KR20180052927A - virtual machine placement method in a virtual machine based service function chaining - Google Patents

Abstract

Disclosed are a method and a device for arranging a virtual machine in virtual machine-based service function chaining. A method for arranging service nodes, comprises: a step of recognizing required resources of service nodes consisting of a service chain; a step of calculating link costs per link between the service nodes consisting of the service chain, and aligning links between the service nodes in accordance with the calculated link costs; a step of checking remaining resources of at least one computing node; and a step of constructing a pseudo service node by grouping the service nodes based on the remaining resources of the at least one computing node, and allocating the pseudo service node to at least one computing node among the at least one computing node.

Description

[0001] The present invention relates to a virtual machine placement method, and more particularly, to a virtual machine placement method in a service function chaining based on a virtual machine,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a service function chaining (SFC) based on a virtual machine, and more particularly to a method of efficiently arranging service nodes based on virtual machines constituting a service chain And apparatus.

The Network Functions Virtualization (NFV) technology is a new technology that can be deployed on existing hardware devices such as Network Address Translation (NAT), Firewall, Intrusion Prevention System (IPS) And Virtual Private Network (VPN), which is a virtualization technology based on software. Thus, using NFV technology can increase the flexibility of network services using a software-based framework.

In addition, there is an increasing interest in service chaining or service function chaining (SFC) technology that selectively combines and executes necessary network functions according to traffic to implement one network service.

The service function chaining (SFC) is a concept in which the traffic of the network user selectively passes through the functions necessary for the user among network services such as NAT, Firewall, IPS, etc., and these services may operate in a physical server It may also run on a virtual machine. Therefore, SFC can be understood as a concept closely related to NFV (Network Function Virtualization).

With interest in virtualized network services, a variety of public and commercial virtualization services platforms are being provided. Virtual machines are deployed in such a virtual service platform to provide service functions. At this time, the physical computer on which the virtual machine is placed is called a computing node and the virtual machine on the computing node is called placement or scheduling. Service using virtual machine is defined by means of service chaining. This service chaining expresses the connection between each virtual machine to provide service. At this time, the virtual machine that performs service chaining is called a service node.

However, communication between virtual machines, whose purpose is to provide services, has significant performance limitations. That is, a virtual machine placed in another computing node is subjected to communication load in various ways (VLAN, VxLAN, GRE, etc.). Therefore, proper placement of service nodes composed of virtual machines has a significant effect on the performance of the provided network service.

In order to solve the above problems, an object of the present invention is to provide a method and system for efficiently allocating each virtual machine service node to a computing node in an NFV-based service function chaining environment, And a method of arranging the machine service nodes.

It is another object of the present invention to solve the above problems and to provide a method and system for efficiently deploying a virtual machine service node in a computing node in an NFV-based service functioning environment, And to provide a placement apparatus for a virtual machine service node.

According to an aspect of the present invention, there is provided a method for arranging service nodes constituting a service chain in service function chaining, comprising the steps of: (a) identifying required resources of service nodes constituting the service chain; (B) calculating a link cost per link between service nodes constituting the service chain, and arranging a link between the service nodes according to the calculated link cost; Identifying (c) the remaining resources of at least one computing node; And configuring a pseudo service node by grouping the service nodes based on the requested resources of the service nodes, the cost per link, and the remaining resources of the at least one computing node, (D) of assigning one computing node to one of the computing nodes.

Here, the requesting resources of the service nodes may include at least one of a computing resource, a disk capacity, and a memory capacity required for operation of each service node.

Here, the link cost may be calculated based on at least one of the number of required transactions, the required bandwidth, and the importance between links between the service nodes. Also, the link cost may be calculated using a cost function based on at least one of the number of required transactions, the required bandwidth, and the importance between the links between the service nodes.

Here, in the allocating step, the service nodes constituting the link may be allocated to the same computing node from the link having the highest link cost among the links arranged in the order of the link cost to configure the similar service node have.

