CN108897606B - Multi-tenant container cloud platform virtual network resource adaptive scheduling method and system - Google Patents
Multi-tenant container cloud platform virtual network resource adaptive scheduling method and system Download PDFInfo
- Publication number
- CN108897606B CN108897606B CN201810827859.8A CN201810827859A CN108897606B CN 108897606 B CN108897606 B CN 108897606B CN 201810827859 A CN201810827859 A CN 201810827859A CN 108897606 B CN108897606 B CN 108897606B
- Authority
- CN
- China
- Prior art keywords
- tenant
- container
- data center
- cloud
- resource
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 title claims abstract description 17
- 230000007246 mechanism Effects 0.000 claims abstract description 18
- 238000004891 communication Methods 0.000 claims abstract description 15
- 238000005457 optimization Methods 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000035945 sensitivity Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000007405 data analysis Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000010410 layer Substances 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 239000012792 core layer Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/24—Accounting or billing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/14—Charging, metering or billing arrangements for data wireline or wireless communications
- H04L12/1403—Architecture for metering, charging or billing
- H04L12/1407—Policy-and-charging control [PCC] architecture
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M15/00—Arrangements for metering, time-control or time indication ; Metering, charging or billing arrangements for voice wireline or wireless communications, e.g. VoIP
- H04M15/66—Policy and charging system
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M15/00—Arrangements for metering, time-control or time indication ; Metering, charging or billing arrangements for voice wireline or wireless communications, e.g. VoIP
- H04M15/70—Administration or customization aspects; Counter-checking correct charges
- H04M15/765—Linked or grouped accounts, e.g. of users or devices
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M15/00—Arrangements for metering, time-control or time indication ; Metering, charging or billing arrangements for voice wireline or wireless communications, e.g. VoIP
- H04M15/70—Administration or customization aspects; Counter-checking correct charges
- H04M15/785—Reserving amount on the account
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M15/00—Arrangements for metering, time-control or time indication ; Metering, charging or billing arrangements for voice wireline or wireless communications, e.g. VoIP
- H04M15/80—Rating or billing plans; Tariff determination aspects
- H04M15/8016—Rating or billing plans; Tariff determination aspects based on quality of service [QoS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
本发明属于网络技术领域，公开了一种多租户容器云平台虚拟网络资源自适应调度方法及系统，基于可用性和用户偏好的数据中心选择策略，使用容器对描述容器间的通信路径，从云服务提供商可用的备选数据中心集合出发，将每一个租户指配到最近的数据中心，确定多租户数据放置最佳的数据中心子集；进行基于应用效能的服务器选择策略和多租户容器云数据中心下的网络资源自适应调度机制。本发明提出一种多租户多数据中心的云服务环境下，容器云平台中各种网络资源之间以协同方式进行自适应调度，全局考虑云计算环境下的网络资源管理问题，在保证服务等级协议前提下，实现云服务供需双方的利益均衡。
The invention belongs to the field of network technology, and discloses a multi-tenant container cloud platform virtual network resource adaptive scheduling method and system. Based on a data center selection strategy of availability and user preference, a container pair is used to describe a communication path between containers, and a cloud service Starting from the set of alternative data centers available to the provider, assign each tenant to the nearest data center to determine the best subset of data centers for multi-tenant data placement; implement server selection strategies based on application performance and multi-tenant container cloud data The network resource adaptive scheduling mechanism under the center. The present invention proposes a multi-tenant and multi-data center cloud service environment, where various network resources in the container cloud platform perform adaptive scheduling in a coordinated manner, and considers the network resource management problem in the cloud computing environment as a whole. Under the premise of the agreement, realize the balance of interests between the supply and demand sides of cloud services.
Description
Technical Field
The invention belongs to the technical field of networks, and particularly relates to a method and a system for self-adaptive scheduling of virtual network resources of a multi-tenant container cloud platform.
