CN109412865A - A kind of virtual network resource allocation method, system and electronic equipment - Google Patents
A kind of virtual network resource allocation method, system and electronic equipment Download PDFInfo
- Publication number
- CN109412865A CN109412865A CN201811430362.9A CN201811430362A CN109412865A CN 109412865 A CN109412865 A CN 109412865A CN 201811430362 A CN201811430362 A CN 201811430362A CN 109412865 A CN109412865 A CN 109412865A
- Authority
- CN
- China
- Prior art keywords
- virtual
- task image
- network
- network information
- critical path
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Information Transfer Between Computers (AREA)
Abstract
This application involves a kind of virtual network resource allocation method, system and electronic equipments.The described method includes: step a: generating task image according to the network information of virtual machines all in physical machine, and divided using multilevel scheme partitioning algorithm to the task image, obtain k mutually disjoint segmentation subsets;Step b: the critical path with the longest deadline in each segmentation subset is calculated, the virtual machine positioned at critical path node is obtained;Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.The application optimizes network in such a way that half virtualization technology is combined with single input and output virtualization technology, default each virtual machine and is assigned paravirtualized virtual network interface card, and the virtual unit of single input and output virtualization is only assigned to the virtual machine positioned at critical path node after multilevel scheme divides, to achieve the purpose that improve universe network bandwidth availability ratio and reduce Parallel application overall execution time.
Description
Technical field
The application belongs to computer virtualized technology field, in particular to a kind of virtual network resource allocation method, system
And electronic equipment.
Background technique
It is existing due to the continuous attenuating of computer hardware cost in recent years and the continuous reinforcement of single computer computation ability
It has become more and more popular and has been widely accepted in parallel computation.Meanwhile for the efficient utilization of data center resource, often using void
Quasi-ization technology is to improve the resource utilization of data center.So-called virtualization technology is exactly to refer to run in a physical machine
More virtual machine techniques, and the virtual machine does not almost have area with physical machine in user, application software even operating system
Not.Although this can increase substantially resource utilization, the problem of also thus having drawn resource allocation between different virtual machine, because
How this, effectively distribute virtual resource and raising resource utilization is constantly subjected to widely pay close attention to and study to the greatest extent.
In current parallel network and distributed network environment, promotion net is had become to the Parallel Scheduling of Internet resources
One extremely important technology of network performance.But since data-center applications layer and bottom disconnect, using de- with network at this stage
Section, this will can cause the semantic gap of network resource scheduling, and the operation of whole system is made to lack harmony, so as to cause net
The utilization rate of network resource is not high.
Summary of the invention
This application provides a kind of virtual network resource allocation method, system and electronic equipments, it is intended at least in certain journey
One of above-mentioned technical problem in the prior art is solved on degree.
To solve the above-mentioned problems, this application provides following technical solutions:
A kind of virtual network resource allocation method, comprising the following steps:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and uses multilevel scheme partitioning algorithm
The task image is divided, k mutually disjoint segmentation subsets are obtained;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, obtains being located at critical path diametral pitch
The virtual machine of point;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
The technical solution that the embodiment of the present application is taken further include: before the step a further include: collect institute in each physical machine
There is the network information of virtual machine;The collection mode of the network information are as follows: each virtual machine by read UNIX operating system be
User accesses the pseudo file system that kernel provides and obtains the network information, and the network information that the virtual machine obtains is put into storage system
In, each physical machine storage system reads the network information of all virtual machines, and the network information is uploaded to host.
The technical solution that the embodiment of the present application is taken further include: described according to void all in physical machine in the step a
The network information of quasi- machine generates task image, and is divided using multilevel scheme partitioning algorithm to task image specifically: the host
The network information that all physical machines upload is received, the task image for having parameter is generated according to the network information, and using multistage
Recurrence bipartition point method divides the task image.