At this time, the sum of the requested resources of the service nodes constituting the similar service node is smaller than the spare resources of the same computing node.

Here, the service node may be operated based on a virtual machine.

Here, steps (c) and (d) may be repeatedly performed until no similar service node is constructed.

According to another aspect of the present invention, there is provided a service function chaining based on network function virtualization (NFV), including: a memory device in which a program code is loaded; and at least one processor for executing the program code And arranging service nodes constituting a service chain, wherein the program code comprises the steps of: identifying requested resources of service nodes constituting the service chain from an orchestrator or a predetermined specification a); (B) calculating a link cost per link between service nodes constituting the service chain, and arranging a link between the service nodes according to the calculated link cost; (C) identifying the remaining resources of at least one computing node present on a Network Function Virtualization Infrastructure (NFVI); And configuring a pseudo service node by grouping the service nodes based on the requested resources of the service nodes, the cost per link, and the remaining resources of the at least one computing node, (D) of assigning one computing node to one of the computing nodes.

Here, the requesting resources of the service nodes may include at least one of a computing resource, a disk capacity, and a memory capacity required for operation of each service node.

Here, the link cost may be calculated based on at least one of the number of required transactions, the required bandwidth, and the importance between links between the service nodes. Also, the link cost may be calculated using a cost function based on at least one of the number of required transactions, the required bandwidth, and the importance between the links between the service nodes.

Here, in the allocating step, the service nodes constituting the link may be allocated to the same computing node from the link having the highest link cost among the links arranged in the order of the link cost to configure the similar service node have.

At this time, the sum of the required resources of the service nodes constituting the similar service node may be smaller than the spare resources of the same computing node.

Here, steps (c) and (d) may be repeatedly performed until no similar service node is constructed.

Here, the device may be included in a Virtualized Infrastructure Manager (VIM).

Using the method and apparatus for arranging service nodes according to the present invention as described above, it is possible to construct similar service nodes and efficiently arrange service nodes in the computing nodes, so that when a virtual machine communicates with a virtual machine placed in another computing node The load can be reduced.

1 is a conceptual diagram illustrating a configuration of a service chain to which a service node arrangement method according to the present invention can be applied.
FIG. 2 illustrates a link cost concept of inter-service node links constituting a service chain according to an embodiment of the present invention.
FIG. 3 is a table illustrating link costs calculated for links between service nodes constituting a service chain according to an embodiment of the present invention.
4 is a table illustrating exemplary computing nodes to which service nodes configuring a service chain can be allocated according to an exemplary embodiment of the present invention.
FIG. 5 is a conceptual diagram showing a concept of a similar service node in link cost according to an embodiment of the service node arrangement method of the present invention.
FIG. 6 is a flowchart illustrating a method of arranging service nodes according to the present invention.
7 is a general conceptual diagram of a method of arranging service nodes according to the present invention.
FIG. 8 is a block diagram illustrating a concept of a service node arrangement method of the present invention performed on an NFV architecture.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing.

The terms first, second, A, B, etc. may be used to describe various elements, but the elements should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a conceptual diagram illustrating a configuration of a service chain to which a service node arrangement method according to the present invention can be applied.

FIG. 1 illustrates a service chain 100 as a general service chaining graph ().

Referring to FIG. 1, a service chain 100 may include n service nodes 101, 102, 103, and 104. At this time, each service node may refer to each unit function (for example, a server) for providing a service on a virtual machine. That is, each service node is a functional block for performing a specified process on a received packet, and can be understood as a service service function (SF: Service Function) constituting a service function chain. For example, each service node may be a functional unit that performs at least one of a network such as a firewall, an Intrusion Prevention System (IPS), a Deep Packet Inspection (DPI), and a Network Address Translation (NAT).

On the other hand, a request resource necessary for operation can be defined for each service node. For example, the requested resource may mean resources such as computing resources (e.g., CPU usage, number of cores required), disk capacity, and memory capacity required for each service node.