Background
Currently, the current state of the art commonly used in the industry is such that:
cloud brokers are a new trend in recent years to provide services to users using multi-cloud and hybrid-cloud, and are proposed as a basic cloud service model that enables cloud network selection by renting cloud service provider instances and multiplexing relatively small tenant demands to achieve cost minimization and profit maximization.
The multi-tenant multi-data center selects common scenes with 9 tenants and 10 data centers, and spans 4 continents. Different tenants select multiple data centers according to the availability and the preference of the data centers, and the prior art only uses a mechanism of one data center, cannot meet the requirement of comprehensive data analysis, and cannot ensure faster local data analysis.
In summary, the problems of the prior art are as follows:
(1) the prior art only uses a mechanism of one data center, cannot meet the requirement of data comprehensive analysis, cannot ensure faster local data analysis, and does not have lower cost.
(2) The prior art does not consider the price difference of reserved and real-time resources and cannot reflect the dynamic characteristics of tenant demands in time.
(3) The selection is limited in a plurality of cloud platforms owned by the same cloud service provider, and the selection is difficult to be expanded among the plurality of cloud service providers.
The difficulty and significance for solving the technical problems are as follows:
the scale is huge, the failure rate is high: at present, the number of interconnected servers in a public cloud data center exceeds 105Of the order of magnitude of (1), the number of switching nodes also reaching 104The data centers with increasingly large scale put new requirements on network architecture, transmission protocols and system management. Moreover, the network failure rate increases rapidly with increasing system size, with network configuration failures (38%) and failures of unknown origin (23% such as sudden switch stops forwarding traffic).
The flow is complicated, and the longitudinal extension cost is high: due to the Incast problem caused by a high-burst and high-dynamic many-to-one communication mode, the development of computation-intensive applications such as MapReduce and Hadoop and the wide use of virtualization technologies, not only is the complexity of network traffic behaviors caused, but also a severe transmission load is brought. Meanwhile, the data center has high occupied east-west flow rate and tree structure convergence rate, so that the longitudinal expansion cost of the data center is extremely high and cannot be continued.
Low resource utilization rate, various forms: the traditional data exchange (such as VLAN) and communication identification (such as IP) technology effectively avoids mutual interference when a plurality of applications in a data center are deployed at the same time, but also limits the flexibility of network resource multiplexing, so that the utilization rate of network resources is generally low. In addition, due to the traction of different performance requirements, the situation of coexistence of networks with different forms is formed, including an enhanced ethernet network, an InfiniBand high-speed interconnection storage network, a dedicated high-speed network and the like.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method and a system for self-adaptive scheduling of virtual network resources of a multi-tenant container cloud platform.
The invention is realized in such a way that a self-adaptive scheduling method for virtual network resources of a multi-tenant container cloud platform comprises the following steps:
based on a data center selection strategy of availability and user preference, describing communication paths among containers by using container pairs [ src docker, dstDocker ], assigning each tenant to the nearest data center from an alternative data center set available to a cloud service provider, and determining a data center subset with optimal multi-tenant data placement;
server selection policy based on application performance: the container tenant specifies the efficiency parameters of the application according to the application characteristics, and provides network services of different levels by setting efficiency coefficient combinations [ alpha, beta ];
the network resource self-adaptive scheduling mechanism under the multi-tenant container cloud data center comprises the following steps: on-line self-adaptive cloud network selection algorithm based on historical information, resource reservation charging period T is estimated and divided according to historical information of each tenant on resource usage of cloud service provideriAnd resource multiplexing charging time slot taui(ii) a Let niNumber of instances enabled for the ith multiplex charging slot, CiFor the total cost of the ith multiplexing charging time slot, the optimization target of the online self-adaptive cloud network selection algorithm based on the historical information is to determine the longest resource reservation charging period and the longest resource multiplexing charging time slot tau according to the historical information of the tenanti。
Further, a container pair [ srcDocker, dstDocker ] is used]Describing the communication path between containers, the number of paths for a container pair i is denoted piThe bandwidth allocated to the container pair is represented by a vector
Is represented by the formula, wherein xijRepresents the bandwidth allocated on path j for the ith container pair; current number ofThe global bandwidth allocation vector is represented as n according to the number of container pairs in the center
The routing matrix is represented as:
further, in the server selection policy based on application performance, the container tenant specifies the performance parameters of the application according to the application characteristics, and the format is as follows [ application id, srcDocker, dst docker, Bmin,α,β]In the formula BminExpressing the minimum bandwidth requirement of the application, alpha and beta respectively expressing the throughput and the delay sensitivity coefficient of the application, and combining alpha, beta by setting the efficiency coefficient]Providing different levels of network services, the performance function is:
where u represents the set of paths used by the container pair; v represents the set of links used by the path; x is the number ofkwRepresenting the bandwidth allocated to application k on link w, 1/ywRepresenting the expected value of the congestion delay on the link w; coefficient of performance αkAnd betakRespectively representing the throughput and delay sensitive characteristics of application k.
