers, generate a container to be relocated that is the same as the communication volume excess container generated by the other server, and having the server, which has a smaller difference, delete the communication volume excess container; A container relocation program that executes the following:
13. A container rearrangement program executed in an information processing device, the program comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated in different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, when a communication volume excess container with a data communication volume exceeding the threshold owned by the other server is generated, determining whether the number of containers in both of the two servers in which the communication volume excess container is generated is equal to or less than the maximum number of containers assigned to each of the servers; when it is determined that the numbers of containers of both servers will be equal to or less than the maximum number of containers, if a first difference between the current number of containers and the current maximum number of containers is the same for the two servers, causing the server of the two servers, which has a smaller second difference between the current number of containers and the current minimum number of containers, to generate a container to be relocated that is the same as the communication volume excess container generated in the other server, and causing the server, which has a larger second difference, to delete the communication volume excess container; A container relocation program that executes the following:
14. A container rearrangement program executed in an information processing device, the program comprising: acquiring, from a plurality of servers, data traffic between a plurality of containers generated in each of the plurality of servers; determining whether or not a data communication amount between the containers generated in different servers exceeds a threshold; when it is determined that there is a data communication volume exceeding the threshold, determining whether or not there is a server in which a data communication volume excess container having a data communication volume exceeding the threshold is generated, the number of containers in the other server is equal to or less than a maximum number of containers assigned to each of the servers when the other server generates the data communication volume excess container; when it is determined that there is no server for which the number of containers is equal to or less than the maximum number of containers, determining whether there is any other server other than the two servers for which the number of containers when both of the communication volume excess containers of the two servers are generated is equal to or less than the maximum number of containers assigned to the other server; when it is determined that there is another server whose number of containers is equal to or less than the maximum number of containers, generating two containers to be relocated that are identical to the two servers' containers whose communication volume exceeds the limit, and causing the two servers to delete the containers whose communication volume exceeds the limit; A container relocation program that executes the following:
Citation Information
Patent Citations
Container network management system and method with high scalability
CN105591820A
Virtual machine management method, program, and management server
JP2010140134A
In-cluster migration management system, in-cluster migration management method, management server and program
JP2017146791A
Live migration of containers based on geo-location
US20180020077A1
