CN107070709A - A kind of NFV implementation methods based on bottom NUMA aware - Google Patents
A kind of NFV implementation methods based on bottom NUMA aware Download PDFInfo
- Publication number
- CN107070709A CN107070709A CN201710209194.XA CN201710209194A CN107070709A CN 107070709 A CN107070709 A CN 107070709A CN 201710209194 A CN201710209194 A CN 201710209194A CN 107070709 A CN107070709 A CN 107070709A
- Authority
- CN
- China
- Prior art keywords
- nfv
- information
- implementation methods
- methods based
- mrow
- 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
-
- 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
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/485—Task life-cycle, e.g. stopping, restarting, resuming execution
- G06F9/4856—Task life-cycle, e.g. stopping, restarting, resuming execution resumption being on a different machine, e.g. task migration, virtual machine migration
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
-
- 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
Abstract
The invention discloses a kind of NFV implementation methods based on bottom NUMA aware, comprise the following steps:Access delay and bandwidth information between step 1, acquisition Node, form information matrix;Step 2, parsing NFV arrangement informations, obtain the specifying information of NFV service chainings;Step 3, according to the access delay and bandwidth information between Node, generate resource dispatching strategy;Step 4, according to resource dispatching strategy perform resources of virtual machine scheduling;Step 5, monitoring module monitoring system resource load, are to need transition state by the vertex ticks loaded higher than set threshold value;It step 6, will be shielded in information matrix of the node obtained by step 1 being labeled in step 5, be then back to step 3.A kind of NFV implementation methods based on bottom NUMA aware of the present invention, with reference to the actual performance and the specifying information of NFV arranging services of real server, dynamic dispatching distributes distribution and resource of the virtual machine thread on physical cpu and used, so as to reach the purpose of optimization overall performance.
Description
Technical field
Realized the present invention relates to computer virtualized technology field, more particularly to a kind of NFV based on bottom NUMA aware
Method.
Background technology
Virtualization is the key technology for realizing network function virtualization.Virtualization technology be it is a kind of by it is existing calculate, storage,
The various actual resources such as network carry out the resource management techniques of abstract conversion.Virtualization technology is the one of IT technology trends
Part, can bring organization and administration mode preferably using actual resource for user, and these actual resources are not by existing
Erection mode, region or the configuration for having resource are limited.Just because of this, virtualization technology, which turns into, realizes that telecommunication service is provided
Source pond, dynamic layout, the key technology of quick-expansion.
ETSI (ETSI) has issued NFV ginsengs as NFV initiation normal structure in 2015 beginning of the years
Examine the serial manuscript such as framework, although ETSI NFV phased Achievements are not the standards enforced, but have obtained the universal of industry
Accreditation, the fact that have become industry standard.Current NFV standard card cage is basicly stable, as shown in Figure 1.NFV templates
Frame mainly has NFV infrastructure, virtual network function and NFV management and layout (NFV Management and
Orchestration,NFV MANO).The current groundwork of industrial quarters is still concentrated in NFV management and layout, and on logical
Still have to be solved with the Performance And Reliability problem of server.
The proprietary hardware device of the communications industry is changed into general multiple-core server, merely for the angle of hardware, property
Can necessarily it decline with reliability aspect.For example in core net virtual equipment, its performance bottleneck is concentrated mainly on I/O interfaces
On the scheduling virtual machine of data forwarding and multiple-core server, probably there is 30%~40% performance loss compared with legacy equipment,
NFV target is to be reduced to that the demand of commercialization could be met within 10%.
In addition, the framework that the polycaryon processor of the generic server of existing main flow is used is nonuniformity internal storage access
(Non-Uniform Memory Access, abbreviation NUMA) framework.The characteristics of NUMA architecture is the associated hardware resource of system
(such as internal memory, CPU) is divided into multiple nodes.In NUMA architecture, processor access local node internal memory speed than
The speed for accessing the internal memory of remote node is fast.Specific communication equipment virtual machine will be run in such generic server, it is right
It is a kind of brand-new application scenarios for generic server, it is original for being normally applied the NUMA scheduling strategies optimized
It may be failed with load-balancing mechanism, and sensitiveness meeting of the virtual network function operated in generic server to IO performances
New challenge is brought to the general-purpose operating system and operation platform.
Therefore, those skilled in the art is directed to a kind of NFV implementation methods based on bottom NUMA aware of exploitation, logical
Under NUMA architecture server, based on dispatching method of virtual machine, with reference to the actual performance and NFV arranging services of real server
Specifying information, dynamic dispatching distributes distribution and resource of the virtual machine thread on physical cpu and used, so as to reach that optimization is overall
The purpose of performance.
The content of the invention
In view of the drawbacks described above of prior art, the technical problems to be solved by the invention, which are that exploitation is a kind of, is based on bottom
The NFV implementation methods of NUMA aware, under general NUMA architecture server, based on dispatching method of virtual machine, with reference to actual clothes
The actual performance of device of being engaged in and the specifying information of NFV arranging services, dynamic dispatching distribute distribution of the virtual machine thread on physical cpu
Used with resource, so as to reach the purpose of optimization overall performance.
To achieve the above object, the invention provides a kind of NFV implementation methods based on bottom NUMA aware, including it is following
Step:
Access delay and bandwidth information between step 1, acquisition Node, form information matrix;
Step 2, parsing NFV arrangement informations, obtain the specifying information of NFV service chainings;
Step 3, according to the access delay and bandwidth information between Node, generate resource dispatching strategy;
Step 4, according to resource dispatching strategy perform resources of virtual machine scheduling;
Step 5, monitoring module monitoring system resource load, are needs by the vertex ticks loaded higher than set threshold value
Transition state;
It step 6, will be shielded in information matrix of the node obtained by step 1 being labeled in step 5, be then back to step
3。
Further, step 1 obtains each Node access delay and band by VMM and host internal memory and I O access detection
Wide information, memory access latencies and the information matrix of bandwidth are organized the formation of by data normalization, according to the information square after normalization
Battle array, draws the compatibility equivalence class partition matrix of each physical core, and record in monitoring module.
Further, the information that step 1 is obtained is bottom hardware information.
Further, step 2 obtains the virtual machine information of NFV service chainings, and record by parsing NFV arrangement informations
Monitoring module.
Further, virtual machine information include by virtual robot arm into the virtual machine quantity of service chaining and tissue topology, each
The resource requirement feature of connected mode, each virtual network function between virtual network function.
Further, step 3 is specially the information obtained according to step 1 and step 2, according to formula
Obtain resource dispatching strategy.
Further, step 4 is completed by scheduler finger daemon.
Further, threshold value set in step 5 is arranged to physical cpu utilization rate higher than 90%, when the thing of node
When managing CPU usage higher than 90%, trigger virtual cpu on the CPU and migrate, and be marked as needing transition state.
Technique effect
1st, the rational management to virtual cpu resource is passed through so that the internal storage access efficiency of each thread is maximized, average retardation
Time is minimum, so that the bandwidth of I/O resources can be fully used;
2nd, thread dynamic migration cost is smaller, and optimal scheduling result can be obtained according to formula 1;
3rd, the characteristic of NFV service chainings is fully taken into account, decision-making is scheduled with reference to the service organization information of concrete application, is protected
Demonstrate,prove the applicability of this method in varied situations.
The technique effect of the design of the present invention, concrete structure and generation is described further below with reference to accompanying drawing, with
It is fully understood from the purpose of the present invention, feature and effect.
Brief description of the drawings
Fig. 1 is a kind of probability of NFV implementation methods based on bottom NUMA aware of the preferred embodiment of the present invention
Figure.
Fig. 2 is a kind of NUMA of NFV implementation methods based on bottom NUMA aware of the preferred embodiment of the present invention
Node structure schematic diagram.
Fig. 3 is a kind of flow of NFV implementation methods based on bottom NUMA aware of the preferred embodiment of the present invention
Schematic diagram.
Fig. 4 is a kind of NFV marks of NFV implementation methods based on bottom NUMA aware of the preferred embodiment of the present invention
Collimator frame schematic diagram.
Embodiment
As shown in figure 1, the preferred embodiment of the present invention provides a kind of NFV implementation methods based on bottom NUMA aware
It is briefly to obtain hardware information and service chain information, and forms information matrix, then by information matrix record in monitoring module,
According to the record of monitoring module, scheduler module is scheduled to resources of virtual machine.
As shown in Fig. 2 the preferred embodiment of the present invention provides a kind of physical machine processing based on bottom NUMA aware
Device node and high-performance network interface card topological structure schematic diagram, the asymmetric connection mode between processor node and network interface card therein is mesh
One of framework that preceding main flow processor is generally used.
As shown in figure 3, the preferred embodiment of the present invention provides a kind of NFV realization sides based on bottom NUMA aware
Method, comprises the following steps:
Access delay and bandwidth information between step 1, acquisition Node, form information matrix;Specifically, VMM and place are passed through
Host memory and I O access detection obtain each node access delay and bandwidth information, and data normalization is organized the formation of into internal memory visits
The information matrix of delay and bandwidth is asked, according to the matrix after normalization, the compatibility equivalence class partition square of each physical core is drawn
Battle array, is recorded in monitoring module.The information that this step is obtained is bottom hardware information.
The information mainly includes:Node inter access and the delay (Latency) of outside access and bandwidth
(Bandwidth) as shown in table 1, what is gone out marked in wherein external and internal two is main influence factor.
System break is as unified influence factor therefore no longer especially marks, by the data value Unified Characterization measured.Internal access
Postpone information and bandwidth statistics as shown in table 2, form delay and bandwidth matrices respectively, be used for as input parameter in formula 1.This
Data are recorded in monitoring module after normalization, are stored as system platform essential information.Due to this information only with it is hard
Part is related, therefore when expanding to different platform, only need to update the information.
Table 1
External | Internal | |
Latency | I/O | QPI |
Bandwidth | I/O | QPI |
Step 2, parsing NFV arrangement informations, obtain the specifying information of NFV service chainings;Specifically, by parsing NFV layouts
Information, obtains the virtual machine information of NFV service chainings, and record in monitoring module.Virtual machine information include by virtual robot arm into
Connected mode, the money of each virtual network function between the virtual machine quantity and topological, each virtual network function of tissue of service chaining
Source characteristics of demand.
By this information record in monitoring module, and it is used as repeatedly according in the state transition equation of this information in equation 1
For selection gist.
Table 2
Node0 | Node1 | Node2 | Node3 | … | |
Node0 | |||||
Node1 | |||||
Node2 | |||||
Node3 | |||||
… |
Step 3, according to the access delay and bandwidth information between Node, generate resource dispatching strategy;According to below equation
Generate resource dispatching strategy:
Step 4, according to resource dispatching strategy perform resources of virtual machine scheduling;This process is completed by scheduler finger daemon.
Step 5, monitoring module monitoring system resource load, by load, higher than set threshold value, (node physical cpu is used
Rate average value is to need transition state higher than vertex ticks 90%);
It step 6, will be shielded in information matrix of the node obtained by step 1 being labeled in step 5, be then back to step
3。
After migration performing module completes the migration to virtual cpu, for virtual cpu be also just migrated to pair
On the node answered, the scheduling to virtual machine virtual processor resources this completes.
As shown in figure 4, the standard NFV based on bottom NUMA aware of the preferred embodiment of the present invention realizes structural representation
Figure is briefly to instruct a kind of specific reality based on bottom physical server performance aware under framework based on etsi standard NFV
It is existing.The integrity demands of NFV functions had both been can guarantee that based on such a framework, performances of the NFV under concrete application scene can be also improved,
Better meet industrial quarters realizes target.
In summary, a kind of NFV implementation methods based on bottom NUMA aware of the invention take in general NUMA architecture
It is engaged under device, based on dispatching method of virtual machine, with reference to the actual performance and the specifying information of NFV arranging services of real server, moves
Distribution and resource of the state dispatching distribution virtual machine thread on physical cpu are used, so as to reach the purpose of optimization overall performance.It is logical
Cross the rational management to virtual cpu resource and I O access so that the internal memory and I O access efficiency of each virtual cpu are maximized, it is average
Time delay is minimum, so that the bandwidth of I/O resources can be fully used.Thread dynamic migration cost is smaller, according to
Formula 1 can obtain optimal scheduling result.Further, calculated by introducing two information matrixs and I O access delay vector
It can make it possible priority scheduling of resource under virtual environment.
Preferred embodiment of the invention described in detail above.It should be appreciated that one of ordinary skill in the art without
Need creative work just can make many modifications and variations according to the design of the present invention.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical scheme, all should be in the protection domain being defined in the patent claims.
Claims (8)
1. a kind of NFV implementation methods based on bottom NUMA aware, it is characterised in that comprise the following steps:
Access delay and bandwidth information between step 1, acquisition Node, form information matrix;
Step 2, parsing NFV arrangement informations, obtain the specifying information of NFV service chainings;
Step 3, according to the access delay and bandwidth information between Node, generate resource dispatching strategy;
Step 4, according to the resource dispatching strategy perform resources of virtual machine scheduling;
Step 5, monitoring module monitoring system resource load, are to need migration by the vertex ticks loaded higher than set threshold value
State;
It step 6, will be shielded in described information matrix of the node obtained by step 1 being labeled in step 5, be then back to step
3。
2. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 1, it is characterised in that the step 1
Each Node access delay and bandwidth information is obtained by VMM and host internal memory and I O access detection, by data normalization group
The information matrix to form memory access latencies and bandwidth is knitted, according to the information matrix after normalization, the parent for drawing each physical core
With property equivalence class partition matrix, and record in monitoring module.
3. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 2, it is characterised in that the step 1
The information of acquisition is bottom hardware information.
4. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 1, it is characterised in that the step 2
By parsing NFV arrangement informations, the virtual machine information of NFV service chainings is obtained, and record in monitoring module.
5. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 4, it is characterised in that described virtual
Machine information include by virtual robot arm into service chaining virtual machine quantity and topological, each virtual network function of tissue between connection
Mode, the resource requirement feature of each virtual network function.
6. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 1, it is characterised in that the step 3
The information specially obtained according to the step 1 and the step 2, according to formula
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mi>f</mi>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
<mo>,</mo>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>f</mi>
<mo>(</mo>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
<mo>+</mo>
<mi>M</mi>
<mi>a</mi>
<mi>x</mi>
<mo>{</mo>
<mi>M</mi>
<mi>a</mi>
<mi>t</mi>
<mi>r</mi>
<mi>i</mi>
<msub>
<mi>x</mi>
<mrow>
<mi>n</mi>
<mi>o</mi>
<mi>d</mi>
<mi>e</mi>
<mi>A</mi>
<mi>f</mi>
<mi>f</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>&lsqb;</mo>
<mi>i</mi>
<mo>-</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>&lsqb;</mo>
<mi>i</mi>
<mo>&rsqb;</mo>
<mo>,</mo>
<mi>A</mi>
<mi>r</mi>
<mi>r</mi>
<mi>a</mi>
<msub>
<mi>y</mi>
<mrow>
<mi>I</mi>
<mi>O</mi>
<mi>A</mi>
<mi>f</mi>
<mi>f</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>&lsqb;</mo>
<mi>i</mi>
<mo>&rsqb;</mo>
<mo>}</mo>
<mo>,</mo>
<mi>i</mi>
<mo>></mo>
<mn>0</mn>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Obtain resource dispatching strategy.
7. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 1, it is characterised in that the step 4
Completed by scheduler finger daemon.
8. a kind of NFV implementation methods based on bottom NUMA aware as claimed in claim 1, it is characterised in that the step 5
In set threshold value be arranged to physical cpu utilization rate higher than 90%, when the physical cpu utilization rate of node is higher than 90%,
Trigger virtual cpu on the CPU to migrate, and be marked as needing transition state.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710209194.XA CN107070709B (en) | 2017-03-31 | 2017-03-31 | NFV (network function virtualization) implementation method based on bottom NUMA (non uniform memory Access) perception |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710209194.XA CN107070709B (en) | 2017-03-31 | 2017-03-31 | NFV (network function virtualization) implementation method based on bottom NUMA (non uniform memory Access) perception |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107070709A true CN107070709A (en) | 2017-08-18 |
CN107070709B CN107070709B (en) | 2020-06-26 |
Family
ID=59601603
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710209194.XA Active CN107070709B (en) | 2017-03-31 | 2017-03-31 | NFV (network function virtualization) implementation method based on bottom NUMA (non uniform memory Access) perception |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107070709B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107967180A (en) * | 2017-12-19 | 2018-04-27 | 上海交通大学 | Based on resource overall situation affinity network optimized approach and system under NUMA virtualized environments |
CN110545228A (en) * | 2019-07-26 | 2019-12-06 | 北京邮电大学 | service function chain request processing method and system |
CN110738316A (en) * | 2018-07-20 | 2020-01-31 | 北京三星通信技术研究有限公司 | Operation method and device based on neural network and electronic equipment |
CN112306628A (en) * | 2020-10-12 | 2021-02-02 | 上海交通大学 | Virtual network function resource management framework based on multi-core server |
CN113518016A (en) * | 2021-06-22 | 2021-10-19 | 新华三大数据技术有限公司 | Message sending method, VNFM and computer readable storage medium |
CN115794317A (en) * | 2023-02-06 | 2023-03-14 | 天翼云科技有限公司 | Processing method, device, equipment and medium based on virtual machine |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104050043A (en) * | 2014-06-17 | 2014-09-17 | 华为技术有限公司 | Share cache perception-based virtual machine scheduling method and device |
CN104572307A (en) * | 2015-01-30 | 2015-04-29 | 无锡华云数据技术服务有限公司 | Method for flexibly scheduling virtual resources |
CN104615480A (en) * | 2015-02-04 | 2015-05-13 | 上海交通大学 | Virtual processor scheduling method based on NUMA high-performance network processor loads |
US20160306675A1 (en) * | 2015-04-17 | 2016-10-20 | Vmware, Inc. | Proactive high availability in a virtualized computer system |
CN106095576A (en) * | 2016-06-14 | 2016-11-09 | 上海交通大学 | Under virtualization multi-core environment, nonuniformity I/O accesses resources of virtual machine moving method |
CN106293944A (en) * | 2016-08-11 | 2017-01-04 | 上海交通大学 | System and optimization method is accessed based on nonuniformity I/O under virtualization multi-core environment |
CN106533723A (en) * | 2015-09-11 | 2017-03-22 | 中兴通讯股份有限公司 | Virtual resource scheduling method, device and system |
-
2017
- 2017-03-31 CN CN201710209194.XA patent/CN107070709B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104050043A (en) * | 2014-06-17 | 2014-09-17 | 华为技术有限公司 | Share cache perception-based virtual machine scheduling method and device |
CN104572307A (en) * | 2015-01-30 | 2015-04-29 | 无锡华云数据技术服务有限公司 | Method for flexibly scheduling virtual resources |
CN104615480A (en) * | 2015-02-04 | 2015-05-13 | 上海交通大学 | Virtual processor scheduling method based on NUMA high-performance network processor loads |
US20160306675A1 (en) * | 2015-04-17 | 2016-10-20 | Vmware, Inc. | Proactive high availability in a virtualized computer system |
CN106533723A (en) * | 2015-09-11 | 2017-03-22 | 中兴通讯股份有限公司 | Virtual resource scheduling method, device and system |
CN106095576A (en) * | 2016-06-14 | 2016-11-09 | 上海交通大学 | Under virtualization multi-core environment, nonuniformity I/O accesses resources of virtual machine moving method |
CN106293944A (en) * | 2016-08-11 | 2017-01-04 | 上海交通大学 | System and optimization method is accessed based on nonuniformity I/O under virtualization multi-core environment |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107967180A (en) * | 2017-12-19 | 2018-04-27 | 上海交通大学 | Based on resource overall situation affinity network optimized approach and system under NUMA virtualized environments |
CN107967180B (en) * | 2017-12-19 | 2019-09-10 | 上海交通大学 | Based on resource overall situation affinity network optimized approach and system under NUMA virtualized environment |
CN110738316A (en) * | 2018-07-20 | 2020-01-31 | 北京三星通信技术研究有限公司 | Operation method and device based on neural network and electronic equipment |
CN110545228A (en) * | 2019-07-26 | 2019-12-06 | 北京邮电大学 | service function chain request processing method and system |
CN110545228B (en) * | 2019-07-26 | 2020-11-20 | 北京邮电大学 | Service function chain request processing method and system |
CN112306628A (en) * | 2020-10-12 | 2021-02-02 | 上海交通大学 | Virtual network function resource management framework based on multi-core server |
CN112306628B (en) * | 2020-10-12 | 2022-10-14 | 上海交通大学 | Virtual network function resource management system based on multi-core server |
CN113518016A (en) * | 2021-06-22 | 2021-10-19 | 新华三大数据技术有限公司 | Message sending method, VNFM and computer readable storage medium |
CN113518016B (en) * | 2021-06-22 | 2022-08-30 | 新华三大数据技术有限公司 | Message sending method, VNFM and computer readable storage medium |
CN115794317A (en) * | 2023-02-06 | 2023-03-14 | 天翼云科技有限公司 | Processing method, device, equipment and medium based on virtual machine |
Also Published As
Publication number | Publication date |
---|---|
CN107070709B (en) | 2020-06-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107070709A (en) | A kind of NFV implementation methods based on bottom NUMA aware | |
Tong et al. | A hierarchical edge cloud architecture for mobile computing | |
Lai et al. | Cost-effective app user allocation in an edge computing environment | |
Lin et al. | QoS-aware data replication for data-intensive applications in cloud computing systems | |
Mercier et al. | Improving MPI applications performance on multicore clusters with rank reordering | |
CN105933408B (en) | A kind of implementation method and device of Redis universal middleware | |
Redekopp et al. | Optimizations and analysis of bsp graph processing models on public clouds | |
CN104754008B (en) | Network storage node, network store system and the device and method for network storage node | |
US10666570B2 (en) | Computing infrastructure resource-workload management methods and apparatuses | |
CN107729138B (en) | Method and device for analyzing high-performance distributed vector space data | |
CN113641457A (en) | Container creation method, device, apparatus, medium, and program product | |
CN110308984B (en) | Cross-cluster computing system for processing geographically distributed data | |
CN104536937A (en) | Big data appliance realizing method based on CPU-GPU heterogeneous cluster | |
Duong et al. | A framework for dynamic resource provisioning and adaptation in iaas clouds | |
EP3273664B1 (en) | Data processing method and device, server, and controller | |
CN104753706B (en) | A kind of distributed type assemblies configuring management method and device | |
Maqsood et al. | Congestion-aware core mapping for network-on-chip based systems using betweenness centrality | |
CN102662740A (en) | Asymmetric multi-core system and realization method thereof | |
CN109614227A (en) | Task resource concocting method, device, electronic equipment and computer-readable medium | |
CN102929769A (en) | Virtual machine internal-data acquisition method based on agency service | |
Tseng et al. | Service-oriented virtual machine placement optimization for green data center | |
CN108512782A (en) | Accesses control list is grouped method of adjustment, the network equipment and system | |
Brandt et al. | Demonstrating improved application performance using dynamic monitoring and task mapping | |
Xu et al. | Amnis: Optimized stream processing for edge computing | |
Han et al. | EdgeTuner: Fast scheduling algorithm tuning for dynamic edge-cloud workloads and resources |
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 |