Further, a network resource adaptive scheduling mechanism under the multi-tenant container cloud data center further includes: modeling a system: the price difference between the resource reservation and the real-time lease is utilized to reduce the cost per se.
The invention also aims to provide a computer program for realizing the self-adaptive scheduling method of the virtual network resources of the multi-tenant container cloud platform.
The invention also aims to provide an information data processing terminal for realizing the self-adaptive scheduling method of the virtual network resources of the multi-tenant container cloud platform.
Another object of the present invention is to provide a computer-readable storage medium, which includes instructions that, when executed on a computer, cause the computer to execute the method for adaptive scheduling of virtual network resources of a multi-tenant container cloud platform.
Another objective of the present invention is to provide a system for adaptively scheduling virtual network resources of a multi-tenant container cloud platform, which implements the method for adaptively scheduling virtual network resources of the multi-tenant container cloud platform, and the system includes:
a data center selection strategy unit based on availability and user preference describes communication paths among containers by using container pairs [ src docker, dstDocker ], each tenant is assigned to the nearest data center from an alternative data center set available to a cloud service provider, and a data center subset with optimal multi-tenant data placement is determined;
the server based on the application efficiency selects a policy unit: the container tenant specifies the efficiency parameters of the application according to the application characteristics, and provides network services of different levels by setting efficiency coefficient combinations [ alpha, beta ];
a network resource adaptive scheduling mechanism unit under a multi-tenant container cloud data center, an online adaptive cloud network selection algorithm based on historical information, and a resource reservation charging period T estimated and divided according to the historical information of each tenant on the use of cloud service provider resourcesiAnd resource multiplexing charging time slot taui(ii) a Let niNumber of instances enabled for the ith multiplex charging slot, CiFor the total cost of the ith multiplexing charging time slot, the optimization target of the online self-adaptive cloud network selection algorithm based on the historical information is to determine the longest resource reservation charging period and the longest resource multiplexing charging time slot tau according to the historical information of the tenanti。
The invention also aims to provide a network charging platform carrying the self-adaptive scheduling system for the virtual network resources of the multi-tenant container cloud platform.
In summary, the advantages and positive effects of the invention are:
the invention provides a cloud service environment with multiple tenants and multiple data centers, which is characterized in that various network resources in a container cloud platform are adaptively scheduled in a cooperative mode, the problem of network resource management in the cloud computing environment is considered globally, and the benefit balance of both cloud service supply and demand parties is realized on the premise of ensuring a service level agreement.
The invention aims to estimate and divide the resource reservation charging period T according to the historical information of each tenant on the resource use of the cloud service provideriAnd resource multiplexing charging time slot taui. If n is setiNumber of instances enabled for the ith multiplex charging slot, CiFor the total charge of the ith multiplexing charging time slot, the optimization goal of the algorithm is to determine the longest resource reservation charging period and resource multiplexing charging time slot tau according to the tenant historical informationiThe benefit of the cloud broker is maximized.
Compared with a mechanism using only one data center, the mechanism using a plurality of data centers can not only meet the requirement of data comprehensive analysis, but also ensure faster local data analysis and have lower cost.
