CN105610715B - A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN - Google Patents
A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN Download PDFInfo
- Publication number
- CN105610715B CN105610715B CN201610144020.5A CN201610144020A CN105610715B CN 105610715 B CN105610715 B CN 105610715B CN 201610144020 A CN201610144020 A CN 201610144020A CN 105610715 B CN105610715 B CN 105610715B
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- migration
- indicate
- virtual
- sdn
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
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)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN, the memory of virtual machine to be migrated is considered, available link bandwidth, host CPU computing capabilitys, has belonged to the various impact factors such as intercommunication flow between the virtual machine of a VDC, as the synthesis determinant for determining virtual machine (vm) migration priority.The advantage using SDN automatic sensing network states is proposed simultaneously, is improved migration bandwidth availability ratio, is optimized Virtual Machine Manager.Present invention is primarily aimed at the overall transit times for reducing cluster virtual machine, the average downtime due to virtual data center caused by migration and each virtual machine is effectively reduced simultaneously, to promote the service quality level of cloud computing application, the performance of data center is promoted.
Description
Technical field
The present invention relates to a kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN, belongs to cloud computing
Technical field.
Background technology
With the progress of cloud computing technology, present more and more users will apply or service arrangement beyond the clouds.Cloud takes
Business provider provides a user respective service by virtualization technology.With the rise of next generation internet business, user's deployment
Using the continuous expansion of scale, as the most important subfoundation service facility cloud data center of cloud computing service, data center
The trend rapidly increased is presented in network flow, and operating pressure is increasing.Since the construction cost of data center is huge, Er Qieguan
Manage difficulty and safeguard it is complicated, and the maturation of virtual machine migration technology provided for the management and maintenance of data center it is highly effective
Technological means.The advantages of virtual machine (vm) migration, is that system maintenance management can be simplified, reduces operational overhead in data center, load
Equilibrium prevents network congestion, reduction operation service downtime etc. from playing the role of highly effective.
Although virtual machine migration technology avoids in the past when service operation environment changes, it is necessary to interrupt industry for a long time
The shortcomings that being engaged in, redeploying running environment still when large-scale virtual machine needs while migrating, still will produce delay machine
The case where overlong time, application service severe disruptions.Effective how virtual migration plan how is formulated, overall move is efficiently reduced
Shift time and downtime play good facilitation to promoting the service quality level of user.And software defined network
The appearance of SDN, providing a kind of novel network exchange model can be automatic using the pattern of control layer and data layer separation
The global bottom-layer network of perception state, realize the faster forwarding of virtual machine (vm) migration flow in cloud computation data center network.
It is main and single as influencing to migrate will generally to migrate available bandwidth for current virtual machine (vm) migration operation plan research
One impact factor is primarily present two kinds of migration schemes:1, the migration based on single virtual machine.This technology is in required migration
In the fewer scene of the number of virtual machine, often use.According to certain rule, each virtual machine is determined using Greedy strategy
The sequence of migration can only migrate a virtual machine every time, it is clear that in this way, overall transit time is longer, causes
Cluster virtual machine downtime it is also long.2, while migration plan.This migration scheme is compared to the first migration
Scheme has certain promotion, but due to the limitation of physical link available bandwidth capacity, for moving to the same purpose master
For virtual machine on machine, the available bandwidth for being assigned to each virtual machine is seldom, at the same time, because virtual machine is also continuous
Stolen goods page is generated, needs the data volume transmitted increasing, it is easy to cause the congestion of network.Therefore, a conjunction how is formulated
The multi-dummy machine migration plan of reason reduces total transit time, reduces service downtime, promotes QoS of customer level,
There is also the spaces of research.
Invention content
In view of the deficiencies of the prior art, the present invention provides a kind of, and the cloud data center multi-dummy machine migration based on SDN is adjusted
Spend method of planning;
The present invention has considered the memory of virtual machine to be migrated, available link bandwidth, host CPU computing capabilitys, same
The various impact factors such as intercommunication flow between virtual machine in one virtual data center VDC of category.Base proposed by the present invention
Framework is migrated in the New Virtual machine of SDN so that controller can automatically perceive the state change of bottom-layer network in real time, for void
The migration of quasi- machine provides global optimum path, the utilization rate of the migration link bandwidth of raising, promoted virtual data center VDC from
Dynamicization management level.The overall transit time for reducing cluster virtual machine, effectively reduces due to virtual data caused by migration
The average downtime at center and each virtual machine improves data to improve the service quality level of cloud computing application
The performance at center.