1, the request resource of the service node 101 is defined as r1, the request resource of the service node 102 is defined as r2, the request resource of the service node 103 is defined as r3, The request resource of node 104 may be defined as rn.

Hereinafter, the request resource is represented by one variable (r1, r2, r3, rn), but this represents a representative value which is a combination of various resources. That is, depending on the embodiment, the requested resource may be represented by two or more variables.

In addition, at least one of a number of transactions, a required bandwidth, and a weight required for each link between service nodes may be defined. For example, the number of transactions required for the link between the service node 101 and the service node 102 can be expressed as t1, the required bandwidth is b1, and the importance is w1. In a similar manner, the number of transactions for the link between the service node 102 and the service node 103 can be expressed as t2, the required bandwidth is b2, and the importance is w2. In a similar manner, for a link between a given service node and a service node 104, the number of transactions may be expressed as ti, the requested bandwidth may be bi, and the importance may be expressed as wi.

FIG. 2 illustrates a link cost concept of inter-service node links constituting a service chain according to an embodiment of the present invention.

2, the link cost between the service node 101 and the service node 102 can be expressed as cost 1, and the link cost between the service node 102 and the service node 103 can be expressed as cost 2 . In addition, the link cost between the service node 104 and a predetermined service node can be expressed as costi.

At this time, the link cost can be calculated by a cost function based on at least one of the number of transactions, the required bandwidth, and the importance required for each of the links mentioned above. For example, a value obtained by multiplying at least two values among the transaction times, the required bandwidths, and the importance values by multiplying the respective weights (or the scale values), and then adding the values, or a value obtained by applying a predefined cost function, . In various embodiments of the present invention, the cost function used to calculate the link cost is not limited to a particular function

For example, an example of a cost function for calculating a link cost used in an embodiment of the present invention can be defined as Equation 1 below.

[Equation 1]

In Equation (1), Ci represents the link cost for link i, and wi and bi denote the importance and bandwidth of link i, respectively.

On the other hand, in the case of a transaction, the deviation must be considered for each link, and the overall performance of the service may be different for each transaction to be exchanged. Therefore, ti can be transformed into nti to which the normalization is applied and applied to Equation (1). The normalization can be expressed by the following equation (2).

&Quot; (2) "

That is, the link cost cost1 between the service node 101 and the service node 102 can be calculated by a cost function based on at least one of t1, b1, and w1 described above, The link cost cost2 between the service node 103 and the service node 103 can be calculated by a cost function based on the value of at least one of t2, The costi can be calculated by a cost function based on at least one of ti, bi, and wi described above.

In this case, the link cost may be changed depending on the definition of the cost function used. However, a link with a high link cost may process more transactions than a link cost with a low link cost, require a larger bandwidth, have.

FIG. 3 is a table illustrating link costs calculated for links between service nodes constituting a service chain according to an embodiment of the present invention.

The table illustrated in FIG. 3 is a table created for service nodes and links constituting the service chain illustrated in FIG. 1 and FIG.

In the table shown in Fig. 3, the first column 301 represents the calculated link costs, the second column 302 represents the service node pair constituting each link, and the third column 303 represents The request bandwidth of each link in the fourth column 304, the importance of each link in the fifth column 305, the request resource of the service node pair constituting each link in the sixth column 306, Is described.

At this time, the links can be arranged in order of links having a high link cost. For example, in the table illustrated in FIG. 3, cost 1, cost 2, ... , and costi are shown as having high link costs in that order.

That is, the link between the service node 101 and the service node 102 has a higher link cost (i.e., cost 1> cost 2) than the link between the service node 102 and the service node 103, Nodes 103 have a higher link cost (i. E., Cost 2 > costi) than links of a given service node and service node 104.

4 is a table illustrating exemplary computing nodes to which service nodes configuring a service chain can be allocated according to an exemplary embodiment of the present invention.

