KR20090059851A - System and method for service level management in virtualized server environment - Google Patents

Abstract

A system for managing the service level in a virtual server environment and a method thereof are provided to operate the resource utility and service availability effectively by dynamically offering system resources according to service level requirements. A virtualization target system node(130) produces a plurality of virtual systems(140) and assigns or collects/manages system resources. The system node comprises the followings. A virtual system manager(132) generates and manages the virtual systems. A system monitoring unit(133) monitors and manages resources of the virtual systems.

Description

[0001] SYSTEM AND METHOD FOR SERVICE LEVEL MANAGEMENT IN VIRTUALIZED SERVER ENVIRONMENT [0002]

The present invention relates to a system and method for managing service levels in a virtual server environment, and more particularly, to a system and method for managing a service level in a virtual server environment, The present invention relates to a service level management system and a method thereof in a virtual server environment.

Conventionally, the system specifications and service requirements that the system should provide at the time of the service level agreement (SLA) with the customer for the service level management (SLM) in the data center are determined. A service level contract is a contract between a service provider and a customer that includes services to be provided, metrics related to the service, levels of acceptable or unacceptable services, And service level management can be defined as a set of people and systems to ensure that service level agreements are satisfied and that the necessary resources are provided efficiently.

For example, while hosted services that offer today's services vary slightly from one hosted service provider to another, personalized web hosting is not limited to the maximum capacity of available storage devices, the amount of network traffic that can be transferred during a day, , The number of e-mail accounts, and so on. In addition, there are web hosting services through server leasing. In this case, fixed service can be provided according to predefined service products according to the provision of CPU, memory, storage device, network traffic, and other maintenance service.

Therefore, in order to maintain the service quality within the scope of the previously agreed system resources, it is necessary for the customer to provide the system resource at a higher level than the normal level, that is, the peak time It is inevitable to waste system resources. Or statically fixed services, there is a disadvantage that it can not be dynamically coped with when the services are concentrated in an unspecified manner.

Therefore, in recent server computing, server virtualization technology is getting attention again in order to increase the utilization rate of system resources and it is rapidly spreading to the market. Virtualization is an abstraction of computer resources in a broad sense, a technology that hides the physical characteristics of computer resources from other systems, applications, or users that interact with that computer resource. This includes both making one physical resource appear as multiple virtualized logical resources, or making multiple physical resources appear as a single virtualized logical resource.

By using the virtualization technology, it is possible not only to create and operate a plurality of virtual machines having various system specifications by sharing or partitioning resources in one physical system, A shared computer can dynamically freely move the virtual machine to another physical system while it is in operation. This has the advantage of increasing system performance, scalability, reliability, availability, flexibility, and optimizing resources. With the appeal of these virtualization technologies, today's data centers Virtualization technology is rapidly spreading as an underlying technology that can reduce total cost of ownership (TCO), one of the biggest challenges.

These virtualization technology services enter into contracts based on system resource requirements that are usually more peak time than ever for service availability, according to service level agreements between customers and service providers. As a result, since the system resources provided from the service provider are not required at all times, the waste of the system resources, that is, the resource utilization rate is kept very low, and the total cost of ownership of the data center is increased.

The present invention provides a system and method for efficiently managing resource utilization and service availability by dynamically providing and retrieving system resources according to service level requirements in order to guarantee Quality of Service (QoS) for each individual service in a data center A service level management system and a method thereof in a virtual server environment are provided.

According to another aspect of the present invention, there is provided a service level management system in a virtual server environment, comprising: a virtualization target system node for generating a plurality of virtual systems at the request of an external administrator and allocating / And a virtual system management node that manages system resources of the virtualization target system node and controls and manages a workload according to a service request.

The virtualization target system node may include a plurality of virtual systems, a virtual system management unit for creating and managing the virtual systems, a system monitoring unit for monitoring and managing the virtual system resources, It is desirable to include a host operating system that manages and controls the components.