Term is explained
1,SP:The abbreviation of Service Providers refers to service provider.
2,VDCs migration request:It refer to cluster virtual machine migration request.
3,Sequencer module:Sort algorithm module is migrated, the migration priority sequence for generating virtual machine to be moved is calculated
Row.
4,VM:Virtual machine.
5,VDC:Virtual data center.
6,SDN;Software defined network.
7,SDN global orchestrator;SDN global controllers.
8,Server Manager:Server manager is responsible for the application state of implementing monitoring physical node, is carried for migration
Preceding preparation materials and equipment resources memory, CPU, link bandwidth instantiate VM.
9,OpenFlowcontroller:OpenFlow controllers issue migration by OpenFlow agreements to interchanger
Flow forwarding strategy, control migration network flow forwarding.
10,VDCs Controller:Virtual data center controller is responsible for opening and managing the migration of cluster virtual machine,
Network performance is provided by APIs interfaces for different application services to ensure.
The technical scheme is that:
A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN, this method is in the cloud number based on SDN
It is run according to center multi-dummy machine migration scheduling planning system, the cloud data center multi-dummy machine migration scheduling meter based on SDN
The system of drawing includes SDN global controllers, Sequencer modules, VDC controllers, interchanger, virtual machine, the SDN overall situations control
Device include Sequencer module, Server Manager, OpenFlow controller, Network Info databases,
VDCs Info databases;The Server Manager are mainly responsible for instantiation virtual machine or virtual switch, and in real time
Obtain and monitor the application state of physical node;The OpenFlow controller are responsible for exchanging with convergence-level and marginal layer
Machine (supporting OpenFlow agreements) direct communication issues migration flow forwarding strategy, control migration network flow forwarding;
Network Info databases mainly store the topology information of network, the state of bottom-layer network and Sequencer module and generate
Multi-dummy machine migration scheme strategy;VDCs Info databases mainly store the topology information of VDC networks, the shape of dummy node
State.Core in entire global orchestrator is Sequencer module, multi-dummy machine migration scheme strategy
Algorithm is generated to be embedded in the module.Connection relation is as shown in Figure 1, arrow indicates mutual interaction and connection, specific steps
Including:
(1) SDN global controllers receive the cluster virtual machine migration request that service provider SP is submitted;The virtual machine collection
Group migration request includeviIndicate virtual machine to be migrated,Indicate virtual machine place to be migrated
Source end host,Indicate the destination host of virtual machine to be migrated,Indicate the minimum transition band of virtual machine to be migrated
It is wide;
(2) SDN global controllers collect bottom-layer network information;
(3) the bottom-layer network information input for being collected into step (2) is to Sequencer (migration sequence) module,
Sequencer modules generate the cluster virtual machine migration scheme strategy after optimization;
(4) SDN global controllers by after the optimization of generation cluster virtual machine migration scheme and migration flow forwarding strategy
Issue VDC controllers and each bottom switch;
(5) according to the cluster virtual machine migration scheme strategy after optimization, VDC controllers are opened and manage cluster virtual machine and move
It moves;
(6) in same virtual data center VDC, the virtual link between virtual machine for intercommunication is remapped to phase
It answers on physical link;
(7) virtual machine is reactivated, virtual data center VDC restores service.
According to currently preferred, in the step (2), the bottom-layer network information include bottom physical network information,
Virtual network information, G (Ns, Ss, Ls) indicate the bottom physical network information, NsIndicate physical node, SsIndicate physical exchange
Machine, LsIndicate the set of physical link;G(Nv, Sv, Lv) indicate the virtual network information, NvIndicate virtual machine, SvIndicate virtual
Interchanger, LvIndicate the set of virtual link;
The bottom physical network information is stored in Network Info (network information) database, by the virtual net
Network information storage is in VDCs Info (virtual data center information) database.