Drawings
Fig. 1 is a flowchart of a self-adaptive scheduling method for virtual network resources of a multi-tenant container cloud platform according to an embodiment of the present invention.
Fig. 2 is a data center Tree topology structure diagram represented by a Fat-Tree provided in the embodiment of the present invention.
Fig. 3 is an exemplary diagram of a manner in which resources of a cloud broker network are reused according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a virtual network resource adaptive scheduling system of a multi-tenant container cloud platform provided in an embodiment of the present invention.
In the figure: 1. selecting a policy unit based on the availability and the data center of the user preference; 2. selecting a strategy unit based on the server of the application efficiency; 3. and a network resource self-adaptive scheduling mechanism unit under the multi-tenant container cloud data center.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The prior art only uses a mechanism of one data center, cannot meet the requirement of data comprehensive analysis, cannot ensure faster local data analysis, and does not have lower cost.
The self-adaptive scheduling method for the virtual network resources of the multi-tenant container cloud platform provided by the embodiment of the invention comprises the following steps:
s101: based on the data center selection strategy of availability and user preference, using a container pair to describe a communication path between containers, starting from an alternative data center set available to a cloud service provider, assigning each tenant to the nearest data center, and determining a data center subset with optimal multi-tenant data placement;
s102: server selection policy based on application performance: the container tenant specifies the efficiency parameters of the application according to the application characteristics, and provides network services of different levels by setting efficiency coefficient combinations;
s103: the network resource self-adaptive scheduling mechanism under the multi-tenant container cloud data center comprises the following steps: an online self-adaptive cloud network selection algorithm based on historical information estimates and divides a resource reservation charging period and a resource multiplexing charging time slot according to the historical information of each tenant on the resource use of a cloud service provider; and setting the number of enabled instances of the ith multiplexing charging time slot as the total cost of the ith multiplexing charging time slot, and determining the longest resource reservation charging period and the longest resource multiplexing charging time slot according to the tenant historical information by using the optimization target of the online adaptive cloud network selection algorithm based on the historical information.
In step S101, a container pair [ src docker, dstDocker ] is used to describe a communication path between containers, each tenant is assigned to a nearest data center from an alternative data center set available to a cloud service provider, and a data center subset with optimal placement of multi-tenant data is determined;
in step S102, network services of different levels are provided by setting an efficiency coefficient combination [ alpha, beta ];
step S103: resource reservation charging period Ti and resource multiplexing charging time slot tau are estimated and divided according to historical information of cloud service provider resource usage of each tenanti(ii) a Setting ni as the number of enabled instances of the ith multiplexing charging time slot and Ci as the total cost of the ith multiplexing charging time slot, determining the longest resource reservation charging period and resource multiplexing charging time slot tau according to tenant historical information by the optimization objective of the online adaptive cloud network selection algorithm based on the historical informationi。
The invention is further described below with reference to specific assays.
One, data center selection policy based on availability and user preferences:
the multi-tenant multi-data center selects common scenes with 9 tenants and 10 data centers, and spans 4 continents. Different tenants select multiple data centers according to the availability and the preference of the data centers, and compared with a mechanism using only one data center, the mechanism using multiple data centers can meet the requirement of comprehensive data analysis, can ensure faster local data analysis and has lower cost.
A data center Tree topology represented by a Fat-Tree is shown in fig. 2 as an example, and is composed of an access layer, a convergence layer, and a core layer. Wherein the number of communication paths across the container cluster is determined by the number of core layer switches and the number of communication paths within the container cluster is determined by the number of switches of the aggregation layer within the cluster.
The topology may be described in a weighted undirected graph G ═ N, L, where N represents a set of switches; l ═ 1,2.., L } (L ≧ 2) represents a set of physical links; the bandwidth capacity and the residual capacity of the link are respectively used as vectors
And
and (4) showing.