Here, the virtual system may include: a cluster configuration unit for executing and storing a system configuration of a cluster configuration for each virtual system; a system monitoring unit for monitoring and managing system resources assigned to the virtual system; And a virtual system operating system for operating and controlling the virtual machine operating system.

The virtual system management node includes a workload reconciliation management unit that receives a request input by the external administrator, a virtual system instruction execution unit that transmits a command to the virtual system, A resource pool manager for monitoring and providing resources on a pool basis, a database for storing management information on the managed virtual system and system resources, and a management policy applied to the managed virtual system, And a cluster management unit configured to dynamically construct a cluster according to a service state using the virtual system to provide load balancing.

According to another aspect of the present invention, there is provided a method for managing a service level in a virtual server environment, comprising: installing and providing a virtual system corresponding to a basic specification for each service level; Monitoring the provided virtual system performance and resource status; Determining whether the threshold value is deviated from the threshold value in the monitoring step; Determining a service level rating provided when the threshold value is deviated; Further providing or retrieving a cluster-based virtual system according to a threshold value type when the service level rating is high; And terminating the service level management after additionally providing or retrieving the virtual system.

In the step of determining the service level class, it is determined whether the resource level is adjustable when the service level class is normal. And if the resource adjustment is possible, allocating or recovering resources of the virtual system according to the threshold value type.

Here, the step of further providing the virtual system may include registering and storing virtualization target system node information in the virtual system management node; Determining whether to designate a virtual system to be created for a requested system resource on a virtualization target system node based on the registered information; If the virtualization target system node is designated, checking whether the virtualization target system node is a resource satisfying the requested system resource requirement; Requesting the virtualization target system node to create a virtual system when the requested system resource requirement is satisfied; Storing the generated virtual system information; And storing the virtual system information, and notifying the result to the higher-level manager.

In the step of determining whether to designate a virtual system to be created as a virtualization target system node, the virtualization target system node is not designated as a virtualization target system node; And acquiring virtualization target system node information that will satisfy the requested system resource requirement.

The step of constructing a cluster using two or more virtual systems when distributing a load of a service in the step of providing and providing a virtual system according to the type of the threshold value comprises the steps of: Requesting a cluster configuration using the cluster configuration information; Transmitting a system setup request and cluster service information for the cluster creation; Executing the cluster service after setting up the cluster; Returning the execution result and storing cluster creation and service information; And storing the service information, and notifying the upper manager of the final result.

A service level management system and a method thereof in a virtual server environment of the present invention dynamically provide system resources according to service level requirements to guarantee Quality of Service (QoS) for each individual service in a data center And resource utilization and service availability can be effectively managed.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It is to be understood that the following embodiments are provided so that those skilled in the art can understand the present invention without departing from the scope and spirit of the present invention. no.

FIG. 1 and FIG. 2 are views schematically showing the configuration of a service level management system in a virtual server environment according to the present invention. 1 and 2, the service level management system includes a virtualization target system node 130 that generates a plurality of virtual systems at the request of an external administrator and allocates or reclaims system resources, A virtual system management node 120 that manages system resources of the target system node 130 and controls and manages a workload according to a service request and a hierarchical structure of the virtual system management node managing the virtualization target system node And the upper virtual system management node 110 which manages the upper virtual system.

In general, a service provider, such as a data center, has a system node to provide services, that is, a system node 130 to be virtualized. The virtualization target system node may be a physical computer system, such as a rack-mounted server, a personal desktop computer, a blade server, or the like.

Virtualization technology may be applied to these virtualization target system nodes 130 so that a plurality of virtual systems exist in one physical computer system to provide services.

A virtual system management node 120 exists for managing system resources of the virtualization target system node 130 and effectively managing workloads according to service requirements.

One virtual system management node 120 can control and manage dozens to hundreds of virtualization target system nodes 130 according to the installation and operation environment of the system. In addition, the virtual system management node 120 may have a hierarchical structure depending on the size of the virtualization target system node 130.

That is, when there are hundreds of virtualization target system nodes 130, there are dozens of virtual system management nodes 120 that manage dozens of virtualization target system nodes 130, and the top virtual system management node 110 may be superimposed on each other.

The virtual system management node 120 includes a workload tuning management unit 121 that receives a request input from a system provided or utilized by a higher virtual system management node 110, a system administrator, or a third party, A virtual system instruction execution unit 122 for sending a command to the virtual system 140 according to an autonomous action of the system based on the received request or policy, a monitoring unit 122 for monitoring the resource status of all the virtualization target system nodes 120 to be managed, A resource pool management unit 123 for managing and providing management information on the basis of a pool, a database 124 for storing management information on all the managed systems and system resources, Management rules that can be acted upon according to judgment, and so on are defined as policies, and the real-time situation of the system or service A policy management unit 125 for determining a management action to be applied to the management target system based on knowledge stored in an expression base, and requesting execution; a virtual system And a cluster manager 126 for providing service availability and reliability as well as load balancing by constructing a plurality of system nodes or virtual systems 140 as clusters dynamically according to service conditions .

The virtualization target system node 130 receives a request from a system provided or utilized by the virtual system management node 120, the system administrator, or a third party to create and delete the virtual system 140, A system monitoring unit 133 for monitoring and managing system resources of the virtualization target system node 130 and a virtual system management unit 132 and a system monitoring unit 133. The virtual system management unit 132, And a plurality of virtual systems 140 generated through the virtual system management unit 132 exist in one physical system.

The virtual system 140 includes a cluster configuration unit 142 for executing and storing system settings related to a cluster configuration for each virtual system 140 so that a cluster can be configured by a plurality of virtual systems 140, A system monitoring unit 143 for monitoring and managing system resources allocated to the virtual system 140, and a virtual system operating system 141 for operating and controlling the virtual system 140.

FIG. 3 is a flowchart of a service level management method in a virtual server environment according to the present invention. As shown in FIG. 3, a virtual system is installed and provided in accordance with the basic specification for each service class (S201).

More specifically, after preparing the virtualization target system node 130, such as a data center, the service provider determines the virtual system 140 according to the service level agreement established with the customer. At this time, the basic system resource size of the virtual system 140 is already determined according to the service class defined in advance by the service provider and agreed with the customer, and the virtual system 140 is allocated to the specific virtualization target system node 130 selected by the service provider And start to provide services to customers.

Then, the provided virtual system performance and resource status are monitored (S202). That is, when the virtual system 140 is installed and started to be serviced, the performance of the virtual system 140, the state of the system resources, and the state of the service are continuously (continuously) transmitted through the resource pool management unit 123 and the system monitoring units 133 and 143 .

Next, it is determined whether or not the threshold value set in the monitoring is deviated (S203). More specifically, the policy management unit 125 continuously refers to the virtual system 140 and the service status thereof collected by the resource pool management unit 123, and determines a threshold value, that is, a status of the virtual system resource Or when the status of the service reaches a certain point, it deviates from the reference value set so that additional resources can be allocated or retrieved.

Then, when the threshold value is deviated, the service level class is determined (S204). That is, if the virtual system 140 acquired at a certain time or its service state deviates from a predefined threshold value exceeding the upper threshold value, the policy management unit 125 may additionally request the service level agreement Determine the action to be taken, first identify the service level rating according to the service level agreement with the customer.

If the service level grade is high, the virtual system based on the cluster is additionally provided or retrieved according to the threshold value type (S205). More specifically, if the service level of the customer is high, the virtual system 140 having the basic specification prepared in advance for the purpose of switching over in the event of a system failure must be ensured to ensure maximum service availability in a short period of time. And then the existing virtual system 130 and the cluster are configured to immediately distribute the load and maintain the availability.

The system resource requirements and the like of the virtual system 140 to be further added may be variously changed according to various circumstances such as the service characteristics provided and the policy of the service provider. In this case, before the system administrator operates the system, ). ≪ / RTI > Also, it is possible to dynamically generate a virtual system in accordance with a predicted scale, rather than a virtual system 140 waiting in advance according to the system status, to provide a service by constructing a cluster with an existing virtual system 140 have.

Meanwhile, in the step of determining the service level class, if the service level class is normal, it is determined whether resource adjustment is possible (S206). Here, when the service level grade is normal, there is sufficient resources in the virtualization target system node 130 by using the dynamic resource setting method of the virtual system 140, and the amount of the maximum allocatable resources The resource of the virtual system 140 is additionally dynamically allocated (S207). At this time, the size of the resource to be further allocated is defined in advance by the system administrator through the policy management unit 125 according to the system status and the service status as described above. If there is no additional resource to allocate, but in order to guarantee the service availability, the service level level may be generated according to the content defined in the policy management unit 125, The resource may be further provided by clustering using the dynamically generated virtual system 140.

It may also take no further action if the threshold has been exceeded but the service level is not required to guarantee service availability.

On the other hand, if the virtual system 140 or its service state acquired at a certain point of time deviates from a predefined threshold value below a lower threshold value, system resources are provided in excess of the service requirement, And recovers the virtual system 140 or its resources for optimization. In this case, when the virtual system 140 is additionally provided as a cluster, only the minimum virtual system 130 that can guarantee the service among the virtual systems 130 constituting the cluster according to the estimated size of the service, The system 140 may exclude from the service and use it for other services or delete the virtual system 140 to return all the system resources in use to the virtualization target system node 130.

If the utilization rate of the system resources is very low while a service is provided to one virtual system 140 rather than a cluster, the amount of resources in use among the entire resources allocated to the virtual system 140 is slightly reduced, The system resources that are not occupied by the virtualization target system node 130 may be transferred to another virtual system 140 operating in the same virtualization target system node 130 to provide the utilization rate of the entire system resources.

Finally, after the virtual system is additionally provided or retrieved, the service level management is terminated (S208). In other words, it provides dynamic service level management and terminates all services when service level management is terminated. Otherwise, it continuously monitors system performance, resource status, and service status, and maintains service level management.

FIG. 4 is a flowchart of a method for further providing the virtual system of the present invention. As shown in FIG. 4, in the method of providing a virtual system, the virtualization target system node information is registered and stored in the virtual system management node (S401).

More specifically, the host operating system 131, the virtual system management unit 132, and the system monitoring unit 133 are installed and prepared in the virtualization target system node 130 as a method of creating and operating the virtual system, Various system information such as the system name, the central processing unit, the memory, the storage device, the installed system resource information such as the network device, the IP registration information, and the user account for the prepared virtualization target system node 130, And stores it in the database 124 for management.

In step S402, it is determined whether to assign a virtual system to be created to the virtualization target system node based on the registered information on the requested system resources. That is, the system manager accesses the system through the workload reconciliation management unit 121 of the virtual system management node 120, and based on the information registered in the database 124, 130 or designates a resource pool to be created based on the resource pool information generated in the resource pool management unit 123 and stored in the database 124 and then designates a resource pool of the virtual system to be created And the like.

If the virtualization target system node 130 is specified, the workload reconciliation management unit 121 notifies the resource pool management unit 123 of the virtualization target system node 130 to the virtual system management node 120, the system administrator, The resource pool management unit 123 confirms that the system provided or utilized by the virtualization target system node 130 has sufficient resources to satisfy the requested system resource requirements, (133) to obtain related information.

Next, if the virtualization target system node is designated (S403), it is determined whether it is a resource satisfying the requested system resource requirement (S404).

If there is a resource sufficient to satisfy the requested system resource requirement in the virtualization target system node 130 in operation S405, the workload reconciliation management unit 121 notifies the virtualization target system node 130 through the virtual system command execution unit 122, (Step S406). That is, the virtual system instruction execution unit 122 accesses the virtual system management unit 132 of the virtualization target system node 130 to request the virtual system 140 to be created.

The virtual system management unit 132 creates the virtual system 140 based on the inputted virtual system 140 setting information and transfers the result to the virtual system command execution unit 122. The generated virtual system 140 Is stored in the database 124 (S407).

After storing the virtual system information, the virtual system management node 121, the system administrator, or a system provided or utilized by a third party, which is an upper manager through the workload reconciliation management unit 121, (S408).

On the other hand, if the resource pool is not designated to the virtualization target system node (S409), the virtualization target system node information that satisfies the requested system resource requirement is acquired.

More specifically, in the workload reconciliation management unit 121, a system resource requested by a system provided or utilized by the virtual system management node 120, the system administrator, or a third party in the resource pool specified for the resource pool management unit 123 And acquires the virtualization target system node 130 information having sufficient resources to satisfy the requirements.

That is, the workload reconciliation management unit 121 transmits the information about the virtualization target system node 130 and the virtual system 140 resource requirements received from the resource pool management unit 123 to the virtual system command execution unit 122 Thereby causing the virtual system 140 to be generated.

The procedure for creating the virtual system 140 in the virtual system instruction execution unit 122 is the same as that for specifying the virtualization target system node 130 described above.

On the other hand, the deletion of the virtual system 123 or the modification of the allocated resources is similar to the creation process of the virtual system.

FIG. 5 is a flowchart of a method of constructing a cluster of the present invention. As shown in FIG. 5, in a case where the service load is distributed and provided, a process of configuring a cluster using two or more virtual systems requests a cluster configuration using two or more virtual systems existing in the same resource pool ( S501).

That is, when it is desired to distribute the load of the service provided using the virtual system 140, or to improve the availability and reliability, the cluster may be configured using two or more virtual systems 140. Therefore, when the system provided or utilized by the virtual system management node 120, the system administrator, or a third party requests the cluster configuration to the workload tuning management unit 121 for the purpose of distributing the service load or increasing availability and reliability , The workload reconciliation management unit 121 uses the system resource information generated by the resource pool management unit 123 or stored in the database 124 to configure the cluster using two or more virtual systems 140 existing in the same resource pool To the cluster management unit 126. [

Then, the system setup request and the cluster service information for the cluster creation are transmitted (S502). The cluster management unit 126 accesses the cluster configuration unit 142 of each virtual system 140 requested based on the information of the requested virtual system 140 and the service information to request the system configuration for cluster creation And transmits cluster service information.

After setting the cluster, the cluster service is executed (S503). That is, the cluster configuration unit 142 completes the requested cluster configuration, and then executes the cluster service.

Then, the execution result is returned, and cluster creation and service information are stored (S504). Here, the execution result is returned to the cluster manager 126 of the virtual system management node 120, and the received cluster creation and service information is stored in the database 124.

After storing the service information, the result is transmitted to the system provided or utilized by the virtual system management node 120, the system administrator, or a third party, which is an upper manager through the workload reconciliation management unit 121 do.

On the other hand, the system administrator can set the policy management unit 125 so that the system administrator can define management rules that can be taken immediately or repeatedly or simply from the management task, . The policy management unit 125 continuously monitors the resource statuses of the virtualization target system node 130 and the virtual system 140 through the resource pool management unit 123 and determines whether any conditions predefined by the system administrator are satisfied Based on knowledge of a knowledge base stored in the database 124, a management action to be applied to the managed system is determined and the virtual system command execution unit 122 ). ≪ / RTI >

As described above, embodiments of the present invention have been disclosed through detailed description and drawings. It is to be understood that the terminology used herein is for the purpose of describing the present invention only and is not used to limit the scope of the present invention described in the claims or the claims. Therefore, those skilled in the art will appreciate that various modifications and equivalent embodiments are possible without departing from the scope of the present invention. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

FIG. 1 and FIG. 2 are diagrams schematically showing the configuration of a service level management system in a virtual server environment according to the present invention. FIG.

FIG. 3 is a flow chart of a service level management method in a virtual server environment according to the present invention; FIG.

FIG. 4 is a flow chart of a method for further providing a virtual system of the present invention. FIG.

5 is a flowchart of a method for configuring a cluster of the present invention.

Description of the Related Art

120 --- Virtual Machine Managed Node

130 --- Virtualization target system node

140 --- Virtual systems

Claims (12)

A virtualization target system node for generating a plurality of virtual machines at the request of an external administrator and allocating or recovering system resources; And a virtual system management node for managing system resources of the virtualization target system node and controlling and managing a workload according to a service request. The method according to claim 1, Further comprising a higher virtual system management node managing a hierarchical structure of the virtual system management node managing the virtualization target system node. The virtualization target system node according to claim 1, A plurality of virtual systems, A virtual system management unit for creating and managing the virtual system, A system monitoring unit for monitoring and managing the virtual system resources; And a host operating system for operating and controlling the virtual system management unit and the system monitoring unit. 4. The system of claim 3, A cluster configuration unit for executing and storing a system configuration of the cluster configuration for each virtual system; A system monitoring unit for monitoring and managing system resources allocated to the virtual system; And a virtual system operating system for operating and controlling the virtual system. 2. The system according to claim 1, A workload tuning management unit receiving a request input by the external manager, A virtual system instruction execution unit for transmitting a command to the virtual system; A resource pool management unit for monitoring the resource status of the virtualization target system node and managing and providing resources based on the pool; A database unit for storing management information on the managed virtual system and system resources; A policy management unit for defining a management policy to be applied to the management target virtual system and requesting execution, And a cluster manager for dynamically configuring a cluster according to a service state using the virtual system to provide load balancing. 6. The method according to claim 1 or 5, Wherein the external manager comprises a system provided by a higher virtual system management node, a system administrator, or a third party. Installing and providing a virtual system corresponding to a basic specification for each service class; Monitoring the provided virtual system performance and resource status; Determining whether the threshold value is deviated from the threshold value in the monitoring step; Determining a service level rating provided when the threshold value is deviated; Further providing or retrieving a cluster-based virtual system according to a threshold value type when the service level rating is high; And terminating the service level management after additionally providing or retrieving the virtual system. 8. The method of claim 7, wherein in determining the service level rating, Determining whether resource adjustment is possible if the service level rating is normal; And if the resource can be adjusted, allocating or withdrawing resources of a virtual system according to a threshold value type. 8. The method as claimed in claim 7, Registering and storing virtualization target system node information in a virtual system management node; Determining whether to designate a virtual system to be created for a requested system resource on a virtualization target system node based on the registered information; If the virtualization target system node is designated, checking whether the virtualization target system node is a resource satisfying the requested system resource requirement; Requesting the virtualization target system node to create a virtual system when the requested system resource requirement is satisfied; Storing the generated virtual system information; And storing the virtual system information, and notifying the result to a higher-level manager. 10. The method of claim 9, Determining whether to designate a virtual system to be created as a virtualization target system node, and designating the virtualization target system node as a resource pool without designating the virtualization target system node; And acquiring virtualization target system node information that satisfies the requested system resource requirement. 8. The method of claim 7, further comprising: Wherein when the load of the service is distributed and provided, the cluster is configured using two or more virtual systems. 12. The method of claim 11, wherein configuring the cluster comprises: Requesting a cluster configuration using two or more virtual systems existing in the same resource pool; Transmitting a system setup request and cluster service information for the cluster creation; Executing the cluster service after setting up the cluster; Returning the execution result and storing cluster creation and service information; And storing the service information, and notifying an upper manager of the final result after storing the service information.

Images