According to currently preferred, in the step (3), Sequencer modules are preferential according to the virtual machine (vm) migration of proposition
The comprehensive certainty factor Q (v of gradei) constantly iterative calculation, find out optimal multi-dummy machine migration scheduling scheme Order;Virtual machine
Migrate priority synthesis certainty factor Q (vi) calculation formula such as formula (I) shown in:
In formula (I), CPU (vi) indicate virtual machine v to be migratediCPU sizes, Mem (vi) indicate virtual machine v to be migratedi's
Memory, bw indicate virtual machine v to be migratediMigration available link bandwidth;∑jCPU(vj) indicate and virtual machine v to be migratediMigration
The CPU summations of the identical all virtual machines of purpose physical host, ∑jMem(vj) indicate and virtual machine v to be migratediMove target object
Manage the memory summation of the identical all virtual machines of host;diIndicate that the amount of capacity of migration link bandwidth used, α indicate that CPU is big
The weight of small impact factor, β indicate that the weight of memory impact factor, γ indicate the weight of migration available link bandwidth, alpha+beta+γ
=1, β ∈ [0,1], γ ∈ [0,1].The specific value of α, β, γ can need to be adjusted according to business.
According to currently preferred, in the step (4), virtual machine that OpenFlow controllers will generate in step (3)
Cluster migration scheme policy distribution to each bottom switch, the cluster virtual machine migration scheme strategy include selection path,
The amount of bandwidth of each path distribution;Meanwhile notice VDCs (virtual data center) controller preparation of SDN global controllers starts
The migration of virtual machine.
According to currently preferred, the following operation of execution before the step (5):Server Manager are distribution in advance
Various bottom physical resources needed for virtual machine (vm) migration, the various bottom physical resources include physical host, CPU, memory, object
Manage link.
According to currently preferred, in the step (3), specific steps include:
3-1) initialization algorithm input variable:Migrate start time t ← 0, migration schemeV ← { all
Virtual machine to be migrated };
3-2) judge whether V is empty, if it is not, executing 3-3);If so, output Order, end step (3) execute
Step (4);
It 3-3) creates t moment and meets transition condition, the transition condition refers to:Whether remaining link bandwidth meets migration most
Whether bandwidth requirement low, purpose physical host residue CPU, memory size meet CPU, request memory needed for migration, can carry out
The virtual machine set V of migrationt:It traverses V, for i ∈ V, as long as i meets transition condition and can be migrated, i is added
To set VtIn;
3-4) determine the virtual machine set G that t moment can migrate simultaneously(t):
3-5) judge VtIn whether also have element, if so, initialization priority synthesis certainty factor Q ← 0, executes 3-
6);If it is not, executing 3-8);
3-6) traverse Vt, for vj∈Vt, Q (v are calculated according to formula (I)j)’Q_max←Q(vj), if Q ← Q_max, to Q
Again assignment:Q ← Q_max, the virtual machine v of migrationm←v;If it is not, Q values are constant;
3-7) step 3-6) after traversal, by 3-6) the virtual machine v to be migrated that picks out of traversing operationmWhen being added to t
Carve the virtual machine set G that can be migrated simultaneously(t)In:G(t)←G(t)∪{vm};
Update bottom-layer network status information:The bandwidth migrated before is algorithmically discharged, update residue virtual machine to be moved is in t
The feasibility of moment migration migrates bandwidth for the virtual machine arrangement that epicycle is chosen.
Update the transportable virtual machine set V of t momentt, return to step 3-5).
3-8) update migration plan Order:Order ← Order ∪ { < t, G(t)> }.
3-9) update total virtual machine (vm) migration set V:V←V/G(t)。
3-10) update migration start time t ← t+inter (w), return to step 3-2).
Inter (w) is that two adjacent migration virtual machines terminate to migrate the time difference between the moment in Order.
Beneficial effects of the present invention are:
1, the present invention has considered CPU sizes, memory, the available link bandwidth of virtual machine to be migrated, host CPU meters
Calculation ability belongs to the various impact factors such as intercommunication flow between the virtual machine of a VDC, is determined according to integrated contributory factor
The priority of virtual machine to be moved, it is ensured that meeting the virtual machine of transition condition can migrate simultaneously.
2, the present invention uses the new architecture of SDN so that controller can automatically perceive the state of bottom-layer network in real time
Variation provides global optimum path, the utilization rate of the migration link bandwidth of raising for the migration of virtual machine.