The invention contemplates the use of a container pair srcDocker, dstDocker]Describing the communication paths between containers, if the number of paths available to a container for i is denoted as piThen the bandwidth allocated to the pair of containers may be represented by the vector
Is represented by the formula, wherein xijIndicating the bandwidth allocated on path j for the ith pair of containers. Assuming that the number of container pairs in the current data center is n, the global bandwidth allocation vector may be represented as
The routing matrix may be represented as:
the invention is to store the data of each tenant only on one data center, so that the data can meet the following targets:
preferred data placement: the maximum weighted distance between the tenant and the selected data center is minimized.
Transfer cost minimization data placement: the weighted distance sum between the tenant and the selected data center is minimized.
Fair data placement: the maximum distance between the tenant and the selected data center is minimized.
The total cost is minimized: the sum of the tenant costs is minimized.
Based on the above objectives, the present invention is intended to start with an alternative set of data centers available to the cloud service provider, attempting to assign each tenant to the closest (i.e., the lowest total cost between the user and the selected data center) data center to determine the best subset of data centers for multi-tenant data placement.
Secondly, selecting a strategy based on the server of the application efficiency:
in order to provide a bandwidth allocation service with a fine granularity distinction, the invention designs a broadband allocation mode based on application efficiency for a container tenant. The container tenant may specify the performance parameters of the application according to the application characteristics, and the format is as follows [ application ID, srcDocker, dstDocker, Bmin,α,β]In the formula BminExpressing the minimum bandwidth requirement of the application, alpha and beta respectively expressing the throughput and the delay sensitivity coefficient of the application, and combining alpha, beta by setting the efficiency coefficient]Providing different levels of network services, the invention is designedThe efficacy function is as follows:
where u represents the set of paths used by the container pair; v represents the set of links used by the path; x is the number ofkwRepresenting the bandwidth allocated to application k on link w, 1/ywRepresenting the expected value of the congestion delay on the link w; coefficient of performance αkAnd betakRespectively representing the throughput and delay sensitive characteristics of application k. The performance function designed by the present invention is determined by all the container sets, paths and links used.
Thirdly, a network resource self-adaptive scheduling mechanism under the multi-tenant container cloud data center:
due to the constraint of multiple factors such as a virtualized resource pricing mode, a lease period and data transmission across cloud service providers, the service provided by a cloud broker by using a single cloud service provider instance is a basic mode for cloud network selection at present and in a future period. The cloud broker multiplexes the multi-tenant network requirements through mechanisms such as resource reservation and dynamic adjustment, and maximizes the self income on the premise of ensuring the user service agreement.
The invention is based on the following facts:
tenant requirements: according to the historical information and the demand plan of each tenant, the cloud broker can estimate the total demand of the tenant in a longer period T, and the increase of the total demand curve along with the time T is monotonically increased.
Cloud service provider pricing: the resource reservation cost strictly monotonically increases with increasing time T, while the average cost strictly monotonically decreases with increasing time T.
Modeling a system:
due to the dynamic change of the tenant demand, the cloud broker reduces its own cost by using the price difference between the resource reservation and the real-time lease, and an example of the network resource multiplexing mode commonly adopted by the cloud broker is shown in fig. 3.
An online self-adaptive cloud network selection algorithm based on historical information comprises the following steps:
according to the network resource multiplexing mode shown in fig. 3, the invention proposes to estimate and divide the resource reservation charging period T according to the historical information of each tenant on the resource usage of the cloud service provideriAnd resource multiplexing charging time slot taui. If n is setiNumber of instances enabled for the ith multiplex charging slot, CiFor the total charge of the ith multiplexing charging time slot, the optimization goal of the algorithm is to determine the longest resource reservation charging period and resource multiplexing charging time slot tau according to the tenant historical informationiThe benefit of the cloud broker is maximized.
