CN106648834A - Virtual machine scheduling method based on batch boxing problem - Google Patents
Virtual machine scheduling method based on batch boxing problem Download PDFInfo
- Publication number
- CN106648834A CN106648834A CN201611205454.8A CN201611205454A CN106648834A CN 106648834 A CN106648834 A CN 106648834A CN 201611205454 A CN201611205454 A CN 201611205454A CN 106648834 A CN106648834 A CN 106648834A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- machine
- physical
- bin packing
- scheduling
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 230000005012 migration Effects 0.000 claims abstract description 15
- 238000013508 migration Methods 0.000 claims abstract description 15
- 238000012856 packing Methods 0.000 claims description 37
- 230000008859 change Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 238000005265 energy consumption Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000005059 dormancy Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 239000011800 void material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
-
- 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
-
- 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
- G06F2009/45595—Network integration; Enabling network access in virtual machine instances
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a virtual machine scheduling method based on a batch boxing problem. The method comprises the following steps that S1, a virtual machine scheduler regularly receives new virtual machine requests submitted by a user and collects state information of a virtual machine running on each physical machine in a system at the same time, and the state information includes information of the virtual machines about to finish running; S2, the virtual machine scheduler carries out adjustment through a batch boxing algorithm aiming at each set of new virtual machine requests, and the corresponding relationship table of the new virtual machines and the physical machines is obtained; S3, the difference between front and rear corresponding relationship tables is scheduled through a virtual machine scheduler comparison algorithm, a transfer instruction is formulated and sent to the specific physical machine, and the relevant physical machine completes virtual machine transfer according to the instruction. Compared with the prior art, according to the method, an idle server is turned off to reduce energy consumption, the virtual machine migration frequency is effectively reduced, and distribution efficiency is effectively improved.
Description
Technical field
The present invention relates to cloud computing resources dispatching technique field, more particularly to a kind of new dispatching method of virtual machine.
Background technology
Scheduling of resource is one of key issue in cloud computing, and scheduling of resource includes static scheduling and dynamic dispatching.Typically
For, static scheduling is only performed once in virtual machine selection target server host first according to existing scheduling strategy;And
Dynamic dispatching refers to situation of the scheduler according to current system load dynamic change, and by technologies such as virtual machine (vm) migrations clothes are avoided
Business device transships the appearance of phenomenon, and reaches the target of energy-conservation.
The process of virtual machine is created or migrated, is exactly the process of scheduling of resource, in system implementation, scheduling of resource
Device is according to new resource request and the system load situation adjustresources of dynamic change.Energy-efficient resource dispatching strategy,
It is the most important thing for building a cloud computing system.
In the prior art, greatly research all scheduling virtual machine problem is modeled as into online bin packing.Tool
Body, virtual machine is regarded as article, server host regards chest as, and system is according to the request of endlessly virtual machine and dynamic change
The load of change is scheduled.In the existing bin packing algorithm for solving resource scheduling, asked according to virtual machine
Arrival one by one formulate scheduling strategy.In reality, virtual machine request is often Batch Arrival, existing for each
Virtual machine request all makes the strategy of scheduling and is possible to that unnecessary virtual machine (vm) migration can be caused.
Therefore, how according to user's request it is flexible and efficient realize that the energy-saving distribution of virtual machine is the pass of current urgent need to resolve
Key technology problem.
The content of the invention
Present invention is primarily targeted at overcoming the shortcoming and deficiency of prior art, there is provided a kind of new scheduling virtual machine side
Method so that system can according to user's request it is flexible and efficient realize scheduling virtual machine.The method is based on batch bin packing
The dispatching algorithm of (Batched Bin Packing Problem).
In order to achieve the above object, the present invention is employed the following technical solutions:
Based on the dispatching method of virtual machine of batch bin packing, as shown in figure 1, comprising the steps:
S1:The new virtual machine request that the regular receive user of scheduling virtual machine device is submitted to, while every thing in collection system
The status information of the virtual machine run on reason machine, including the virtual machine information of the operation that closes to an end;
S2:The virtual machine information asked for new virtual machine and change, scheduling virtual machine device adopts bin packing algorithm
It is adjusted, draws the mapping table of new virtual machine and physical machine;
S3:Difference before and after the scheduling of scheduling virtual machine device comparison algorithm between mapping table, formulates and sends migration and refer to
Make to specified physical machine, related physical machine completes virtual machine (vm) migration according to instruction.
As preferred, in step sl, the packet that scheduling virtual machine device should be collected in specified time interval T
Include:New virtual machine request sequence that user submits to, physical machine load information and by the virtual machine for changing and its hosted service
Device information.
As preferred, scheduling virtual machine device combined with virtual machine request sequence information and load information generation in step s 2
The virtual machine and its home server information of change, calls bin packing algorithm to obtain new virtual machine and physical machine mapping table.
Used as preferred, physical machine is isomorphism server, if physical machine is heterogeneous server, can be by physical machine point
For some process, each section physical machine is all isomorphism server.
Used as preferred, the bin packing algorithm for being adopted is batch bin packing algorithm online.Compared with other online bin packing algorithms, no
It is that it is a collection of bulk arrivl that what online batch bin packing algorithm was processed is article with part, rather than reach one by one.In other words
Say, when scheduling strategy is formulated, we also know in addition in addition to knowing the virtual machine load information for operating in physical machine
The one group of virtual machine request sequence that will be allocated.
The present invention compared with prior art, has the advantage that and beneficial effect:
1. the present invention can flexibly control the quantity of physical machine according to the life cycle of virtual machine, while dormancy is in the free time
The physical machine of state, effectively reduces energy consumption;
2.2. the present invention processes in batches the virtual machine request of user, it is possible to reduce unnecessary using batch bin packing algorithm
Virtual machine (vm) migration, saves migration energy consumption;
Description of the drawings
Fig. 1 is the flow chart of the present invention;
Fig. 2 is the system logic structure figure of embodiment;
Fig. 3 is the information transmission figure of the present invention.
Specific embodiment
With reference to embodiment and accompanying drawing, the present invention is described in further detail, but embodiments of the present invention are not limited
In this.
The system logic structure of embodiment is as shown in Figure 2.Wherein, virtual machine is distributed in physical machine, using virtualization phase
Pass technology, every physical machine can simultaneously run multiple stage virtual machine, and virtual machine serves primarily in user, provide the user calculating clothes
Business;Run a virtual machine manager in every physical machine, the virtual machine manager can be collected and all run on the physical machine
On virtual machine load information, i.e., occupation condition of the every virtual machine to physical machine, and load information is informed virtually
Machine scheduler, meanwhile, virtual machine manager can receive the migration instruction that scheduling virtual machine device is sent, and coordinate scheduling virtual machine device complete
Into virtual machine (vm) migration work.
Information transmission of the present invention during scheduling of resource is as shown in figure 3, all run a void in every physical machine
Plan machine manager, the load data of virtual machine and can be sent to scheduling virtual machine device in the timed collection physical machine, and virtual machine is adjusted
Degree device receives the load information from physical machine, with reference to the request that user submits to, migration is formulated and sent according to scheduling strategy and is referred to
Make to physical machine.
Technical scheme is as follows:
Scheduling virtual machine device periodic operation.First, on the one hand, scheduling virtual machine device receives one group of new void from user
Plan machine request sequence, on the other hand, scheduling virtual machine device receives the load information of the virtual machine in every physical machine.Then,
Scheduling virtual machine device calls batch bin packing algorithm to obtain new virtual machine and physical machine mapping table, and compares virtual before and after scheduling
Machine formulates virtual machine (vm) migration plan with the difference of physical machine corresponding relation;Finally, scheduling virtual machine device is sent to command information
The physical machine of each related physical machine and dormancy in idle condition;Virtual machine manager in each physical machine receives migration and refers to
Order, coordinates scheduling virtual machine device to complete virtual machine (vm) migration.
Key technology of the present invention is further analyzed below.
(1) conversion of bin packing
Scheduling virtual machine problem is converted into bin packing by the present invention using a kind of common method:Physical machine is regarded as case
Son, size is unit 1, and virtual machine is regarded as article, and size accounts for respectively physical machine respective resources capacity equal to each resource of virtual machine
Ratio maximum.For example, it is contemplated that three kinds of (considering that situation duplicates during more kinds of resources) resources:CPU, internal memory and magnetic
Disk.Now with a virtual machine this three classes resource load account for each resource capacity of physical machine ratio be respectively 0.20,0.25,
0.15, then the corresponding article size of this virtual machine is three's maximum 0.25.Can virtual machine tune by above-mentioned conversion process
Degree problem changes into bin packing.The constraints of bin packing causes the resource capacity of every physical machine to meet to operate in
Virtual machine above it, and the target of bin packing causes physics function as much as possible to be in idle condition.Problem of the present invention
Model is a kind of special case of bin packing, criticizes bin packing, is a kind of online bin packing.