3, the present invention can reduce the overall transit time of cluster virtual machine simultaneously, while effectively reduce since migration causes
The average downtime of virtual data center and each virtual machine carried to promote the service quality level of cloud computing application
Rise the performance of data center.
4, when legacy migration strategy migrates virtual machine and determines the priority of virtual machine at the same time, only consider single influence because
Grade determines factor to son as priority.The present invention takes the combinations of a variety of impact factors, and grade determines factor as priority, and
The adjusting of various impact factor difference weights when can be according to virtual machine (vm) migration is suitable for a variety of different application scenarios, phase
For legacy migration strategy, the use scope of this method is wider, and scalability is more preferable.
Description of the drawings
Fig. 1 is that the present invention is based on the cloud data center multi-dummy machine migration scheduling planning system block diagrams of SDN.
Fig. 2 is that the present invention is based on the cloud data center multi-dummy machine migration scheduling planning procedures figures of SDN.
Fig. 3 is comprehensive priority algorithm operational flow diagram in Sequencer modules.
Specific implementation mode
The present invention is further qualified with embodiment with reference to the accompanying drawings of the specification, but not limited to this.
Embodiment
A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN, this method is in the cloud number based on SDN
It is run according to center multi-dummy machine migration scheduling planning system, the cloud data center multi-dummy machine migration scheduling meter based on SDN
The system of drawing includes SDN global controllers, Sequencer modules, VDC controllers, interchanger, virtual machine, the SDN overall situations control
Device include Sequencer module, Server Manager, OpenFlow controller, Network Info databases,
VDCs Info databases;The Server Manager are mainly responsible for instantiation virtual machine or virtual switch, and in real time
Obtain and monitor the application state of physical node;The OpenFlow controller are responsible for exchanging with convergence-level and marginal layer
Machine (supporting OpenFlow agreements) direct communication issues migration flow forwarding strategy, control migration network flow forwarding;
Network Info databases mainly store the topology information of network, the state of bottom-layer network and Sequencer module and generate
Multi-dummy machine migration scheme strategy;VDCs Info databases mainly store the topology information of VDC networks, the shape of dummy node
State.Core in entire global orchestrator is Sequencer module, multi-dummy machine migration scheme strategy
Algorithm is generated to be embedded in the module.The cloud data center multi-dummy machine migration scheduling planning system block diagram based on SDN
As shown in Figure 1, arrow indicates that the interaction and connection between disparate modules, specific steps include in Fig. 1:
(1) SDN global controllers receive the cluster virtual machine migration request that service provider SP is submitted;The virtual machine collection
Group migration request includeviIndicate virtual machine to be migrated,Indicate virtual machine place to be migrated
Source end host,Indicate the destination host of virtual machine to be migrated,Indicate the minimum transition band of virtual machine to be migrated
It is wide;
(2) SDN global controllers collect bottom-layer network information;
(3) the bottom-layer network information input for being collected into step (2) is to Sequencer (migration sequence) module,
Sequencer modules generate the cluster virtual machine migration scheme strategy after optimization;
(4) SDN global controllers by after the optimization of generation cluster virtual machine migration scheme and migration flow forwarding strategy
Issue VDC controllers and each bottom switch;
(5) according to the cluster virtual machine migration scheme strategy after optimization, VDC controllers are opened and manage cluster virtual machine and move
It moves;
(6) in same virtual data center VDC, the virtual link between virtual machine for intercommunication is remapped to phase
It answers on physical link;
(7) virtual machine is reactivated, virtual data center VDC restores service.The present invention is based on the cloud data center of SDN is more
Virtual machine (vm) migration operation plan flow chart is as shown in Figure 2.
In the step (2), in the step (2), the bottom-layer network information includes bottom physical network information, virtual
The network information, G (Ns, Ss, Ls) indicate the bottom physical network information, NsIndicate physical node, SsIndicate physical switches, Ls
Indicate the set of physical link;G(Nv, Sv, Lv) indicate the virtual network information, NvIndicate virtual machine, SvIndicate virtual switch
Machine, LvIndicate the set of virtual link;
The bottom physical network information is stored in Network Info (network information) database, by the virtual net
Network information storage is in VDCs Info (virtual data center information) database.