First, a computing node is a physical node in which a virtual machine is disposed, which manages all resources, distributes the resource to each virtual machine, and manages a virtual machine. That is, since the service nodes on the service chain according to the present invention can be configured as instances of a server operating on a virtual machine, each service node can be created and assigned to a computing node. At this time, at least one service node may be disposed in one computing node.

In the example illustrated in FIG. 4, the first column 401 specifies the number of computing nodes that make up the computing node pool. For example, FIG. 4 illustrates that there are j computing nodes.

The second column 402 represents the total resources possessed by each computing node. Here, as with the requesting resources of the service node described above, the resources include resources such as computing resources (e.g., CPU usage, number of required cores), disk capacity, and memory capacity that each computing node can provide It can mean. For example, the first computing node # 1 has the entire resource Rt1, the second computing node # 2 has the prerequisite resource Rt2, and the computing node # Has a total resource Rtj.

The third column 403 represents resources (allocated resources) already used by each computing node. For example, the first computing node has already allocated as many resources as Ra1, the second computing node has already allocated as many resources as Ra2, and the jth computing node has already allocated as many resources as Raj.

The fourth column 404 represents the remaining resources of each computing node. For example, the first computing node holds the remaining resources as R1, the second computing node holds the remaining resources as R2, and the j computing node holds the remaining resources as Rj.

Thus, the entire resources of the second column 402, the allocated resources of the third column 403, and the remaining resources of the fourth column 404 have a relationship of remaining resources = total resource-allocated resources.

As mentioned above, all of the resources / allocated resources / remaining resources are all represented by one variable (rt1, ra1, R1, ...), which represents a representative value of various resources. That is, depending on the embodiment, the entire resource / allocated resource / remaining resource may be represented by two or more variables.

The procedure of the present invention includes calculating a link cost by using a number of transactions, a required network bandwidth, a priority, and the like for each service between nodes in a configured service chaining; Calculating resources of the two service nodes constituting the link having the highest value of the calculated link cost, and searching for a candidate computing node that can be accommodated by a remaining resource among the computing nodes; Arranging two service nodes and updating the remaining resources of the corresponding computing node; And configuring a similar service chain by reconfiguring the service chain with two disposed service nodes as one similar service node. Such a series of operations is repeated so that the steps are repeatedly performed until a service node or a similar service node is no longer able to deploy to the computing node.

An embodiment according to the present invention described above will be described below with reference to FIG. 6 and FIG.

FIG. 6 is a flowchart for explaining a service node arrangement method of the present invention, and FIG. 7 is a conceptual diagram of a service node arrangement method of the present invention.

First, in the service node arrangement method according to the present invention, the requested resources of each service node constituting the service chain are grasped (S610).

For example, the service chain 710 illustrated in FIG. 7 comprises four service nodes SN ₁ , SN ₂ , SN ₃ , and SN ₄ , and the request resources of each service node are r ₁ , r ₂ , r ₃ , r ₄ < / RTI > In addition, the number of required transactions, the required bandwidth, and the importance of the links between the respective service nodes can be defined. For example, a link between a service node SN ₁ and a service node SN ₂ may be defined as a number of request transactions t 1, a request bandwidth b 1, an importance w ₁ , a link between a service node SN ₂ and a service node SN ₃ may be defined as a request transaction number t2, required bandwidth, b2, importance w2, the link between the service node (SN ₃₎ and the service node (SN ₄₎ may be defined as a request transaction count t3, required bandwidth b3, importance w3 .

Next, the link cost of the link for each service node can be calculated (S620).

For example, in FIG. 7, the link cost between the service node SN ₁ and the service node SN ₂ may be defined as cost ₁ , and the link cost between the service node SN ₂ and the service node SN ₃ may be defined as cost 2. Also, the calculated link cost can be sorted according to its value.

Next, the remaining resources of the available computing nodes may be grasped (S630).

7 illustrates a case where five computing nodes CN ₁ , CN ₂ , CN ₃ , CN ₄ and CN ₅ form a computing node pool 740, the remaining resources can be expressed as ₁ ar, ar _{2, 3} ar, ar _4, and ₅ ar.