(2) input of bin packing algorithm and output
Specifically, the input of bin packing algorithm includes:In certain time T, scheduling virtual machine device receive using by oneself
One group of virtual machine request sequence Q at family, the virtual machine load data in each physical machine of physical machine is (including current virtual machine
Table B is answered with the relation pair of physical machine and will terminate the virtual machine set C for running in this section of time T);The output of bin packing algorithm
It is the mapping table of one group of virtual machine and physical machine.In the present invention, bin packing algorithm input per treatment is one group virtual
Machine set.
(3) loading strategy
According to the virtual machine set C of the operation that closes to an end, close respective virtual machine and update load information in each physical machine,
Now there may be the virtual machine in some physical machines just all to terminate to run and idle condition of entering.Loading strategy is as follows:
Article is sorted from big to small according to residual capacity, while article is sorted from big to small, two has been respectively obtained
Sorted sequence.For the largest item in article sequence, judge that the maximum chest of residual capacity whether may be used in chest sequence
To accommodate lower largest item, if it is then putting in chest and updating the article into chest capacity, otherwise, new case is opened
Son is simultaneously put article in chest into.
The present invention adopts this data structure storage bin sub-information of Max-heap, i.e., set up most according to chest remaining capacity value
Big value heap.The maxima operation that takes every time of heap only needs to constant time, and inserts node and only need to O (logn).Hence set up most
Big value heap can improve the efficiency of bin packing algorithm in the present invention.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can within the scope of the claims make various modifications or modification, this not shadow
Ring the flesh and blood of the present invention.
Claims (5)
1. based on the dispatching method of virtual machine for criticizing bin packing model, it is characterised in that comprise the steps:
S1:The new virtual machine request that the regular receive user of scheduling virtual machine device is submitted to, while every physical machine in collection system
The status information of the virtual machine of upper operation, including the virtual machine information of the operation that closes to an end;
S2:The virtual machine information asked for new virtual machine and change, scheduling virtual machine device is carried out using bin packing algorithm
Adjustment, draws the mapping table of new virtual machine and physical machine;
S3:Difference before and after the scheduling of scheduling virtual machine device comparison algorithm between mapping table, formulate and send migration instruction to
Specified physical machine, related physical machine completes virtual machine (vm) migration according to instruction.
2. it is according to claim 1 based on the dispatching method of virtual machine for criticizing bin packing model, it is characterised in that described
In step S1, new virtual machine request sequence that user submits to, physical machine load information and by the virtual machine for changing and its place
Master Server Info.
3. it is according to claim 1 based on the dispatching method of virtual machine for criticizing bin packing model, it is characterised in that described
Scheduling virtual machine device combined with virtual machine request sequence information and load information change in step S2 virtual machine and its host
Server info, calls bin packing algorithm to obtain new virtual machine and physical machine mapping table.
4. it is according to claim 1 based on the dispatching method of virtual machine for criticizing bin packing model, it is characterised in that the thing
Reason machine is isomorphism server, if physical machine is heterogeneous server, physical machine can be divided into into some and be processed, each portion
Point physical machine is isomorphism server.
5. the dispatching method of virtual machine based on batch bin packing model according to claim 1 or 3, it is characterised in that institute
The bin packing algorithm for adopting is batch bin packing algorithm online, and the article that bin packing algorithm process is criticized online is a collection of bulk arrivl, rather than
Reach one by one, that is to say, that when scheduling strategy is formulated, we bear except knowing the virtual machine operated in physical machine
Information carrying breath is outer, and the one group of virtual machine request sequence that will be allocated also is known in addition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611205454.8A CN106648834B (en) | 2016-12-23 | 2016-12-23 | Virtual machine scheduling method based on batch packaging problem |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611205454.8A CN106648834B (en) | 2016-12-23 | 2016-12-23 | Virtual machine scheduling method based on batch packaging problem |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106648834A true CN106648834A (en) | 2017-05-10 |
CN106648834B CN106648834B (en) | 2020-07-28 |
Family
ID=58827011
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611205454.8A Active CN106648834B (en) | 2016-12-23 | 2016-12-23 | Virtual machine scheduling method based on batch packaging problem |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106648834B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108073449A (en) * | 2017-11-21 | 2018-05-25 | 山东科技大学 | A kind of virtual machine dynamic laying method |
CN109947530A (en) * | 2019-01-25 | 2019-06-28 | 西安交通大学 | A kind of various dimensions virtual machine mapping method for cloud platform |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101719081A (en) * | 2009-12-01 | 2010-06-02 | 北京大学 | Method for scheduling virtual machines |
US20130086272A1 (en) * | 2011-09-29 | 2013-04-04 | Nec Laboratories America, Inc. | Network-aware coordination of virtual machine migrations in enterprise data centers and clouds |
CN103605578A (en) * | 2013-12-09 | 2014-02-26 | 武汉理工大学 | Load balance scheduling method based on virtual machine migration |
CN104298339A (en) * | 2014-10-11 | 2015-01-21 | 东北大学 | Server integration method oriented to minimum energy consumption |
US20150160672A1 (en) * | 2007-05-09 | 2015-06-11 | Gridpoint, Inc. | Method and system for scheduling the discharge of distributed power storage devices and for levelizing dispatch participation |
-
2016
- 2016-12-23 CN CN201611205454.8A patent/CN106648834B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150160672A1 (en) * | 2007-05-09 | 2015-06-11 | Gridpoint, Inc. | Method and system for scheduling the discharge of distributed power storage devices and for levelizing dispatch participation |
CN101719081A (en) * | 2009-12-01 | 2010-06-02 | 北京大学 | Method for scheduling virtual machines |
US20130086272A1 (en) * | 2011-09-29 | 2013-04-04 | Nec Laboratories America, Inc. | Network-aware coordination of virtual machine migrations in enterprise data centers and clouds |
CN103605578A (en) * | 2013-12-09 | 2014-02-26 | 武汉理工大学 | Load balance scheduling method based on virtual machine migration |
CN104298339A (en) * | 2014-10-11 | 2015-01-21 | 东北大学 | Server integration method oriented to minimum energy consumption |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108073449A (en) * | 2017-11-21 | 2018-05-25 | 山东科技大学 | A kind of virtual machine dynamic laying method |
CN108073449B (en) * | 2017-11-21 | 2021-08-27 | 山东科技大学 | Dynamic virtual machine placement method |
CN109947530A (en) * | 2019-01-25 | 2019-06-28 | 西安交通大学 | A kind of various dimensions virtual machine mapping method for cloud platform |
CN109947530B (en) * | 2019-01-25 | 2021-09-07 | 西安交通大学 | Multi-dimensional virtual machine mapping method for cloud platform |
Also Published As
Publication number | Publication date |
---|---|
CN106648834B (en) | 2020-07-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | Adaptive asynchronous federated learning in resource-constrained edge computing | |
CN104503838B (en) | A kind of virtual cpu dispatching method | |
US7689996B2 (en) | Method to distribute programs using remote Java objects | |
CN105893148B (en) | A kind of accidental task low energy consumption dispatching method based on RM strategy | |
Mehta et al. | Managing intra-operator parallelism in parallel database systems | |
CN103279390B (en) | A kind of parallel processing system (PPS) towards little optimization of job | |
CN101169731A (en) | Multiple-path multiple-core server and its CPU virtualization processing method | |
CN102364447B (en) | Operation scheduling method for optimizing communication energy consumption among multiple tasks | |
CN103916438B (en) | Cloud testing environment scheduling method and system based on load forecast | |
CN106919445A (en) | A kind of method and apparatus of the container of Parallel Scheduling in the cluster | |
CN105045658A (en) | Method for realizing dynamic dispatching distribution of task by multi-core embedded DSP (Data Structure Processor) | |
CN105320570B (en) | Method for managing resource and system | |
CN105187327A (en) | Distributed message queue middleware | |
WO2015051685A1 (en) | Task scheduling method, device and system | |
CN107861796A (en) | A kind of dispatching method of virtual machine for supporting cloud data center energy optimization | |
CN106354574A (en) | Acceleration system and method used for big data K-Mean clustering algorithm | |
CN105955809A (en) | Thread scheduling method and system | |
CN106648834A (en) | Virtual machine scheduling method based on batch boxing problem | |
CN105912383A (en) | High-reliability dependent task scheduling and resource configuration method | |
CN108429784B (en) | Energy efficiency priority cloud resource allocation and scheduling method | |
CN107918676B (en) | Resource optimization method for structured query and database query system | |
CN103645941A (en) | Novel wireless sensor network operation system | |
CN106900007B (en) | Overlapping clustering-oriented complex task allocation method for wireless sensor network | |
CN105930202B (en) | A kind of virtual machine migration method of three threshold values | |
Chen et al. | MRSIM: mitigating reducer skew In MapReduce |
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 |