In the step (3), Sequencer modules integrate certainty factor Q according to the virtual machine (vm) migration priority of proposition
(vi) constantly iterative calculation, find out optimal multi-dummy machine migration scheduling scheme Order;Virtual machine (vm) migration priority synthesis is true
Determine factor Q (vi) calculation formula such as formula (I) shown in:
In formula (I), CPU (vi) indicate virtual machine v to be migratediCPU sizes, Mem (vi) indicate virtual machine v to be migratedi's
Memory, bw indicate virtual machine v to be migratediMigration available link bandwidth;∑iCPU(vj) indicate and virtual machine v to be migratediMigration
The CPU summations of the identical all virtual machines of purpose physical host, ∑jMem(vj) indicate and virtual machine v to be migratediMove target object
Manage the memory summation of the identical all virtual machines of host;diIndicate that the amount of capacity of migration link bandwidth used, α indicate that CPU is big
The weight of small impact factor, β indicate that the weight of memory impact factor, γ indicate the weight of migration available link bandwidth, alpha+beta+γ
=1, β ∈ [0,1], γ ∈ [0,1].The specific value of α, β, γ can need to be adjusted according to business.
In the step (4), OpenFlow controllers will be under the cluster virtual machine migration scheme strategy that generated in step (3)
Be dealt into each bottom switch, the cluster virtual machine migration scheme strategy include the path of selection, each path distribution bandwidth
Size;Meanwhile SDN global controllers notice VDCs (virtual data center) controller prepares to start the migration of virtual machine.
Following operation is executed before the step (5):Server Manager are needed for distribution virtual machine (vm) migration in advance
Various bottom physical resources, the various bottom physical resources include physical host, CPU, memory, physical link.
In the step (3), specific steps include:As shown in Figure 3:
3-1) initialization algorithm input variable:Migrate start time t ← 0, migration schemeY ← { all
Virtual machine to be migrated };
3-2) judge whether V is empty, if it is not, executing 3-3);If so, output Order, end step (3) execute
Step (4);
It 3-3) creates t moment and meets transition condition, the transition condition refers to:Whether remaining link bandwidth meets migration most
Whether bandwidth requirement low, purpose physical host residue CPU, memory size meet CPU, request memory needed for migration, can carry out
The virtual machine set V of migrationt:It traverses V, for i ∈ V, as long as i meets transition condition and can be migrated, i is added
To set VtIn;
3-4) determine the virtual machine set G that t moment can migrate simultaneously(t):
3-5) judge VtIn whether also have element, if so, initialization priority synthesis certainty factor Q ← 0, executes 3-
6);If it is not, executing 3-8);
3-6) traverse Vt, for vj∈Vt, Q (v are calculated according to formula (I)j)’Q_max←Q(vj), if Q < Q_max, to Q
Again assignment:Q ← Q_max, the virtual machine v of migrationm←v;If it is not, Q values are constant;
3-7) step 3-6) after traversal, by 3-6) the virtual machine v to be migrated that picks out of traversing operationmWhen being added to t
Carve the virtual machine set G that can be migrated simultaneously(t)In:G(t)←G(t)∪{vm};
Update bottom-layer network status information:The bandwidth migrated before is algorithmically discharged, update residue virtual machine to be moved is in t
The feasibility of moment migration migrates bandwidth for the virtual machine arrangement that epicycle is chosen.
Update the transportable virtual machine set V of t momentt, return to step 3-5).
3-8) update migration plan Order:Order←Order∪{<T, G(t)>}。
3-9) update total virtual machine (vm) migration set V:V←V/G(t)。
3-10) update migration start time t ← t+inter (w), return to step 3-2).
Inter (w) is that two adjacent migration virtual machines terminate to migrate the time difference between the moment in Order.
Claims (4)
1. a kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN, this method is in the cloud data based on SDN
Center multi-dummy machine migration scheduling planning system operation, the cloud data center multi-dummy machine migration scheduling plan based on SDN
System includes SDN global controllers, VDC controllers, interchanger, virtual machine, and the SDN global controllers include Sequencer
Module, Server Manager, OpenFlow controller, Network Info databases, VDCs Info databases;
It is characterized in that, specific steps include:
(1) SDN global controllers receive the cluster virtual machine migration request that service provider SP is submitted;The cluster virtual machine moves
Moving request includesviIndicate virtual machine to be migrated,Indicate the source where virtual machine to be migrated
Host,Indicate the destination host of virtual machine to be migrated,Indicate the minimum transition bandwidth of virtual machine to be migrated;
(2) SDN global controllers collect bottom-layer network information;
(3) the bottom-layer network information input for being collected into step (2) is to Sequencer module, Sequencer module lifes
At the cluster virtual machine migration scheme strategy after optimization;Including:Sequencer module are excellent according to the virtual machine (vm) migration of proposition
First grade synthesis certainty factor Q (vi) constantly iterative calculation, find out optimal multi-dummy machine migration scheduling scheme Order;Virtually
Machine migrates priority synthesis certainty factor Q (vi) calculation formula such as formula (I) shown in:
In formula (I), CPU (vi) indicate virtual machine v to be migratediCPU sizes, Mem (vi) indicate virtual machine v to be migratediMemory,
Bw indicates virtual machine v to be migratediMigration available link bandwidth;∑jCPU(vj) indicate and virtual machine v to be migratediMove target
The CPU summations of the identical all virtual machines of physical host, ∑jMem(vj) indicate and virtual machine v to be migratediMove target physics master
The memory summation of the identical all virtual machines of machine;diIndicate that the amount of capacity of migration link bandwidth used, α indicate CPU size shadows
The weight of the factor is rung, β indicates that the weight of memory impact factor, γ indicate the weight of migration available link bandwidth, alpha+beta+γ=1, β
∈[0,1],γ∈[0,1];
(4) SDN global controllers issue the cluster virtual machine migration scheme after the optimization of generation with migration flow forwarding strategy
VDC controllers and each bottom switch;
(5) according to the cluster virtual machine migration scheme strategy after optimization, VDC controllers are opened and manage cluster virtual machine migration;
(6) in same virtual data center VDC, the virtual link between virtual machine for intercommunication is remapped to homologue
Manage chain road;
(7) virtual machine is reactivated, virtual data center VDC restores service.
2. a kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN according to claim 1, special
Sign is, in the step (2), the bottom-layer network information includes bottom physical network information, virtual network information, 5 (Ns,
Ss,Ls) indicate the bottom physical network information, NsIndicate physical node, SsIndicate physical switches, LsIndicate physical link
Set;5(Nv,Sv,Lv) indicate the virtual network information, NvIndicate virtual machine, SvIndicate virtual switch, LvIndicate virtual chain
The set on road;
The bottom physical network information is stored in Network Info databases, the virtual network information storage is existed
In VDCs Info databases.
3. a kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN according to claim 1, special
Sign is, in the step (4), OpenFlow controllers will be under the cluster virtual machine migration scheme strategy that generated in step (3)
Be dealt into each bottom switch, the cluster virtual machine migration scheme strategy include the path of selection, each path distribution bandwidth
Size;Meanwhile SDN global controllers notice VDCs controllers prepare to start the migration of virtual machine.
4. a kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN according to claim 1, special
Sign is, following operation is executed before the step (5):Server Manager are each needed for distribution virtual machine (vm) migration in advance
Kind bottom physical resource, the various bottom physical resources include physical host, CPU, memory, physical link.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610144020.5A CN105610715B (en) | 2016-03-14 | 2016-03-14 | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610144020.5A CN105610715B (en) | 2016-03-14 | 2016-03-14 | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105610715A CN105610715A (en) | 2016-05-25 |
CN105610715B true CN105610715B (en) | 2018-10-23 |
Family
ID=55990257
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610144020.5A Active CN105610715B (en) | 2016-03-14 | 2016-03-14 | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105610715B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105808352A (en) * | 2016-03-07 | 2016-07-27 | 广东睿江云计算股份有限公司 | Cloud resource allocation method and device |
