KR20140117905A - Virtual machine allcoation of cloud service for fuzzy logic driven virtual machine resource evaluation apparatus and method - Google Patents

Abstract

An embodiment of the present invention relates to a virtual resource allocation apparatus and method in a cloud service using a fuzzy logic driven resource evaluation method based on a virtual resource allocation technique which defines characteristics of elements constituting a cloud environment in the cloud service and correlations between the characteristics of each element. Provided is the fuzzy logic driven resource evaluation apparatus for a virtual resource allocation of the cloud service, comprising an application program service request part receiving a service request in the cloud service; an application program allocation part allocating an application program according to the received service request; a virtual resource allocation part allocating a virtual resource according to the allocation of the application program; a resource infrastructure part processing work according to the allocation of the virtual resource; and a virtual resource evaluation part evaluating the availability of the virtual resource.

Description

FIELD OF THE INVENTION [0001] The present invention relates to a fuzzy logic-based resource evaluation apparatus and method for allocating virtual resources of a cloud service,

The embodiment of the present invention is based on the virtual resource allocation technique that defines the characteristics of the elements constituting the cloud environment in the cloud service and the correlations between the characteristics of the respective elements and uses the fuzzy logic- Apparatus and method.

Increasingly large amounts of data and methods for dealing with them are becoming issues. Cloud computing uses virtualized environments to organize resource infrastructures, reduce consumption costs of computing environments, increase scalability, flexibility, and manage resources integrally. It is receiving great attention.

Virtualization can improve resource utilization by enabling multiple virtual resources to be shared by more than one physical server, thereby improving the overall performance of the cluster by reducing operating costs. However, in a virtualization environment, resource utilization may be concentrated on a specific server among the physical servers constituted by clusters, which causes deterioration of the overall performance. To solve this problem, we need a resource evaluation method that can analyze the factors that degrade overall performance of virtual resources and reduce bias on specific resources for the next task.

The resource evaluation method can reduce the performance degradation problem because the resource load can be evenly distributed. However, in order to evaluate the resources of virtual resources, it is necessary to evaluate resources according to the state of CPU, RAM, and Net I / O of each virtual node, but the criteria are unclear.

Therefore, it is necessary to determine the load balancing according to the performance of the virtual resource. If the next load balancing is requested according to the determined load balancing, the work should be allocated to the virtual resource having a low load. In other words, for job scheduling in the cloud environment, it is necessary to evaluate the specific availability of the ambiguous state according to the performance of the virtual resources.

As an example of a technology for a work allocation technique of a virtual resource, Korean Patent Laid-Open No. 10-2009-0068833 (published on June 29, 2009) entitled "Program Execution System Based on Virtual Machine Monitor and Its Control Method" A virtual machine monitor based program execution system for executing an application program in any one of the plurality of domains in accordance with a trust level of an application program executed using a virtual monitor is described.

On the other hand, fuzzy logic for solving an uncertain state is a representative technique that can be inferred from ambiguous data, and has many merits in making decisions about uncertain data. Therefore, fuzzy logic is needed for resource evaluation of virtual resources. In addition, to optimize cloud system availability, virtual resource allocation methods are required, and new methods and policies must be defined to design cloud computing methods.

Since the required data throughput varies according to the service provision method, the virtual resource evaluation method must consider the situation that meets each condition. Otherwise, it can not provide high quality service and the load of the cloud server device increases.

Therefore, fuzzy logic-based resource evaluation devices and methods for virtual resource allocation are needed to optimize availability in cloud services.

One embodiment of the present invention is to analyze components and features for virtual resource evaluation and method in order to smoothly provide cloud service and to increase availability, and to analyze the status of each virtual resource using a fuzzy logic method, A fuzzy logic technique based virtual resource evaluation apparatus and method for a service are provided.

A fuzzy logic-based resource evaluation apparatus for allocating virtual resources of a cloud service, the apparatus comprising: an application program service request unit for receiving a service request from a cloud service; An application allocation unit for allocating an application program according to the obtained service request; A virtual resource allocation unit allocating a virtual resource according to application program allocation; A resource infrastructure unit for processing work according to the allocation of virtual resources; And a virtual resource evaluation unit for evaluating the availability of the virtual resource.

In one aspect, the resource infrastructure unit includes a plurality of virtual resources that actually process a job, the virtual resource includes a task queue module that prioritizes a task; And a job processing module that substantially processes the job.

According to another aspect of the present invention, a virtual resource evaluation unit includes: a virtual resource monitor module for checking a status of a virtual resource; And a virtual resource evaluation module for evaluating the virtual resource according to the status of the virtual resource.

In another aspect, the virtual resource evaluation module comprises: a fuzzy input module for inputting a state according to a state of a virtual resource; An inference module for evaluating virtual resources; An inverse purge module for providing inference results; A fuzzy function module including a fuzzy function for inference; And a rule base module for allocation of virtual resources for inference.

In another aspect, the inference module receives the CPU, RAM, Net I / O, Processing Time, and Queue Size of the virtual resource and evaluates the virtual resource as fuzzy logic. The second reasoning can be performed on the result of the first order reasoning, the processing time, and the queue size.

In another aspect, the fuzzy function module includes a fuzzy membership function for CPU, RAM, Net I / O, Processing Time and Queue Size of a virtual resource, and the rule base module includes a fuzzy membership function for the first- Each rule base can be included.

In another aspect, the virtual resource evaluation unit deduces the state of the virtual resource in the fuzzy logic method, calculates the processing time and the queue size for allocating the virtual resource, and prioritizes the virtual resource, and the virtual resource allocation unit You can assign tasks to high-ranking virtual resources.

A fuzzy logic-based resource evaluation method for allocating virtual resources of a cloud service, the method comprising: receiving a service request for a cloud service; Allocating an application program according to the acquired service request; Allocating a virtual resource according to an application program allocation; Processing an operation according to allocation of virtual resources; And evaluating the availability of the virtual resource based on the fuzzy logic-based resource evaluation method.

According to an embodiment of the present invention, a high utilization rate and a throughput of a cloud environment are provided by determining a best virtual resource through a fuzzy logic method based on a virtual resource status and a service type for resource evaluation of a virtual resource.

Also, it is possible to select the best virtual resource that can satisfy the performance index of the virtual resource calculated through the fuzzy logic with respect to the performance index having a weight depending on the state of the virtual resource and the type of the service, It can effectively provide cloud services.

1 is a block diagram showing a configuration including a configuration of a fuzzy logic-based resource evaluation device as an internal configuration of a cloud server in an embodiment of the present invention.
FIG. 2 is a block diagram illustrating an operation process of a resource evaluation apparatus according to an exemplary embodiment of the present invention. Referring to FIG.
3 is a block diagram for explaining a detailed configuration of a virtual resource evaluation unit according to an embodiment of the present invention.
4 is a block diagram for explaining a detailed configuration of a fuzzy logic model in an embodiment of the present invention.
5 is a diagram for explaining the configuration of a membership function in a fuzzy function module in an embodiment of the present invention.
6 is a table for explaining the detailed configuration of a rule base in a rule base module in an embodiment of the present invention.
FIG. 7 is a flowchart illustrating a fuzzy logic-based resource evaluation method according to an embodiment of the present invention.

Hereinafter, a fuzzy logic-based resource evaluation apparatus and method for allocating virtual resources of a cloud service will be described in detail with reference to the accompanying drawings.

FIG. 2 is a block diagram showing a configuration including a configuration of a fuzzy logic-based resource evaluation device as an internal configuration of a cloud server. FIG. Figure 1 illustrates a cloud server 100 for providing cloud services in a cloud computing environment. The cloud server 100 according to an embodiment performs work processing according to resource selection of a virtual resource in a system for providing an application service in a cloud environment.

In this embodiment, a virtual resource can be allocated through a method of allocating a virtual resource to which a fuzzy logic method is applied according to a service request in a resource allocation device in the cloud server 100. [

1, the cloud server 100 according to an embodiment includes an application program request unit 200, an application program allocation unit 300, a virtual resource allocation unit 400, a virtual resource evaluation unit 500 And a resource infrastructure unit 600.

First, the application service request unit 200 receives a service request from a user who is provided with a cloud service. Here, the user may correspond to a user terminal, for example, a smart phone, a tablet PC, and the like, and the service request may be received through a wired or wireless network. In addition, application program service and execution notification information according to the service request can be transmitted to the application program allocation unit 300.

The application program assigning unit 300 may determine an application program for the service request transmitted from the application program requesting unit 200, calculate the load and transmit the calculated load to the virtual resource allocating unit 400.

Then, the virtual resource allocation unit 400 selects a data center having the application program according to the application program request transmitted from the application program allocation unit 300. [ In addition, the virtual resource evaluation unit 500 can select a virtual resource in the data center according to a predetermined virtual resource evaluation. After selecting a data center in the resource infrastructure unit 600, virtual resources of the corresponding data center can be utilized.

The virtual resource evaluation unit 500 may evaluate the state of the virtual resources in the resource infrastructure unit 600 and may transmit the evaluated result to the virtual resource allocation unit 400. [

Accordingly, the resource infrastructure unit 600 can actually include virtual resources that process a job. When a virtual resource to be processed is designated from the virtual resource allocation unit 400, the designated virtual resource processes a work for the service request.

FIG. 2 is a block diagram for explaining a work process operation process of the resource evaluation apparatus according to an embodiment of the present invention. Specifically, FIG. 2 explains operations performed in the virtual resource evaluation unit 500 and the resource infrastructure unit 600 Fig.

As shown in the figure, the resource infrastructure unit 600 may include a plurality of virtual resources including the virtual resources 610. In addition, the virtual resources including the virtual resources 610 may be configured to include a work queue module 611 and a work processing module, respectively.

The configuration of the virtual resource evaluation unit 500 may include a virtual resource monitor module 510 and a virtual resource evaluation module 520.

In an embodiment, the virtual resource allocator 400 may allocate a request task by selecting a virtual resource 610 of a resource infrastructure unit 600 that is appropriate for a request task. Here, when the virtual resource 610 is allocated a request task, the virtual resource 610 stores the task sequence in the task queue module 611 to prioritize the received task.

The work queue module 611 sequentially stores the requests received from the first-in first-out method. Here, the requested jobs are sent to the job processing module 612 when the waiting time ends, and the job processing module 612 is a module that processes the requested jobs according to actual service requests. The virtual resources that can be processed at the request of the application may be different.

The virtual resource evaluation unit 500 is a module for checking and evaluating the status of the virtual resource 610 in real time. The virtual resource monitor module 510 can check the current state of the virtual resource 610. In an embodiment, the virtual resource monitor module 510 may monitor the CPU, RAM, Net I / O, processing time, and queue size of the virtual resource 610.

The virtual resource monitor module 510 senses the state of the virtual resource 610 and information on the detected state can be transmitted to the virtual resource evaluation module 520. [ The virtual resource evaluation module 520 calculates a virtual resource having a high availability by applying a fuzzy logic method according to the status of the virtual resource 610. Then, As a result of calculation, the virtual resource having the highest availability can be assigned to the corresponding job.

3 is a block diagram for explaining a detailed configuration of the virtual resource evaluation unit 500 according to an embodiment of the present invention. The virtual resource evaluation module 520 may include a fuzzy input module 521 and an inference module 522, an inverse fuzzy module 523, a fuzzy function 524, and a rule base 525 .

If the status of the virtual resource 610 is confirmed through the virtual resource monitor module 510, the virtual resource evaluation unit 520 can receive the corresponding information. As described above, the CPU of the virtual resource 610, RAM, Net I / O, Processing Time, and Queue Size information.

The fuzzy input module 521 receives a cusp function for the CPU, RAM, Net I / O, processing time, and queue size information of the virtual resource 610, and refers to the fuzzy function 524.

The inference module 522 performs calculations with reference to the fuzzy input module 521, the fuzzy function 524, and the rule base 525. The state of the virtual resource 610 can be known through the derived result.

In addition, the inverse purge module 523 deduces the state of the virtual resource 610 from the inference module 522 and performs inverse fuzzy processing so that the analysis can be performed. When the reverse purge is completed, the virtual resource allocation unit 400 provides the status information of the virtual resource 610.

The embodiment of FIGS. 4 to 6 is performed in the virtual resource evaluation module 520, and can be referred to as a diagram for explaining a virtual resource evaluation method using the expansion logic.

The fuzzy logic model S520 of FIG. 4 can be included as an operation model in the virtual resource evaluation module 520 and inputs the CPU, RAM, Net I / O, Processing Time, and Queue Size information of the virtual resource 610 And accepts the crys input according to the membership function stored in the accept fuzzy function module 524.

In the embodiment for the membership functions S5241 to S5246 in the fuzzy function module 524 and the rules S5251 to S5252 in the rule base module 525 in Fig. 5, the CPU S5241, the RAM S5242, the Net I / O (S5243), and the inference result can be derived as a virtual weight value (S5246). Here, the first rule S5251 in the rule base module 525 may be referred to for the primary inference.

When the virtual resource state is inferred through the first inference result, the fuzzy logic model (S520) makes a second inference. Secondary reasoning infer the virtual resource state, processing time (S5244), and queue size (S5245), which are the result of the first inference. Here, Processing Time can be input using the formula of Equation (1).

Here, i denotes a virtual resource 610 to be evaluated at present, and indicates a value of a virtual resource to be evaluated according to the processing time of all virtual resources.

Also, the queue size is input using the formula shown in Equation (2).

Likewise, i denotes a virtual resource 610 to be evaluated at present, and indicates a value of a virtual resource to be evaluated according to the queue size of all virtual resources.

The second rule S5252 in the rule base module 525 can be inferred for the second-order reasoning. The inference result is that the availability of virtual resources is finally deduced.

The estimated content of the estimated virtual resource availability is transmitted to the virtual resource allocation unit 400. The virtual resource to which the task is allocated in the new job request is selected based on Equation (3) as follows.

Here, the availability of each virtual resource is calculated based on the processing time, the queue size, and the inferred evaluation result (VM weight) of each virtual resource, and finally, the most available virtual resource can be selected.

FIG. 7 is a flowchart illustrating a fuzzy logic-based resource evaluation method according to an embodiment of the present invention. The method according to the embodiment can be performed through the resource evaluation apparatus in the cloud server 100 described with reference to FIG.

In step 710, a service request is received from the terminal of the user receiving the cloud service. In addition, application program service and execution notification information according to the service request can be transmitted to the application program allocation unit 300.

In step 720, an appropriate application program is determined for the service request, the load is calculated, and the calculated load is transferred to the virtual resource allocation unit 400.

In step 730, the data center having the application program is selected according to the application program request transmitted through step 720. In addition, virtual resources in the data center can be selected according to a predetermined virtual resource evaluation. After selecting a data center, virtual resources in that data center can be utilized.

Accordingly, in step 740, when a virtual resource to process a job is designated, the specified virtual resource can process a job for the service request.

In addition, in step 750, the real-time status of the virtual resources included in the apparatus may be evaluated, and the evaluated result may be transmitted to the virtual resource allocation unit 400 so as to be reflected in the priority order. At this time, the transferred evaluation result can be used to allocate a higher priority virtual resource as a virtual resource for performing a task when a later task request is displayed.

As described above, according to the embodiments of the present invention, it is possible to increase the utilization rate of virtual resources by calculating the fuzzy logic method for evaluating virtual resources in service cloud service provision, .

The fuzzy logic-based resource evaluation method according to the embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

100: Cloud server
200: application service request unit
300: application allocation unit
400: virtual resource allocation unit
500: virtual resource evaluation unit
600: Resource Infrastructure Department

Claims (8)

1. A fuzzy logic-based resource evaluation apparatus for allocating virtual resources of a cloud service,
An application service requesting unit receiving a service request from the cloud service;
An application program allocator for allocating an application program according to the obtained service request;
A virtual resource allocation unit for allocating virtual resources according to the application program allocation;
A resource infrastructure unit for processing an operation according to the allocation of the virtual resources; And
A virtual resource evaluation unit for evaluating availability of the virtual resource,
Wherein the fuzzy logic-based resource evaluation device comprises: The method according to claim 1,
Wherein the resource infrastructure unit includes a plurality of virtual resources that actually process the task,
The virtual resource
A work queue module for determining a priority for the job; And
Comprising a work processing module for substantially processing the work
Wherein the fuzzy logic-based resource evaluation device is characterized by: The method according to claim 1,
Wherein the virtual resource evaluation unit comprises:
A virtual resource monitor module for confirming the status of the virtual resource; And
And a virtual resource evaluation module for evaluating the virtual resource according to the status of the virtual resource
A resource evaluation device based on a fuzzy logic The method of claim 3,
Wherein the virtual resource evaluation module comprises:
A fuzzy input module for inputting a state according to a state of the virtual resource;
An inference module for evaluating the virtual resource;
A reverse purge module for providing the speculation result;
A fuzzy function module including a fuzzy function for the reasoning; And
And a rule base module for allocating virtual resources for the reasoning
Wherein the fuzzy logic-based resource evaluation device is characterized by: 5. The method of claim 4,
The reasoning module,
The virtual resource is evaluated by fuzzy logic based on the CPU, RAM, Net I / O, Processing Time, and Queue Size of the virtual resource,
Performing a first-order inference on the CPU, RAM, and Net I / O, and performing a second-order inference on the result of the first-order inference and the processing time and the queue size
Wherein the fuzzy logic-based resource evaluation device is characterized by: 6. The method of claim 5,
The fuzzy function module includes a fuzzy membership function for the CPU, RAM, Net I / O, Processing Time, and Queue Size of the virtual resource,
The rule base module includes each rule base for the first-order reasoning and the second order reasoning
Wherein the fuzzy logic-based resource evaluation device is characterized by: The method according to claim 1,
Wherein the virtual resource evaluation unit comprises:
Estimating a state of the virtual resource in a fuzzy logic method, computing a processing time and a queue size for allocating the virtual resource,
The virtual resource allocation unit
Assigning a task to a virtual resource having a higher priority for a new task
Wherein the fuzzy logic-based resource evaluation device is characterized by: A fuzzy logic-based resource evaluation method for virtual resource allocation of a cloud service,
Receiving a service request for the cloud service;
Assigning an application program according to the obtained service request;
Allocating a virtual resource according to the application program allocation;
Processing an operation according to the allocation of the virtual resource; And
Evaluating the availability of the virtual resource
Based on the fuzzy logic-based resource evaluation method.

Images