The technical solution that the embodiment of the present application is taken further include: described to divide method to task image using multiple recurrence bipartition
It is divided specifically: the multiple recurrence bipartition divides method to include roughening, initial division and refine three phases, described thick
The change stage generates the smaller polymerization figure of a series of complex degree after incoming task figure, passes through the largest contiguous pair for polymerizeing figure
Vertex reduces the complexity of task image, terminates the roughening stage when the complexity of the task image drops to default vertex quantity;Institute
State the initial division stage using heuritic approach to task image carry out initial division, the elaboration phase by after initial division appoint
Business figure gradually restores.
The technical solution that the embodiment of the present application is taken further include: in the step b, the calculation formula of the critical path
Are as follows:
PathCompTime(CP)≥PathCompTime(P)
In above-mentioned formula, path refers to a certain path in task image, and PathSet refers to all roads in task image
Diameter set, PathCompTime refer to that all nodes execute the time completed in path, and symbol P refers to that free routing, CP refer to
Critical path.
Another technical solution that the embodiment of the present application is taken are as follows: a kind of virtual network resource distribution system, including physical machine
And host, the host include:
Multilevel scheme division module: it for generating task image according to the network information of virtual machines all in physical machine, and uses
Multilevel scheme partitioning algorithm divides the task image, obtains k mutually disjoint segmentation subsets;
Critical path computing module: it for calculating the critical path with the longest deadline in each segmentation subset, obtains
To the virtual machine for being located at critical path node;
Internet resources distribution module: empty for distributing single input and output for the virtual machine in critical path node
The virtual unit of quasi-ization.
The technical solution that the embodiment of the present application is taken further include: the physical machine includes Internet resources monitoring module, described
Internet resources monitoring module is used to collect the network information of all virtual machines in physical machine, and the network information of collection is uploaded to
Network resource managing module;The collection mode of the network information are as follows: each virtual machine is to use by reading UNIX operating system
Family accesses the pseudo file system that kernel provides and obtains the network information, and the network information that the virtual machine obtains is put into storage system
In, each physical machine storage system reads the network information of all virtual machines, and the network information is uploaded to Internet resources
Management module.
The technical solution that the embodiment of the present application is taken further include: the host further includes network resource managing module, described
Network resource managing module is used to receive the network information that all physical machines upload, and network information input multilevel scheme is divided mould
Block;The multilevel scheme division module generates the task image for having parameter according to the network information, and using multiple recurrence bipartition point
Method divides the task image.
The technical solution that the embodiment of the present application is taken further include: described to divide method to task image using multiple recurrence bipartition
It is divided specifically: the multiple recurrence bipartition divides method to include roughening, initial division and refine three phases, described thick
The change stage generates the smaller polymerization figure of a series of complex degree after incoming task figure, passes through the largest contiguous pair for polymerizeing figure
Vertex reduces the complexity of task image, terminates the roughening stage when the complexity of the task image drops to default vertex quantity;Institute
State the initial division stage using heuritic approach to task image carry out initial division, the elaboration phase by after initial division appoint
Business figure gradually restores.
The technical solution that the embodiment of the present application is taken further include: the calculation formula of the critical path are as follows:
PathCompTime(CP)≥PathCompTime(P)
In above-mentioned formula, path refers to a certain path in task image, and PathSet refers to all roads in task image
Diameter set, PathCompTime refer to that all nodes execute the time completed in path, and symbol P refers to that free routing, CP refer to
Critical path.
The another technical solution that the embodiment of the present application is taken are as follows: a kind of electronic equipment, comprising:
At least one processor;And
The memory being connect at least one described processor communication;Wherein,
The memory is stored with the instruction that can be executed by one processor, and described instruction is by described at least one
Device is managed to execute, so that at least one described processor is able to carry out the following operation of above-mentioned virtual network resource allocation method:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and uses multilevel scheme partitioning algorithm
The task image is divided, k mutually disjoint segmentation subsets are obtained;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, obtains being located at critical path diametral pitch
The virtual machine of point;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
Compared with the existing technology, the beneficial effect that the embodiment of the present application generates is: the virtual network of the embodiment of the present application
The side that resource allocation methods, system and electronic equipment use half virtualization technology to combine with single input and output virtualization technology
Formula optimizes network, defaults each virtual machine and is assigned paravirtualized virtual network interface card, and single input and output virtualization
Virtual unit is only assigned to the virtual machine positioned at critical path node after multilevel scheme divides, and improves universe network bandwidth to reach
Utilization rate and the purpose for reducing Parallel application overall execution time.