As shown in fig. 4, an embodiment of the present invention provides a system for adaptive scheduling of virtual network resources of a multi-tenant container cloud platform, including:
the data center selection strategy unit 1 based on availability and user preference describes communication paths among containers by using container pairs [ src docker, dstDocker ], assigns each tenant to the nearest data center from an alternative data center set available to a cloud service provider, and determines a data center subset with optimal multi-tenant data placement;
the server selection policy unit 2 based on the application efficiency: the container tenant specifies the efficiency parameters of the application according to the application characteristics, and provides network services of different levels by setting efficiency coefficient combinations [ alpha, beta ];
a network resource adaptive scheduling mechanism unit 3 under a multi-tenant container cloud data center, an online adaptive cloud network selection algorithm based on historical information, and resource reservation charging period T estimated and divided according to the historical information of each tenant on the use of cloud service provider resourcesiAnd resource multiplexing charging time slot taui(ii) a Let niNumber of instances enabled for the ith multiplex charging slot, CiFor the total cost of the ith multiplexing charging time slot, the optimization target of the online self-adaptive cloud network selection algorithm based on the historical information is to determine the longest resource reservation charging period and the longest resource multiplexing charging time slot tau according to the historical information of the tenanti。
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When used in whole or in part, can be implemented in a computer program product that includes one or more computer instructions. When loaded or executed on a computer, cause the flow or functions according to embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.)). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (5)
1. A self-adaptive scheduling method for virtual network resources of a multi-tenant container cloud platform is characterized by comprising the following steps:
based on a data center selection strategy of availability and user preference, describing communication paths among containers by using container pairs [ src docker, dstDocker ], assigning each tenant to the nearest data center from an alternative data center set available to a cloud service provider, and determining a data center subset with optimal multi-tenant data placement;
using a container pair [ srcDocker, dstDocker]Describing the communication path between containers, the number of paths for a container pair i is denoted piThe bandwidth allocated to the container pair is represented by a vector
Is represented by the formula, wherein xijRepresents the bandwidth allocated on path j for the ith container pair; the current number of container pairs in the data center is n, and the global bandwidth allocation vector is represented as
The routing matrix is represented as:
server selection policy based on application performance: the container tenant specifies the efficiency parameters of the application according to the application characteristics, and provides network services of different levels by setting efficiency coefficient combinations [ alpha, beta ]; the method specifically comprises the following steps:
the container tenant specifies the efficiency parameters of the application according to the application characteristics, and the format is [ application ID, srcDocker, dstDocker, Bmin,α,β]In the formula BminExpressing the minimum bandwidth requirement of the application, alpha and beta respectively expressing the throughput and the delay sensitivity coefficient of the application, and combining alpha, beta by setting the efficiency coefficient]Providing different levels of network services, the performance function is:
where u represents the set of paths used by the container pair; v represents the set of links used by the path; x is the number ofkwRepresenting the bandwidth allocated to application k on link w, 1/ywRepresenting the expected value of the congestion delay on the link w; coefficient of performance αkAnd betakRespectively representing the throughput and delay sensitive characteristics of application k;
multi-tenant container cloudThe network resource self-adaptive scheduling mechanism under the data center comprises the following steps: on-line self-adaptive cloud network selection algorithm based on historical information, resource reservation charging period T is estimated and divided according to historical information of each tenant on resource usage of cloud service provideriAnd resource multiplexing charging time slot taui(ii) a Let niNumber of instances enabled for the ith multiplex charging slot, CiFor the total cost of the ith multiplexing charging time slot, the optimization target of the online self-adaptive cloud network selection algorithm based on the historical information is to determine the longest resource reservation charging period and the longest resource multiplexing charging time slot tau according to the historical information of the tenanti。
2. An information data processing terminal for realizing the self-adaptive scheduling method of the virtual network resources of the multi-tenant container cloud platform of claim 1.
3. A computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the multi-tenant container cloud platform virtual network resource adaptive scheduling method of claim 1.