Therefore, the computing nodes (CN ₂ , CN ₃ , CN ₅ ) having the remaining resources larger than the requested resources of the similar service node 750 among the five computing nodes are selected as the candidate computing nodes AN ₁ , AN ₂ , AN ₃ , (730).

Next, a similar service node (e.g., 750) is configured by grouping the service nodes based on the requested resources of the service nodes, the cost per link, and the remaining resources of the at least one candidate computing node (S640) .

Service nodes _{(SN 1, SN 2, SN} 3) the sum (r1 + r2 + r3) is a similar service node is smaller than the remaining resources of the one of which will be described later, the candidate compute nodes compute nodes (750 of required resources of ). ≪ / RTI >

In this process, the link cost table that is calculated by calculating the link cost for each link illustrated in FIG. 3 and FIG. 4 and the remaining resource table of the computing node can be generated and referenced.

The service nodes SN ₁ , SN ₂ , SN ₃ forming one similar service node may be allocated to the same computing node CN ₂ (S 650). That is, the service node (SN ₁₎ of the requested resource r1, the service node (SN ₂₎ of the requested resource r2, the service node _{(SN 3) (r1 + r2} + r3) the sum of the requested resource r3 of the computing nodes (CN ₂₎ Lt; / _RTI > is less than the remaining resources (ar ₂ ) of the computing node (CN ₂ ).

In this case, when there are a plurality of candidate computing nodes (for example, AN ₁ , AN ₂ , AN ₃ ) in which the remaining resources are larger than the required resources (r 1 + r 2 + r 3) of the similar service node, load balancing balancing can be assigned to the computing node with the most remaining resources. However, depending on the predetermined policy, it is possible that the difference between the remaining resources and the requested resources (r1 + r2 + r3) of the like service node is the smallest (E.g., reduce power consumption), or assign similar service nodes according to the priority assigned to the computing nodes.

Meanwhile, the steps (S630) to (S650) are repeatedly performed until the similar service node is configured as one, or until the similar service node can not be configured because it can not be allocated to the remaining resources any more.

FIG. 8 is a block diagram illustrating a concept of a service node arrangement method of the present invention performed on an NFV architecture.

8, a typical NFV architecture includes an OSS / BSS 810, a Virtualized Network Function (VNF) 820, and a Network Function Virtualization Infrastructure (NFVI) 840 . In this case, the Operations Support System (OSS) and the Business Support System (BSS) are systems to support the provider network operation and the provision and maintenance of the customer service. The BSS (Business Support System) Customer relationship management (CRM), and call center business automation.

 There may be at least one VNF instance 821, 822 on the VNF 820. There are virtualization resources 841, 842, and 843 virtualized by the virtualization layer 844 and the virtualization layer 844 that virtualizes the actual physical hardware resources 845, 846, and 847 on the NVFI 840 .

There may also be an NFV Management and Orchestration component (MANO) 900 on the NFV architecture for managing and coordinating the components.

MANO 900 includes an orchestrator 910 for adjusting the overall operation, a virtual network function manager (VNFM) 920 for managing the VNF 820, and a virtualization infrastructure server A virtualized infrastructure manager (VIM) 930 may exist.

Here, the service node placement method of the present invention can be performed by the virtualization infrastructure administrator 930 on the NFV architecture. That is, the VIM 930 obtains information on the service chain from the orchestrator 910 or a predetermined specification 850 and the requested resources of the service nodes constituting the service chain, The remaining resources of the computing nodes are checked.

Accordingly, the VIM 930 according to the present invention comprises program code for executing the service node arrangement method of the present invention, a memory device in which the program code is stored, and at least one processor for executing the program code . In addition, the service node arrangement method of the present invention may be performed by at least the functional blocks or modules included in the VIM.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the present invention as defined by the following claims It can be understood that

710: service chain, 720: similar service chain
730: candidate computing node 740: computing node pool
750: Pseudo-Service Node
SN ₁ , SN ₂ , SN ₃ , SN ₄ : Service node
AN ₁ , AN ₂ , AN ₃ : Candidate Computing Node
CN ₁ , CN ₂ , CN ₃ , CN ₄ , CN ₅ : computing node

Claims (17)

A method for arranging service nodes constituting a service chain in service function chaining,
(A) identifying requested resources of service nodes constituting the service chain;
(B) calculating a link cost per link between service nodes constituting the service chain, and arranging a link between the service nodes according to the calculated link cost;
Identifying (c) the remaining resources of at least one computing node; And
Grouping the service nodes based on the requesting resources of the service nodes, the cost per link, and the remaining resources of the at least one computing node to construct a pseudo service node, (D) assigning to one of the computing nodes of the plurality of computing nodes,
A method of arranging service nodes. The method according to claim 1,
Wherein the requesting resources of the service nodes include at least one of a computing resource, a disk capacity, and a memory capacity required for operation of each service node.
A method of arranging service nodes. The method according to claim 1,
Wherein the link cost is calculated based on at least one of a number of requested transactions, a requested bandwidth, and an importance level between links between the service nodes,
A method of arranging service nodes. The method of claim 3,
Wherein the link cost is calculated using a cost function based on at least one of a number of required transactions, a requested bandwidth, and an importance level between links between the service nodes,
A method of arranging service nodes. The method according to claim 1,
Wherein the service nodes constituting the link are allocated to the same computing node from the link having the highest link cost among the links arranged in the order of the link cost,
A method of arranging service nodes. The method of claim 5,
The method of claim 1, wherein, in the allocating step, a sum of demand resources of service nodes constituting the similar service node is smaller than a spare resource of the same computing node,
A method of arranging service nodes. The method according to claim 1,
The service node is operated based on a virtual machine,
A method of arranging service nodes. The method according to claim 1,
Wherein steps (c) and (d) are repeatedly performed until no similar service node is configured,
A method of arranging service nodes. A service function chaining based on network function virtualization (NFV), comprising: a memory device in which a program code is loaded; and at least one processor for executing the program code, wherein the arrangement of service nodes constituting the service chain The program code comprising:
(A) identifying requested resources of service nodes constituting the service chain from an orchestrator or a predetermined specification;
(B) calculating a link cost per link between service nodes constituting the service chain, and arranging a link between the service nodes according to the calculated link cost;
(C) identifying the remaining resources of at least one computing node present on a Network Function Virtualization Infrastructure (NFVI); And
Grouping the service nodes based on the requesting resources of the service nodes, the cost per link, and the remaining resources of the at least one computing node to construct a pseudo service node, (D) assigning to one of the computing nodes of the plurality of computing nodes,
Service Node Placement Device. The method of claim 9,
Wherein the requesting resources of the service nodes include at least one of a computing resource, a disk capacity, and a memory capacity required for operation of each service node.
Service Node Placement Device. The method of claim 9,
Wherein the link cost is calculated based on at least one of a number of requested transactions, a requested bandwidth, and an importance level between links between the service nodes,
Service Node Placement Device. The method of claim 11,
Wherein the link cost is calculated using a cost function based on at least one of a number of required transactions, a requested bandwidth, and an importance level between links between the service nodes,
Service Node Placement Device. The method of claim 9,
Wherein the service nodes constituting the link are allocated to the same computing node from the link having the highest link cost among the links arranged in the order of the link cost,
Service Node Placement Device. 14. The method of claim 13,
The method of claim 1, wherein, in the allocating step, a sum of demand resources of service nodes constituting the similar service node is smaller than a spare resource of the same computing node,
Service Node Placement Device. The method of claim 9,
The service node is operated based on a virtual machine,
Service Node Placement Device. The method of claim 9,
Wherein steps (c) and (d) are repeatedly performed until no similar service node is configured,
Service Node Placement Device. The method of claim 9,
The device may be a virtualized infrastructure manager (VIM)
Service Node Placement Device.

Images