Detailed description of the invention
Fig. 1 is the flow chart of the virtual network resource allocation method of the embodiment of the present application;
Fig. 2 is the structural schematic diagram of the virtual network resource distribution system of the embodiment of the present application;
Fig. 3 is the hardware device structural schematic diagram of the virtual network resource allocation method of the embodiment of the present application.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the application, not
For limiting the application.
In view of the problems of the existing technology, the virtual network resource allocation method of the embodiment of the present application is virtualized using half
The side that technology and single input and output virtualization technology (SR-IOV, Single-Root I/O Virtualization) are combined
Formula optimizes network, defaults each virtual machine and is assigned paravirtualized virtual network interface card, and single input and output virtualization
Virtual unit (VF, Virtual Function) is only assigned to be located at critical path (Critical after multilevel scheme divides
Path) the virtual machine of node, to reach the mesh for improving universe network bandwidth availability ratio and reducing Parallel application overall execution time
's.
Specifically, referring to Fig. 1, being the flow chart of the virtual network resource allocation method of the embodiment of the present application.The application
The virtual network resource allocation method of embodiment the following steps are included:
Step 100: collecting the network information of all virtual machines in each physical machine, and the network information of collection is uploaded to
Host;
In step 100, the collection mode of the network information are as follows: each virtual machine is by reading UNIX operating system as user's visit
Pseudo file system/proc/net/dev that kernel provides is asked to obtain the network information, the network information that virtual machine obtains, which is put into, deposits
In storage system (XenStore), Internet resources monitoring module is respectively equipped in each physical machine, Internet resources monitoring module passes through
XenStore reads the network information of each virtual machine, and is sent to network resource managing module.
Step 200: receiving the network information of all virtual machines in each physical machine by host, generate appointing with parameter
Business figure, and division operation is carried out to task image using multilevel scheme partitioning algorithm, obtain k mutually disjoint segmentation subsets;
It include network resource managing module and multilevel scheme division module, network resource managing module on host in step 200
It is responsible for receiving the network information of all virtual machines in each physical machine, and the network information is inputted into multilevel scheme division module, generates
Task image with parameter, multilevel scheme division module divide task image using multiple recurrence bipartition point method, obtain k
A mutually disjoint segmentation subset, the corresponding task of difference segmentation subset are distributed to different physical machines and are calculated.The application
By carrying out multilevel scheme division to task image, to balance the calculation amount between each physical machine, and make between each physical machine
Total traffic capacity it is minimum.
Specifically, multiple recurrence bipartition divides method to include roughening, initial division and refinement three phases.The roughening stage exists
The smaller polymerization figure of a series of complex degree will be generated after incoming task figure, it is therefore intended that by the top for polymerizeing the largest contiguous pair of figure
Point to reduce the complexity of task image, when the complexity of task image drop to default vertex quantity (usual vertex quantity is several hundred,
Specific vertex quantity can be set according to practical application) when terminate the roughening stage.In the initial division stage, because task image
Complexity very little carries out initial division to task image using classical simple heuritic approach such as K-L algorithm.It is refining
Stage gradually restores the task image after initial division, and the vertex to condense together is moved between different segmentation subsets, is mentioned
The division effect of high task image.
Step 300: calculating the critical path with the longest deadline in each segmentation subset, obtain in each physical machine
Positioned at the virtual machine of critical path node;
In step 300, since the virtual machine communication overhead between same physical machine can be considered zero, divided in task image
Critical path afterwards calculates the communication overhead only calculated between different segmentation subsets.Specifically, the calculation formula of critical path are as follows:
PathCompTime(CP)≥PathCompTime(P) (1)
In formula (1), path refers to a certain path in task image, and PathSet refers to all roads in task image
Diameter set, PathCompTime refer to that all nodes execute the time completed in path, and symbol P refers to that free routing, CP refer to
Critical path.
Step 400: distributing single input and output virtualization to be in the virtual machine of critical path node in each physical machine
Virtual unit;
In step 400, the network resource allocation method of the embodiment of the present application is that default allows each virtual machine to be assigned half
The Microsoft Loopback Adapter of virtualization, and the virtual unit of single input and output virtualization is only assigned to be located at key after multilevel scheme divides
The virtual machine of path node.It is wherein transparent between paravirtualized virtual network interface card and the virtual unit of single input and output virtualization
Switching has used more network card binding technologies, and the virtual unit of single input and output virtualization is arranged to major network card, and half virtualizes
Microsoft Loopback Adapter be arranged to from network interface card, virtual machine is communicated using major network card when major network card is effective, only in master
Just use is communicated from network interface card when network interface card is invalid, to reduce network overhead.
Step 500: determining whether all segmentation subsets are finished, if all segmentation subsets are finished, execute step
Rapid 600;Otherwise, the critical path that return step 300 continues to calculate next segmentation subset calculates, until all segmentation subsets execute
It finishes;
Step 600: virtual network resource distribution terminates.
Referring to Fig. 2, being the structure chart of the virtual network resource distribution system of the embodiment of the present application.The embodiment of the present application
Virtual network resource distribution system includes physical machine and host, and physical machine includes Internet resources monitoring module, and host includes network
Resource management module, multilevel scheme division module, critical path computing module, Internet resources distribution module and task execution judge mould
Block.
Internet resources monitoring module: for collecting the network information of all virtual machines in each physical machine, and by collection
The network information is uploaded to network resource managing module;Wherein, the collection mode of the network information are as follows: each virtual machine passes through reading
UNIX operating system is that user accesses pseudo file system/proc/net/dev that kernel provides to obtain the network information, virtual machine
The network information of acquisition is put into XenStore, and Internet resources monitoring module reads the network of each virtual machine by XenStore
Information, and it is sent to network resource managing module.
Network resource managing module: believe for receiving the network information of all virtual machines in each physical machine, and by network
Breath input multilevel scheme division module;
Multilevel scheme division module: for generating the task image for having parameter according to the network information of input, and using multistage
Figure partitioning algorithm carries out division operation to task image, obtains k mutually disjoint segmentation subsets;Wherein, multilevel scheme division module
Task image is divided using multiple recurrence bipartition point method, obtains k mutually disjoint segmentation subsets, difference segmentation
Collect corresponding task and distribute to different physical machines and is calculated.The application by task image carry out multilevel scheme division, thus
The calculation amount between each physical machine is balanced, and keeps the total traffic capacity between each physical machine minimum.
Specifically, multiple recurrence bipartition divides method to include roughening, initial division and refinement three phases.The roughening stage exists
The smaller polymerization figure of a series of complex degree will be generated after incoming task figure, it is therefore intended that by the top for polymerizeing the largest contiguous pair of figure
It puts to reduce the complexity of task image, terminates the roughening stage when the complexity of task image drops to only several hundred a vertex.First
Division stage beginning, because of the complexity of task image very little, using classical simple heuritic approach such as K-L algorithm to appointing
Business figure carries out initial division.In elaboration phase, the task image after initial division is gradually restored, the vertex to condense together is existed
It is moved between different segmentation subsets, improves the division effect of task image.
Critical path computing module: it for calculating the critical path with the longest deadline in each segmentation subset, obtains
It is located at the virtual machine of critical path node on to each physical machine;Wherein, since the virtual machine communication between same physical machine is opened
Pin can be considered zero, therefore calculate the communication overhead between only calculating different segmentation subsets in the critical path after task image division.Tool
Body, the calculation formula of critical path are as follows:
PathCompTime(CP)≥PathCompTime(P) (1)
In formula (1), path refers to a certain path in task image, and PathSet refers to all roads in task image
Diameter set, PathCompTime refer to that all nodes execute the time completed in path, and symbol P refers to that free routing, CP refer to
Critical path.
Internet resources distribution module: it is distributed for the virtual machine to be in critical path node in each physical machine single defeated
Enter the virtual unit of output virtualization;Wherein, the network resource allocation method of the embodiment of the present application is that default allows each virtual machine
It is all assigned paravirtualized virtual network interface card, and after the virtual unit of single input and output virtualization is only assigned to multilevel scheme division
The virtual machine positioned at critical path node.Wherein paravirtualized virtual network interface card is virtually set with what single input and output virtualized
Transparent switching has used more network card binding technologies between standby, and the virtual unit of single input and output virtualization is arranged to major network
Card, paravirtualized virtual network interface card are arranged to from network interface card, and virtual machine is led to using major network card when major network card is effective
Letter, only just use is communicated from network interface card when major network card is invalid, to reduce network overhead.
Task execution judgment module: for determining whether all segmentation subsets are finished, if all segmentation subsets are held
Row finishes, then virtual network resource distribution terminates;Otherwise, continue to calculate next segmentation subset by critical path computing module
Critical path calculates, until all segmentation subsets are finished.
Fig. 3 is the hardware device structural schematic diagram of virtual network resource allocation method provided by the embodiments of the present application.Such as Fig. 3
Shown, which includes one or more processors and memory.It takes a processor as an example, which can also include:
Input system and output system.
Processor, memory, input system and output system can be connected by bus or other modes, in Fig. 3 with
For being connected by bus.
Memory as a kind of non-transient computer readable storage medium, can be used for storing non-transient software program, it is non-temporarily
State computer executable program and module.Processor passes through operation non-transient software program stored in memory, instruction
And module realizes the place of above method embodiment thereby executing the various function application and data processing of electronic equipment
Reason method.
Memory may include storing program area and storage data area, wherein storing program area can storage program area, extremely
Application program required for a few function;It storage data area can storing data etc..In addition, memory may include that high speed is random
Memory is accessed, can also include non-transient memory, a for example, at least disk memory, flush memory device or other are non-
Transient state solid-state memory.In some embodiments, it includes the memory remotely located relative to processor that memory is optional, this
A little remote memories can pass through network connection to processing system.The example of above-mentioned network includes but is not limited to internet, enterprise
Intranet, local area network, mobile radio communication and combinations thereof.
Input system can receive the number or character information of input, and generate signal input.Output system may include showing
Display screen etc. shows equipment.
One or more of module storages in the memory, are executed when by one or more of processors
When, execute the following operation of any of the above-described embodiment of the method:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and uses multilevel scheme partitioning algorithm
The task image is divided, k mutually disjoint segmentation subsets are obtained;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, obtains being located at critical path diametral pitch
The virtual machine of point;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
Method provided by the embodiment of the present application can be performed in the said goods, has the corresponding functional module of execution method and has
Beneficial effect.The not technical detail of detailed description in the present embodiment, reference can be made to method provided by the embodiments of the present application.
The embodiment of the present application provides a kind of non-transient (non-volatile) computer storage medium, and the computer storage is situated between
Matter is stored with computer executable instructions, the executable following operation of the computer executable instructions:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and uses multilevel scheme partitioning algorithm
The task image is divided, k mutually disjoint segmentation subsets are obtained;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, obtains being located at critical path diametral pitch
The virtual machine of point;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
The embodiment of the present application provides a kind of computer program product, and the computer program product is non-temporary including being stored in
Computer program on state computer readable storage medium, the computer program include program instruction, when described program instructs
When being computer-executed, the computer is made to execute following operation:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and uses multilevel scheme partitioning algorithm
The task image is divided, k mutually disjoint segmentation subsets are obtained;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, obtains being located at critical path diametral pitch
The virtual machine of point;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
The virtual network resource allocation method, system and electronic equipment of the embodiment of the present application use half virtualization technology and list
The mode that root input and output virtualization technology combines optimizes network, defaults each virtual machine and is assigned paravirtualized void
Quasi- network interface card, and the virtual unit of single input and output virtualization is only assigned to be located at critical path node after multilevel scheme divides
Virtual machine, to achieve the purpose that improve universe network bandwidth availability ratio and reduce Parallel application overall execution time.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, defined herein
General Principle can realize in other embodiments without departing from the spirit or scope of the application.Therefore, this Shen
These embodiments shown in the application please be not intended to be limited to, and are to fit to special with principle disclosed in the present application and novelty
The consistent widest scope of point.
Claims (11)
1. a kind of virtual network resource allocation method, which comprises the following steps:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and using multilevel scheme partitioning algorithm to institute
It states task image to be divided, obtains k mutually disjoint segmentation subsets;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, is obtained positioned at critical path node
Virtual machine;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
2. virtual network resource allocation method according to claim 1, which is characterized in that before the step a further include: receive
Collect the network information of all virtual machines in each physical machine;The collection mode of the network information are as follows: each virtual machine passes through reading
Taking UNIX operating system is the pseudo file system acquisition network information that user accesses that kernel provides, the network that the virtual machine obtains
Information is put into storage system, and each physical machine storage system reads the network information of all virtual machines, and the network is believed
Breath is uploaded to host.
3. virtual network resource allocation method according to claim 2, which is characterized in that in the step a, described
Task image is generated according to the network information of virtual machines all in physical machine, and task image is divided using multilevel scheme partitioning algorithm
Specifically: the host receives the network information that all physical machines upload, and generates appointing with parameter according to the network information
Business figure, and the task image is divided using multiple recurrence bipartition point method.
4. virtual network resource allocation method according to claim 3, which is characterized in that described to be divided using multiple recurrence two
Division methods divide task image specifically: the multiple recurrence bipartition divides method to include roughening, initial division and thin
Change three phases, the roughening stage generates the smaller polymerization figure of a series of complex degree after incoming task figure, by described poly-
The vertex for closing the largest contiguous pair of figure reduces the complexity of task image, when the complexity of the task image drops to default vertex quantity
When terminate the roughening stage;The initial division stage carries out initial division, the refinement rank to task image using heuritic approach
Section gradually restores the task image after initial division.
5. virtual network resource allocation method according to any one of claims 1 to 4, which is characterized in that in the step b
In, the calculation formula of the critical path are as follows:
PathCompTime(CP)≥PathCompTime(P)
In above-mentioned formula, path refers to a certain path in task image, and PathSet refers to all path sets in task image
It closes, PathCompTime refers to that all nodes execute the time completed in path, and symbol P refers to that free routing, CP refer to key
Path.
6. a kind of virtual network resource distribution system, including physical machine and host, which is characterized in that the host includes:
Multilevel scheme division module: for generating task image according to the network information of virtual machines all in physical machine, and using multistage
Figure partitioning algorithm divides the task image, obtains k mutually disjoint segmentation subsets;
Critical path computing module: it for calculating the critical path with the longest deadline in each segmentation subset, obtains in place
In the virtual machine of critical path node;
Internet resources distribution module: for distributing single input and output virtualization for the virtual machine in critical path node
Virtual unit.
7. virtual network resource distribution system according to claim 6, which is characterized in that the physical machine includes network money
Source monitoring module, the Internet resources monitoring module is used to collect the network information of all virtual machines in physical machine, and will collect
The network information be uploaded to network resource managing module;The collection mode of the network information are as follows: each virtual machine passes through reading
UNIX operating system is the pseudo file system acquisition network information that user accesses that kernel provides, the network letter that the virtual machine obtains
Breath is put into storage system, and each physical machine storage system reads the network information of all virtual machines, and by the network information
It is uploaded to network resource managing module.
8. virtual network resource distribution system according to claim 7, which is characterized in that the host further includes network money
Source control module, the network resource managing module are used to receive the network information that all physical machines upload, and by the network information
Input multilevel scheme division module;The multilevel scheme division module generates the task image for having parameter according to the network information, and uses
Multiple recurrence bipartition point method divides the task image.
9. virtual network resource distribution system according to claim 8, which is characterized in that described to be divided using multiple recurrence two
Division methods divide task image specifically: the multiple recurrence bipartition divides method to include roughening, initial division and thin
Change three phases, the roughening stage generates the smaller polymerization figure of a series of complex degree after incoming task figure, by described poly-
The vertex for closing the largest contiguous pair of figure reduces the complexity of task image, when the complexity of the task image drops to default vertex quantity
When terminate the roughening stage;The initial division stage carries out initial division, the refinement rank to task image using heuritic approach
Section gradually restores the task image after initial division.
10. according to the described in any item virtual network resource distribution systems of claim 6 to 9, which is characterized in that the critical path
The calculation formula of diameter are as follows:
PathCompTime(CP)≥PathCompTime(P)
In above-mentioned formula, path refers to a certain path in task image, and PathSet refers to all path sets in task image
It closes, PathCompTime refers to that all nodes execute the time completed in path, and symbol P refers to that free routing, CP refer to key
Path.
11. a kind of electronic equipment, comprising:
At least one processor;And
The memory being connect at least one described processor communication;Wherein,
The memory is stored with the instruction that can be executed by one processor, and described instruction is by least one described processor
It executes, so that at least one described processor is able to carry out above-mentioned 1 to 5 described in any item virtual network resource allocation methods
It operates below:
Step a: task image is generated according to the network information of virtual machines all in physical machine, and using multilevel scheme partitioning algorithm to institute
It states task image to be divided, obtains k mutually disjoint segmentation subsets;
Step b: the critical path with the longest deadline in each segmentation subset is calculated, is obtained positioned at critical path node
Virtual machine;
Step c: the virtual unit of single input and output virtualization is distributed for the virtual machine in critical path node.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811430362.9A CN109412865B (en) | 2018-11-28 | 2018-11-28 | Virtual network resource allocation method, system and electronic equipment |
PCT/CN2019/121297 WO2020108536A1 (en) | 2018-11-28 | 2019-11-27 | Virtual network resource allocation method and system and electronic device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811430362.9A CN109412865B (en) | 2018-11-28 | 2018-11-28 | Virtual network resource allocation method, system and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109412865A true CN109412865A (en) | 2019-03-01 |
CN109412865B CN109412865B (en) | 2020-08-25 |
Family
ID=65455905
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811430362.9A Active CN109412865B (en) | 2018-11-28 | 2018-11-28 | Virtual network resource allocation method, system and electronic equipment |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109412865B (en) |
WO (1) | WO2020108536A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020108536A1 (en) * | 2018-11-28 | 2020-06-04 | 深圳先进技术研究院 | Virtual network resource allocation method and system and electronic device |
CN111427912A (en) * | 2020-03-31 | 2020-07-17 | 拉卡拉支付股份有限公司 | Task processing method and device, electronic equipment and storage medium |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112330229B (en) * | 2020-12-02 | 2023-09-22 | 北京元心科技有限公司 | Resource scheduling method, device, electronic equipment and computer readable storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015028931A1 (en) * | 2013-08-29 | 2015-03-05 | Ericsson Ab | A method and system to allocate bandwidth based on task deadline in cloud computing networks |
US20150089495A1 (en) * | 2013-09-25 | 2015-03-26 | Arm Limited | Data processing systems |
CN104536806A (en) * | 2014-12-26 | 2015-04-22 | 东南大学 | Workflow application flexible resource supplying method in cloud environment |
CN105357322A (en) * | 2015-12-11 | 2016-02-24 | 中国科学院信息工程研究所 | Virtual machine distribution method based on topology partition |
CN108205461A (en) * | 2016-12-19 | 2018-06-26 | 华耀(中国)科技有限公司 | The virtual platform and dispositions method of a kind of mixed deployment |
CN110113184A (en) * | 2019-04-17 | 2019-08-09 | 中国科学院深圳先进技术研究院 | KVM virtual machine network optimization method and device under SR-IOV environment |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109412865B (en) * | 2018-11-28 | 2020-08-25 | 深圳先进技术研究院 | Virtual network resource allocation method, system and electronic equipment |
-
2018
- 2018-11-28 CN CN201811430362.9A patent/CN109412865B/en active Active
-
2019
- 2019-11-27 WO PCT/CN2019/121297 patent/WO2020108536A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015028931A1 (en) * | 2013-08-29 | 2015-03-05 | Ericsson Ab | A method and system to allocate bandwidth based on task deadline in cloud computing networks |
US20150089495A1 (en) * | 2013-09-25 | 2015-03-26 | Arm Limited | Data processing systems |
CN104536806A (en) * | 2014-12-26 | 2015-04-22 | 东南大学 | Workflow application flexible resource supplying method in cloud environment |
CN105357322A (en) * | 2015-12-11 | 2016-02-24 | 中国科学院信息工程研究所 | Virtual machine distribution method based on topology partition |
CN108205461A (en) * | 2016-12-19 | 2018-06-26 | 华耀(中国)科技有限公司 | The virtual platform and dispositions method of a kind of mixed deployment |
CN110113184A (en) * | 2019-04-17 | 2019-08-09 | 中国科学院深圳先进技术研究院 | KVM virtual machine network optimization method and device under SR-IOV environment |
Non-Patent Citations (1)
Title |
---|
曲桦: "《一种基于SDN的CCN集中控制缓存决策方法》", 《电信科学》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020108536A1 (en) * | 2018-11-28 | 2020-06-04 | 深圳先进技术研究院 | Virtual network resource allocation method and system and electronic device |
CN111427912A (en) * | 2020-03-31 | 2020-07-17 | 拉卡拉支付股份有限公司 | Task processing method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109412865B (en) | 2020-08-25 |
WO2020108536A1 (en) | 2020-06-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10241841B2 (en) | Controlling virtual machine density and placement distribution in a converged infrastructure resource pool | |
US9256452B1 (en) | Providing an instance availability estimate | |
US10324754B2 (en) | Managing virtual machine patterns | |
US9501313B2 (en) | Resource management and allocation using history information stored in application's commit signature log | |
CN103645957B (en) | A kind of resources of virtual machine management-control method and device | |
CN110580197A (en) | Distributed computing architecture for large model deep learning | |
US10489208B1 (en) | Managing resource bursting | |
CN109684074A (en) | Physical machine resource allocation methods and terminal device | |
US20200159663A1 (en) | Controlling a rate of prefetching based on bus bandwidth | |
US9766910B1 (en) | Providing field-programmable devices in a distributed execution environment | |
US10411977B2 (en) | Visualization of workload distribution on server resources | |
US20120323821A1 (en) | Methods for billing for data storage in a tiered data storage system | |
CN109412865A (en) | A kind of virtual network resource allocation method, system and electronic equipment | |
US11188348B2 (en) | Hybrid computing device selection analysis | |
CN109240825A (en) | Elastic method for scheduling task, device, equipment and computer readable storage medium | |
CN103455363B (en) | Command processing method, device and physical host of virtual machine | |
CN112463296B (en) | Method and device for creating virtual machine | |
US11442781B2 (en) | Master image for deploying workloads in a heterogeneous computing environment | |
JP2020532803A (en) | Asynchronous updates of metadata tracks in response to cache hits generated via synchronous ingress and out, systems, computer programs and storage controls | |
US8548881B1 (en) | Credit optimization to minimize latency | |
CN104657087B (en) | A kind of method, apparatus and system of virtual disk mapping | |
US11789512B2 (en) | Increased data storage throttling during power failure | |
CN110383255A (en) | It manages and the client partitions of physical equipment is accessed | |
US11513861B2 (en) | Queue management in solid state memory | |
CN110073321A (en) | A kind of storage control and I/O request processing method |
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 |