CN106302717B (en) * | 2016-08-12 | 2019-07-26 | 浪潮(北京)电子信息产业有限公司 | A kind of method for optimizing resources and device of CEPH system |
CN106453576B (en) * | 2016-10-21 | 2019-05-28 | 福建省华渔教育科技有限公司 | A kind of exchange method, system and control centre based on mixing cloud platform |
CN106681815A (en) * | 2016-12-27 | 2017-05-17 | 中山大学 | Concurrent migration method of virtual machines |
CN108337179B (en) * | 2017-01-19 | 2021-02-05 | 华为技术有限公司 | Link flow control method and device |
CN107301088A (en) * | 2017-07-14 | 2017-10-27 | 郑州云海信息技术有限公司 | A kind of method and apparatus for managing virtual machine batch migration |
CN108011825B (en) * | 2017-11-10 | 2020-07-28 | 深圳市泰信通信息技术有限公司 | Multi-network equipment interconnection reality method and system based on software defined network |
CN108897606B (en) * | 2018-07-25 | 2021-06-29 | 广东石油化工学院 | Self-adaptive scheduling method and system for virtual network resources of multi-tenant container cloud platform |
CN109783199A (en) * | 2019-03-28 | 2019-05-21 | 浪潮商用机器有限公司 | A kind of virtual machine migration method and device |
CN112015518B (en) * | 2020-08-27 | 2022-11-25 | 山东大学 | Method and system for realizing real-time migration of multiple virtual machines in incremental deployment SDN environment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102637138A (en) * | 2012-03-20 | 2012-08-15 | 浪潮电子信息产业股份有限公司 | Method for computing and scheduling virtual machine |
CN102968344A (en) * | 2012-11-26 | 2013-03-13 | 北京航空航天大学 | Method for migration scheduling of multiple virtual machines |
CN104683444A (en) * | 2015-01-26 | 2015-06-03 | 电子科技大学 | Data migration method for multiple virtual machines in data center |
-
2016
- 2016-03-14 CN CN201610144020.5A patent/CN105610715B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102637138A (en) * | 2012-03-20 | 2012-08-15 | 浪潮电子信息产业股份有限公司 | Method for computing and scheduling virtual machine |
CN102968344A (en) * | 2012-11-26 | 2013-03-13 | 北京航空航天大学 | Method for migration scheduling of multiple virtual machines |
CN104683444A (en) * | 2015-01-26 | 2015-06-03 | 电子科技大学 | Data migration method for multiple virtual machines in data center |
Non-Patent Citations (1)
Title |
---|
面向云数据中心的虚拟机调度机制研究;董健康;《中国博士学位论文全文数据库.信息科技辑》;20150415(第4期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN105610715A (en) | 2016-05-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105610715B (en) | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN | |
US9519500B2 (en) | Offering network performance guarantees in multi-tenant datacenters | |
WO2020258920A1 (en) | Network slice resource management method and apparatus | |
CN110297699B (en) | Scheduling method, scheduler, storage medium and system | |
CN113490254B (en) | VNF migration method based on bidirectional GRU resource demand prediction in federal learning | |
Tsai et al. | Two-tier multi-tenancy scaling and load balancing | |
CN105468435A (en) | NFV dynamic resource distribution method | |
US20140344440A1 (en) | Managing Network Utility of Applications on Cloud Data Centers | |
CN104935628B (en) | A kind of method that multiple associated virtual machines are migrated between multiple data centers | |
CN110087250B (en) | Network slice arranging scheme and method based on multi-objective joint optimization model | |
CN108667657B (en) | SDN-oriented virtual network mapping method based on local feature information | |
CN106775949A (en) | A kind of Application of composite feature that perceives migrates optimization method online with the virtual machine of the network bandwidth | |
CN102664814A (en) | Grey-prediction-based adaptive dynamic resource allocation method for virtual network | |
CN105426241A (en) | Cloud computing data center based unified resource scheduling energy-saving method | |
CN106681839B (en) | Elastic calculation dynamic allocation method | |
CN104539744B (en) | A kind of the media edge cloud dispatching method and device of two benches cooperation | |
CN104104621A (en) | Dynamic adaptive adjustment method of virtual network resources based on nonlinear dimensionality reduction | |
CN106681815A (en) | Concurrent migration method of virtual machines | |
WO2020134133A1 (en) | Resource allocation method, substation, and computer-readable storage medium | |
CN113596868A (en) | 5G network slice resource management mechanism based on SDN and NFV | |
CN105430049B (en) | A kind of virtual streaming media cluster collaboration moving method based on DCN | |
Cao et al. | Towards tenant demand-aware bandwidth allocation strategy in cloud datacenter | |
CN110138830A (en) | Across data center task schedule and bandwidth allocation methods based on hypergraph partitioning | |
CN110958192B (en) | Virtual data center resource allocation system and method based on virtual switch | |
CN105577834B (en) | Two layers of bandwidth allocation methods of cloud data center with Predicable performance and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |