JP2018156464A - Optimizing method and system for decentralized container arrangement - Google Patents

Abstract

PROBLEM TO BE SOLVED: To optimize decentralized container arrangement without using a central control system by allowing each of server resources to hold, as a block chain, a management state and fault information in container environment.SOLUTION: A plurality of server resources S1, S2, and S3 are connected to a network NW. For each server resource S, a plurality of sections (containers) functioning in the same manner as a server independent of one OS environment is ensured by container management software, and a different OS environment is realized for each container C. An individual user and a service are assigned to each container C. Each server resource S is provided with a database DB1, DB2, DB3 for managing, in a block chain, a management state of each of the containers C1 to C5 and transaction of fault information. Block addition, removal, change, and update with respect to the block chain and container control are managed by each agent E (E1, E2, E3) that operates on each server resource S.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a method and system for optimizing a distributed container arrangement, and in particular, without using a central management system because each server resource holds transactions such as management status and failure information as a block chain in a container environment. In particular, it relates to a method and system for optimizing distributed container placement.

  In Patent Document 1, server resource information is collected by a load information acquisition device or the like regarding the optimal placement of virtual machines, and the server resources for placing virtual machines are determined by analyzing the information, and a control device is used. A technique for arranging virtual machines is disclosed.

JP 2014-056335 A

  According to Patent Document 1, it is possible to arrange containers on an optimal server resource, but the following technical problems are seen.

  First, it is impossible to implement optimal container placement on server resources when a failure occurs in a control device or load information collection device, when switching occurs, or during maintenance.

  Secondly, it is necessary to appropriately collect information on the server resource group to be managed, which may not only take time for virtual machine placement calculation processing, but also makes it difficult to increase the scale of server resources flexibly. .

  The object of the present invention is to solve the above technical problems and to optimize the distributed container arrangement without using a central management system by holding each server resource as a blockchain in the management environment and failure information in the container environment It is to provide a method and system that can.

  In order to achieve the above object, the present invention is characterized in that it has the following configuration in an optimization system for distributed container arrangement using a block chain.

  (1) In a plurality of server resources on which container management software operates, a block chain for managing blocks that register container management transactions by connecting them in a chain, and an agent for managing the block chain in each server resource, Each server resource agent cooperated to make each blockchain consistent.

  (2) Upon receiving the container creation instruction, the server resource agent determines the server resource for creating the container based on the blockchain under management, requests container creation, and receives a container creation completion notification in response to the request Then, a new block whose registered contents are the hash value calculated for the container-created transaction and the block at the end of the block chain under management is added to the end of the block chain, and the new resource is added to each server resource agent on the network. The agent notified of the addition of a block and notified of the addition of the new block added the new block to the end of the block chain under management.

  (3) A server resource agent newly connected to the network requests a block chain from any other server resource agent on the network, puts the block chain acquired by the request under management, and A new block whose registered contents are a hash value calculated for a transaction related to a secure connection and a block at the end of the block chain under management is added to the end of the block chain under management and the new block is added to each server resource agent on the network. The agent notified of the addition of the block and notified of the addition of the new block adds the new block to the end of the block chain under management.

  According to the present invention, the following effects are achieved.

  (1) In a container environment in which multiple containers are distributed and operated on multiple server resources, the management status and failure information are managed by blockchain on each server resource, so there is no need to use a central management system. Even when a failure occurs in the control device or load information collection device, switching occurs, or maintenance is performed, it is possible to optimally place containers on server resources.

  (2) In a container environment in which multiple containers are distributed and operated on multiple server resources, the management status and failure information are managed by blockchain on each server resource. The resource can recognize the optimum placement destination of the container by referring to the block chain under the management. Therefore, an optimum container arrangement can be achieved in a short time regardless of the server resource that receives the container addition request.

  (3) In a container environment in which multiple containers are distributed and operated on multiple server resources, the management status and failure information are managed by blockchain on each server resource, so server resources newly connected to the network Can acquire the management status and failure information of the container environment simply by acquiring the block chain from other arbitrary server resources, and the scale of the server resources can be flexibly increased.

It is the block diagram which showed the structure of the distributed container arrangement | positioning optimization system by the block chain which concerns on one Embodiment of this invention. It is the figure which showed typically the structure of the block chain which the database of each server resource manages. This is a sequence showing a procedure for optimizing the arrangement of containers to be added to server resources and reflecting the transaction in the block chain. This is a sequence showing a procedure for newly adding a server resource to the network and reflecting the transaction on the block chain under the management of each server resource.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a distributed container placement optimization system using a block chain according to an embodiment of the present invention. Here, configurations unnecessary for explaining the present invention are omitted. Yes.

  A plurality of server resources S (S1, S2, S3) having IP reachability are connected to the network NW. Each server resource S has a plurality of partitions (containers) that behave in the same manner as an independent server in one OS environment by container management software, and different OS environments are realized for each container. Each container C is assigned an individual user or service.

  In this embodiment, the server resources S1, S2, and S3 include database DBs (DB1 and DB2) that manage the management status of each container C (C1, C2, C3, C4, and C5) and transaction of failure information in a block chain , DB3) are provided respectively. Block addition, deletion, change, update, and container control for the block chain are managed by each agent E (E1, E2, E3) operating on each server resource S1, S2, S3.

  FIG. 2 is a diagram schematically showing the configuration of the block chain managed by the DB of each server resource S. The block chain is configured by connecting a plurality of blocks B (B1, B2, B3) in time series. DB manages blockchain with the same contents.

  In each block B, the creation information of the container Cn and the server resource information under management are held together with the hash value of the immediately preceding block Bn-1. When each agent E newly creates a block Bn, the hash value of the immediately preceding block Bn-1 is calculated, and the result is added to the newly created block. After the creation of the block Bn is completed, the notification to all the other agents E is performed, and the agent E that receives the notification updates the block chain. As a result, all DBs can hold consistent information and the same information.

  According to this embodiment, in a container environment in which a plurality of containers C are distributed and operated on a plurality of server resources S, the management status and failure information are managed by the blockchain on each server resource. Without using the system, it is possible to optimally place containers on server resources even when a control device or load information collection device fails, when switching occurs, or during maintenance.

  FIG. 3 is a sequence showing a procedure for optimizing the arrangement of the container C newly added to the server resource and reflecting the transaction on the block chain.

  When the operator logs in to the server resource S (here, the server resource S2) from the operation terminal at time t1 and executes the container creation command at time t2, the container creation is performed on the agent E2 of the server resource S2 at time t3. Instruction is transmitted.

  The agent E2 refers to the block chain held in the DB2 under management at time t4, and determines the number of containers on each server resource S based on the container operation information history and the like. Here, for example, when it is determined that the resource is the most empty because the number of containers on the server resource S1 is the smallest, the agent E2 requests the server resource S1 to create a container at time t5.

  The server resource S1 executes creation of container C in response to this container creation request. When the container creation is completed, at time t6, a container creation completion notification is transmitted to the agent E2 that is the container creation request source.

  Upon receiving the container creation completion notification, the agent E2 creates information indicating that the operator's container C has been created on the server resource S1 at time t7, and further calculates the hash value of the block at the end of the block chain. Then, a new block B having these registered contents is created and added to the end of the block chain stored in the DB2 under management.

  Thereafter, at time t8, the agent E2 notifies all the agents E (E1, E3) except for itself of the block addition information. Each agent E that receives the notification updates the block chain stored in the DB under management at time t9. That is, the notified new block B is added to the end of the block chain.

  According to this embodiment, in a container environment in which a plurality of containers are distributed and operated on a plurality of server resources, the management status and failure information are managed by the block chain on each server resource. The server resource that has received can recognize the optimum container placement location by referring to the block chain under management. Therefore, an optimum container arrangement can be achieved in a short time regardless of the server resource that receives the container addition request.

  FIG. 4 is a sequence showing a procedure for adding a server resource to the network and reflecting the transaction in the block chain under the management of each server resource. Here, the server resources S1 and S2 are already connected. A case where a server resource S3 is newly added to the existing network will be described as an example.

  The agent E3 of the newly connected server resource S3 transmits a block chain acquisition request to another agent E on the network NW selected at random at time t21. Here, the description is continued assuming that the agent E2 is selected.

  When agent E3 receives the block chain from agent E2 at time t22, it calculates information that server resource S3 has been added as a newly available server resource and hash value of the block at the end of the block chain at time t23. To do. Then, a new block B3 having these as registered contents is created and added to the end of the block chain stored in the DB3 under management.

  At time t24, the agent E3 performs block addition notification to all the agents E except itself. At time t25, the block chain in the DB of each of the agents E1 and E2 that has received the notification is updated.

  According to the present embodiment, in a container environment in which a plurality of containers are distributed and operated on a plurality of server resources, the management status and failure information are managed by the block chain on each server resource. The server resource to be connected can share the management state of the container environment and the failure information only by acquiring the block chain from any other server resource, and the scale of the server resource can be flexibly increased.

  In the above embodiment, the present invention has been described by taking the addition of a container and the addition of server resources as examples. However, the present invention is not limited to this, and the scope of the present invention is not deviated. It is possible to replace the constituent elements in the above-described embodiments with known constituent elements as appropriate. The technical scope of the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.

  B (B1, B2, B3) ... Block, C (C1, C2, C3, C4, C5) ... Container, DB (DB1, DB2, DB3) ... Database, E (E1, E2, E3) ... Agent, NW ... Network, S (S1, S2, S3) ... server resources

Claims (5)

In the optimization system of distributed container placement by blockchain,
Multiple server resources running container management software;
In each server resource, a block chain that manages by connecting blocks that register container-managed transactions in a chain, and
Each server resource includes an agent that manages the blockchain,
A distributed container placement optimization system in which agents of each server resource cooperate to make each block chain consistent. The server resource agent that received the container creation instruction
Determine the server resource to create the container by referring to the block chain under management,
Request container creation for the determined server resource,
When a container creation completion notification is received in response to the request, a new block having a hash value calculated for the container creation transaction and the block at the end of the block chain under management is added to the end of the block chain,
Notifying the agent of each server resource on the network that the new block has been added,
The distributed container placement optimization system according to claim 1, wherein the agent notified of the addition of the new block adds the new block to the end of a block chain under management. The server resource agent that is newly connected to the network
Request blockchain from agents of any other server resources on the network,
Place the block chain acquired by the request under management,
Add a new block whose registered content is the hash value calculated for the transaction related to the new connection and the block at the end of the block chain under management to the end of the block chain under management
Notifying the agent of each server resource on the network that the new block has been added,
3. The distributed container arrangement optimization system according to claim 1, wherein the agent notified of the addition of the new block adds the new block to the end of the block chain under management. In the optimization method of distributed container placement by blockchain,
Upon receiving the container creation instruction, the first agent of the first server resource determines the second server resource for creating the container with reference to the block chain under management.
The first agent requests the second server resource to create a container;
The second server resource creates a container and notifies the first agent of the container creation completion notification;
The first agent that has received the completion notice adds a new block having the hash value calculated for the container-created transaction and the block at the end of the block chain under management to the end of the block chain,
The first agent notifies the agent of each server resource on the network of the addition of the new block;
The distributed container arrangement optimizing method, wherein the agent notified of the addition of the new block adds the new block to the end of a block chain under management. In the optimization method of distributed container placement by blockchain,
A first agent of a first server resource newly connected to the network requests a block chain from a second agent of any other server resource on the network;
The first agent puts the block chain acquired by the request under management,
The first agent adds a new block whose registered content is a hash value calculated for a transaction related to the new connection and a block at the end of the block chain under management to the end of the block chain under management,
The first agent notifies the agent of each server resource on the network of the addition of the new block;
The distributed container arrangement optimizing method, wherein the agent notified of the addition of the new block adds the new block to the end of a block chain under management.

Images