4. A multi-tenant container cloud platform virtual network resource adaptive scheduling system for implementing the multi-tenant container cloud platform virtual network resource adaptive scheduling method of claim 1, wherein the multi-tenant container cloud platform virtual network resource adaptive scheduling system comprises:
a data center selection strategy unit based on availability and user preference describes communication paths among containers by using container pairs [ src docker, dstDocker ], each tenant is assigned to the nearest data center from an alternative data center set available to a cloud service provider, and a data center subset with optimal multi-tenant data placement is determined;
the server based on the application efficiency selects a policy unit: the container tenant specifies the efficiency parameters of the application according to the application characteristics, and provides network services of different levels by setting efficiency coefficient combinations [ alpha, beta ];
network under multi-tenant container cloud data centerA network resource adaptive scheduling mechanism unit, an online adaptive cloud network selection algorithm based on historical information, and a resource reservation charging period T estimated and divided according to the historical information of each tenant on the resource use of the cloud service provideriAnd resource multiplexing charging time slot taui(ii) a Let niNumber of instances enabled for the ith multiplex charging slot, CiFor the total cost of the ith multiplexing charging time slot, the optimization target of the online self-adaptive cloud network selection algorithm based on the historical information is to determine the longest resource reservation charging period and the longest resource multiplexing charging time slot tau according to the historical information of the tenanti。
5. A network charging platform provided with the multi-tenant container cloud platform virtual network resource adaptive scheduling system of claim 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810827859.8A CN108897606B (en) | 2018-07-25 | 2018-07-25 | Multi-tenant container cloud platform virtual network resource adaptive scheduling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810827859.8A CN108897606B (en) | 2018-07-25 | 2018-07-25 | Multi-tenant container cloud platform virtual network resource adaptive scheduling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108897606A CN108897606A (en) | 2018-11-27 |
| CN108897606B true CN108897606B (en) | 2021-06-29 |
Family
ID=64352038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810827859.8A Active CN108897606B (en) | 2018-07-25 | 2018-07-25 | Multi-tenant container cloud platform virtual network resource adaptive scheduling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108897606B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109669946B (en) * | 2018-12-14 | 2021-11-16 | 中南设计集团（武汉）工程技术研究院有限公司 | System and method for isolating complex authority system data based on massive users |
| CN111800284B (en) * | 2019-04-08 | 2023-06-20 | 阿里巴巴集团控股有限公司 | Edge cloud node set selection method and device and electronic equipment |
| CN110740168B (en) * | 2019-09-24 | 2022-06-03 | 安徽大学 | An adaptive approach to multi-tenant servers in the cloud |
| CN111162921B (en) * | 2019-11-27 | 2022-06-14 | 华为云计算技术有限公司 | Multi-account cloud service usage package sharing method, device and related equipment |
| CN112202688A (en) * | 2020-09-22 | 2021-01-08 | 临沂大学 | Data evacuation method and system suitable for cloud data center network |
| CN115701074A (en) * | 2021-07-30 | 2023-02-07 | 华为技术有限公司 | Method, device, equipment and medium for selecting cloud platform |
| CN115037956B (en) * | 2022-06-06 | 2023-03-21 | 天津大学 | Traffic scheduling method for cost optimization of edge server |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102651775A (en) * | 2012-03-05 | 2012-08-29 | 国家超级计算深圳中心(深圳云计算中心) | Method, equipment and system for managing shared objects of a plurality of lessees based on cloud computation |
| CN105554015A (en) * | 2015-12-31 | 2016-05-04 | 北京轻元科技有限公司 | Management network and method of multi-tenant container cloud computing system |
| CN105610715A (en) * | 2016-03-14 | 2016-05-25 | 山东大学 | Cloud data center multi-virtual machine migration scheduling method based on SDN (Software Defined Network) |
| CN105812222A (en) * | 2016-03-10 | 2016-07-27 | 汉柏科技有限公司 | Multi-tenant virtual network and realization method based on virtual machine and container |
| CN106131158A (en) * | 2016-06-30 | 2016-11-16 | 上海天玑科技股份有限公司 | Resource scheduling device based on cloud tenant's credit rating under a kind of cloud data center environment |
| CN106453492A (en) * | 2016-08-30 | 2017-02-22 | 浙江大学 | Docker container cloud platform container scheduling method based on fuzzy mode recognition |
| CN107222531A (en) * | 2017-05-23 | 2017-09-29 | 北京科技大学 | A kind of container cloud resource dispatching method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9819565B2 (en) * | 2015-01-26 | 2017-11-14 | Ciena Corporation | Dynamic policy engine for multi-layer network management |
-
2018
- 2018-07-25 CN CN201810827859.8A patent/CN108897606B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102651775A (en) * | 2012-03-05 | 2012-08-29 | 国家超级计算深圳中心(深圳云计算中心) | Method, equipment and system for managing shared objects of a plurality of lessees based on cloud computation |
| CN105554015A (en) * | 2015-12-31 | 2016-05-04 | 北京轻元科技有限公司 | Management network and method of multi-tenant container cloud computing system |
| CN105812222A (en) * | 2016-03-10 | 2016-07-27 | 汉柏科技有限公司 | Multi-tenant virtual network and realization method based on virtual machine and container |
| CN105610715A (en) * | 2016-03-14 | 2016-05-25 | 山东大学 | Cloud data center multi-virtual machine migration scheduling method based on SDN (Software Defined Network) |
| CN106131158A (en) * | 2016-06-30 | 2016-11-16 | 上海天玑科技股份有限公司 | Resource scheduling device based on cloud tenant's credit rating under a kind of cloud data center environment |
| CN106453492A (en) * | 2016-08-30 | 2017-02-22 | 浙江大学 | Docker container cloud platform container scheduling method based on fuzzy mode recognition |
| CN107222531A (en) * | 2017-05-23 | 2017-09-29 | 北京科技大学 | A kind of container cloud resource dispatching method |
Non-Patent Citations (1)
| Title |
|---|
| 容器云平台网络资源配置管理系统的设计;冯文超;《工业仪表与自动化装置》;20180403;第70-73页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108897606A (en) | 2018-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108897606B (en) | Multi-tenant container cloud platform virtual network resource adaptive scheduling method and system | |
| US9497139B2 (en) | Client-allocatable bandwidth pools | |
| US9154589B1 (en) | Bandwidth-optimized cloud resource placement service | |
| Popa et al. | FairCloud: Sharing the network in cloud computing | |
| US9306870B1 (en) | Emulating circuit switching in cloud networking environments | |
| Sandhu et al. | Scheduling of big data applications on distributed cloud based on QoS parameters | |
| CN109981438A (en) | A kind of satellite network load-balancing method towards SDN and NFV cooperation deployment frame | |
| Feng et al. | Jetway: Minimizing costs on inter-datacenter video traffic | |
| US10846788B1 (en) | Resource group traffic rate service | |
| CN104993941B (en) | One kind is based on Openflow network high fault tolerance virtual network mapping algorithms | |
| CN109639498B (en) | A Quality of Service-Oriented Resource Flexible Allocation Method Based on SDN and NFV | |
| KR102747387B1 (en) | Master data placement in distributed storage systems | |
| CN108566659A (en) | A kind of online mapping method of 5G networks slice based on reliability | |
| CN108965014A (en) | The service chaining backup method and system of QoS perception | |
| CN107846371B (en) | QoE resource allocation method for multimedia service | |
| CN117724853B (en) | Data processing method and device based on artificial intelligence | |
| CN110769023B (en) | A point-to-point content distribution network system based on intelligent home gateway | |
| WO2022062981A1 (en) | Resource scheduling method and system, electronic device, and computer-readable storage medium | |
| CN111835849A (en) | Method and device for enhancing service capability of access network | |
| CN111611076B (en) | Fair distribution method for mobile edge computing shared resources under task deployment constraint | |
| Ghazar et al. | Pricing utility-based virtual networks | |
| Chen et al. | A case for pricing bandwidth: Sharing datacenter networks with cost dominant fairness | |
| Han et al. | Survivable virtual network link shared protection method based on maximum spanning tree | |
| Wen et al. | Towards reliable virtual data center embedding in software defined networking | |
| Fajjari et al. | Novel adaptive virtual network embedding algorithm for Cloud’s private backbone network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |