CN104994145B - A kind of load-balancing method based on KVM virtualization cluster - Google Patents
A kind of load-balancing method based on KVM virtualization cluster Download PDFInfo
- Publication number
- CN104994145B CN104994145B CN201510351675.5A CN201510351675A CN104994145B CN 104994145 B CN104994145 B CN 104994145B CN 201510351675 A CN201510351675 A CN 201510351675A CN 104994145 B CN104994145 B CN 104994145B
- Authority
- CN
- China
- Prior art keywords
- server
- load
- calculation server
- calculation
- formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1012—Server selection for load balancing based on compliance of requirements or conditions with available server resources
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/101—Server selection for load balancing based on network conditions
Abstract
The present invention provides a kind of load-balancing methods based on KVM virtualization cluster, in a server construction KVM virtualization cluster, which includes a client, a load-balanced server, several calculation servers and a storage server.When client accesses load-balanced server, load-balanced server is called the libvirt java development interfaces of KVM and is obtained the load of each calculation server using SSH remote protocols, in conjunction with user demand situation, the new task requests for reaching server are distributed to the calculation server of distribution weights minimum by the calculation server that distribution weights minimum is found using dynamic requests allocation algorithm.The present invention solves the problems, such as that load distribution is unbalanced between virtual machine in virtual management environment, improves the utilization rate of server virtual resource.
Description
Technical field
The present invention relates to a kind of load-balancing methods based on KVM virtualization cluster, belong to cloud computing virtual technology field.
Background technology
Cloud computing technology is a field technology revolution of IT industry circle, it can dispose computing resource on demand, and user only needs
To be that used resource is paid, in essence, cloud computing refer to user terminal by remotely connect obtain storage, calculate,
Database computing resource.This variation is so that the O＆M degree of specialization of IT infrastructure constantly collects neutralization raising, to right
More stringent requirements are proposed for architecture level, especially server performance.
Virtualization technology be cloud computing technology core composition one of, be by it is various calculating and storage resource fully integrates with
The key technology efficiently utilized, including server virtualization and application virtualization.The environment and physical of each application deployment
Platform is not related, is managed to reach by virtual platform and the operations such as is extended, migrates, backs up to application.
Current popular virtualization product include mainly KVM (Kernel based Virtual Machine),
VMware, XenServer etc..Wherein, the sharpest edges of KVM are to increase income completely, and KVM is the Full-virtualization based on kernel,
Compared with other virtualization products, there is resource utilization height, advantage of good performance.
Currently, in KVM virtualization cluster, when there is a large amount of client access request, presence service resource is in virtual machine
Between distribute unbalanced problem.
Invention content
In view of the deficiencies of the prior art, the invention discloses a kind of load-balancing methods based on KVM virtualization cluster.
The present invention builds KVM virtualization cluster on a server, which includes client, load balancing
Server, calculation server and storage server.When client accesses load-balanced server, load-balanced server passes through
Call KVM libvirt-java development interfaces simultaneously use SSH remote protocols, obtain each calculation server configuration information,
The loading conditions such as Network load status select Optimal calculation service in conjunction with user demand situation using dynamic requests allocation algorithm
Device.
The technical scheme is that：
A kind of load-balancing method based on KVM virtualization cluster, this method is run on a server, in the clothes
Be engaged in device on build KVM virtualization cluster, if the KVM virtualization cluster include a client, a load-balanced server,
Dry platform calculation server and a storage server；LNMP environment is built in load-balanced server, configures nginx's
Nginx.conf files realize reverse proxy；Jdk is installed in load-balanced server, java running environment is built, passes through
“qemu+ssh:// " mode connect several calculation servers, utilize java programming call KVM libvirt-java exploitation
Interface obtains the configuration information of every calculation server, CPU usage, network broadband utilization rate, response time, load, described
Configuration information includes the CPU socket numbers sockets, each of memory, CPU number cpus of memory, each node
Check figure cores, hyperthread threads, the desired cpu frequency mhz of sockets；
Specific steps include：When client accesses load-balanced server, load-balanced server calls KVM's
Libvirt-java development interfaces and the load that each calculation server is obtained using SSH remote protocols utilize dynamic requests point
The new task requests for reaching server are distributed to the calculating service by the calculation server that distribution weights minimum is found with algorithm
Device.
According to currently preferred, the dynamic requests allocation algorithm, specific steps include：
A, the current average load weights of calculation server are soughtShown in calculation formula such as formula (I)：
In formula (I), m refers to the number of calculation server；LiRefer to calculation server SiCurrent load； Refer to calculation server SiThe time of the current customer in response end request of actual measurement,Pass through webpage
Monitoring software measures；ωiRefer to calculation server SiCapability；It refer to calculation server
SiAvailable resources be 0 when customer in response end request time,It measures in the following manner：To calculation server SiConstantly
Increase client request number, as calculation server SiWhen respond request number reaches maximum value, surveyed by web monitor software
Obtain calculation server S at this timeiResponse time；Refer to calculation server SiThe response time of itself, when client sends one
When basic request, pass through the calculation server S of web monitor software detectioniThe respond request time；It refer to calculation server
SiCurrent load weightsCalculation server SiCurrent load weights refer to calculation server SiCurrent load is solid with it
Competent ratio,Calculation formula such as formula (II) shown in：
Meet following relationship when all calculation servers reach load balancing, as shown in formula (III)：
B, the load weights of each calculation server are calculated after the critical decrement factor of increase：The critical decrement factor refers to
Into the load weights of calculation server when critical condition, load weights increase decrement factor δ, then the meter after decrement factor is added
Calculate server SiShown in current load weight computing formula such as formula (IV)：
In formula (IV), when entering critical condition,Calculation server is caused when client request increases
Response time when will appear shake, then the response time is Tr, section Tr~TmaxReferred to as critical condition；When client request quantity
Less than TrWhen corresponding number of requests, number of requests is directly proportional to the response time；When other states, δ=0；
Server can be inhibited to enter saturation state by increasing critical decrement factor, in the program of realization, forbid bearing
The request assigning process that the calculation server that weights are 1 participates in next stage is carried, avoids original on server ask in this way
The case where asking task to be influenced the response time occur to sharply increase by system capability saturation, and to calculating service that load weights are 1
Device is alerted, if more calculation servers are alerted, the server for starting redundancy backup carrys out sharing request load.
C, the request number that next stage each calculation server should be assigned is calculated：Setting is counted when reaching load balancing state
Calculate server SiThe request number of distribution isLoad isThat is next stage calculation server SiThe request that should be assigned
Number；In calculation server SiThe upper request number n currently distributed by web monitor software detectioni, measure present load
Li, it is identical to set every calculation server customer in response end request number in same time t, wherein t<Tmax, therefore, distribution
Request number variation with load variation be consistent, as shown in formula (V)：
Formula (VI) is obtained in substitution formula (V)：
It is rightRounding is to get next stage calculation server SiThe request number that should be assigned；
D, it calculates each calculation server and distributes weights：Distribution weights refer to the request number that calculation server should be assigned
The ratio for the request number that should reach with KVM virtualization cluster calculates calculation server SiDistribution weightsIt calculates public
Shown in formula such as formula (VII)：
The distribution weights for seeking each calculation server find the minimum corresponding calculation server of distribution weights.
The main thought of dynamic requests allocation algorithm is after dynamic monitoring request task is assigned to each calculation server
Influence caused by live load, by feeding back the actual operating duty of calculation server, in conjunction with the calculating service of setting
The intrinsic processing capacity of device, load-balanced server calculate the current distribution weights of every calculation server, find minimum point
With the corresponding calculation server of weights, to achieve the purpose that load balancing.
According to currently preferred, the web monitor software refers to httpwatch.
Beneficial effects of the present invention are：
The method of the invention is worked as the new task requests for reaching client according to KVM virtualization cluster virtual machine
Preceding load state dynamic select Optimal calculation server, the managerial KVM virtualization PC cluster service with high scalability of structure
Device.Using dynamic requests allocation algorithm, dynamic monitoring request task causes each calculation server live load after being assigned
Influence in conjunction with the calculation server capability of setting, born by feeding back the actual operating duty of calculation server
It carries equalization server and calculates every calculation server distribution weights, find the minimum corresponding calculation server of distribution weights,
And the new task requests for reaching client are distributed into the calculation server, achieve the purpose that load balancing.Solution of the present invention
Having determined in virtual management environment, load distributes unbalanced problem between virtual machine, improves the utilization of server virtual resource
Rate.
Specific implementation mode
The present invention is further qualified with reference to embodiment, but not limited to this.
Embodiment 1
A kind of load-balancing method based on KVM virtualization cluster, this method is run on a server, in the clothes
Be engaged in device on build KVM virtualization cluster, if the KVM virtualization cluster include a client, a load-balanced server,
Dry platform calculation server and a storage server；LNMP environment is built in load-balanced server, configures nginx's
Nginx.conf files realize reverse proxy；Jdk is installed in load-balanced server, java running environment is built, passes through
“qemu+ssh:// " mode connect several calculation servers, utilize java programming call KVM libvirt-java exploitation
Interface obtains the configuration information of every calculation server, CPU usage, network broadband utilization rate, response time, load, described
Configuration information includes the CPU socket numbers sockets, each of memory, CPU number cpus of memory, each node
Check figure cores, hyperthread threads, the desired cpu frequency mhz of sockets；
Specific steps include：When client accesses load-balanced server, load-balanced server calls KVM's
Libvirt-java development interfaces and the load that each calculation server is obtained using SSH remote protocols utilize dynamic requests point
The new task requests for reaching server are distributed to the calculating service by the calculation server that distribution weights minimum is found with algorithm
Device.
Embodiment 2
According to load-balancing method described in embodiment 1, difference lies in, the dynamic requests allocation algorithm, specific steps
Including：
A, the current average load weights of calculation server are soughtShown in calculation formula such as formula (I)：
In formula (I), m refers to the number of calculation server；LiRefer to calculation server SiCurrent load； Refer to calculation server SiThe time of the current customer in response end request of actual measurement,Pass through webpage
Monitoring software measures；ωiRefer to calculation server SiCapability；It refer to calculation server
SiAvailable resources be 0 when customer in response end request time,It measures in the following manner：To calculation server SiConstantly
Increase client request number, as calculation server SiWhen respond request number reaches maximum value, surveyed by web monitor software
Obtain calculation server S at this timeiResponse time；Refer to calculation server SiThe response time of itself, when client sends one
When basic request, pass through the calculation server S of web monitor software detectioniThe respond request time；It refer to calculation server
SiCurrent load weightsCalculation server SiCurrent load weights refer to calculation server SiCurrent load is solid with it
Competent ratio,Calculation formula such as formula (II) shown in：
Meet following relationship when all calculation servers reach load balancing, as shown in formula (III)：
B, the load weights of each calculation server are calculated after the critical decrement factor of increase：The critical decrement factor refers to
Into the load weights of calculation server when critical condition, load weights increase decrement factor δ, then the meter after decrement factor is added
Calculate server SiShown in current load weight computing formula such as formula (IV)：
In formula (IV), when entering critical condition,Calculation server is caused when client request increases
Response time when will appear shake, then the response time is Tr, section Tr~TmaxReferred to as critical condition；When client request quantity
Less than TrWhen corresponding number of requests, number of requests is directly proportional to the response time；When other states, δ=0；
Server can be inhibited to enter saturation state by increasing critical decrement factor, in the program of realization, forbid bearing
The request assigning process that the calculation server that weights are 1 participates in next stage is carried, avoids original on server ask in this way
The case where asking task to be influenced the response time occur to sharply increase by system capability saturation, and to calculating service that load weights are 1
Device is alerted, if more calculation servers are alerted, the server for starting redundancy backup carrys out sharing request load.
C, the request number that next stage each calculation server should be assigned is calculated：Setting is counted when reaching load balancing state
Calculate server SiThe request number of distribution isLoad isThat is next stage calculation server SiThe request that should be assigned
Number；In calculation server SiThe upper request number n currently distributed by web monitor software detectioni, measure present load
Li, it is identical to set every calculation server customer in response end request number in same time t, wherein t<Tmax, therefore, distribution
Request number variation with load variation be consistent, as shown in formula (V)：
Formula (VI) is obtained in substitution formula (V)：
It is rightRounding is to get next stage calculation server SiThe request number that should be assigned；
D, it calculates each calculation server and distributes weights：Distribution weights refer to the request number that calculation server should be assigned
The ratio for the request number that should reach with KVM virtualization cluster calculates calculation server SiDistribution weightsIt calculates public
Shown in formula such as formula (VII)：
The distribution weights for seeking each calculation server find the minimum corresponding calculation server of distribution weights.
The main thought of dynamic requests allocation algorithm is after dynamic monitoring request task is assigned to each calculation server
Influence caused by live load, by feeding back the actual operating duty of calculation server, in conjunction with the calculating service of setting
The intrinsic processing capacity of device, load-balanced server calculate the current distribution weights of every calculation server, find minimum point
With the corresponding calculation server of weights, to achieve the purpose that load balancing.
Embodiment 3
According to load-balancing method described in embodiment 2, difference lies in the web monitor software refers to httpwatch.
Claims (2)
1. a kind of load-balancing method based on KVM virtualization cluster, which is characterized in that this method is transported on a server
Row, builds KVM virtualization cluster on the server, and the KVM virtualization cluster is equal including a client, a load
Weigh server, several calculation servers and a storage server；LNMP environment is built in load-balanced server, is configured
The nginx.conf files of nginx realize reverse proxy；Jdk is installed in load-balanced server, builds java running environment,
Pass through " qemu+ssh:// " mode connect several calculation servers, utilize java programming call KVM libvirt-java
Development interface obtains the configuration information of every calculation server, CPU usage, network broadband utilization rate, response time, load,
The configuration information includes the CPU socket numbers sockets, each of memory, CPU number cpus of memory, each node
Check figure cores, hyperthread threads, the desired cpu frequency mhz of sockets；Qemu indicates used virtual platform,
Ssh is the authentication mode for connecting virtual platform, qemu+ssh:// it is a kind of Libvirt specifications of linking objective server；
Specific steps include：When client accesses load-balanced server, load-balanced server calls the libvirt- of KVM
Java development interfaces and the load that each calculation server is obtained using SSH remote protocols, are looked for using dynamic requests allocation algorithm
To the calculation server of distribution weights minimum, the new task requests for reaching server are distributed into the calculation server；
The dynamic requests allocation algorithm, specific steps include：
A, the current average load weights of calculation server are soughtShown in calculation formula such as formula (I)：
In formula (I), m refers to the number of calculation server；LiRefer to calculation server SiCurrent load； Refer to calculation server SiThe time of the current customer in response end request of actual measurement,It is measured by web monitor software；ωi
Refer to calculation server SiCapability； Refer to calculation server SiAvailable resources be 0 when
The time of customer in response end request,It measures in the following manner：To calculation server SiIt is continuously increased client request
Number, as calculation server SiWhen respond request number reaches maximum value, calculation server S at this time is measured by web monitor softwarei
Response time；Refer to calculation server SiThe response time of itself passes through webpage when client sends a basic request
The calculation server S of monitoring software detectioniThe respond request time；Refer to calculation server SiCurrent load weights
Calculation server SiCurrent load weights refer to calculation server SiThe ratio of current load and its capability,Meter
It calculates shown in formula such as formula (II)：
Meet following relationship when all calculation servers reach load balancing, as shown in formula (III)：
B, the load weights of each calculation server are calculated after the critical decrement factor of increase：The critical decrement factor refers to entering
The load weights of calculation server when critical condition, load weights increase decrement factor δ, then the clothes of the calculating after decrement factor are added
Be engaged in device SiShown in current load weight computing formula such as formula (IV)：
In formula (IV), when entering critical condition,When client request increases the sound of cause calculation server
When being shaken between seasonable, then the response time is Tr, section Tr~TmaxReferred to as critical condition；When client request quantity is less than Tr
When corresponding number of requests, number of requests is directly proportional to the response time；When other states, δ=0；
C, the request number that next stage each calculation server should be assigned is calculated：Setting calculates clothes when reaching load balancing state
Be engaged in device SiThe request number of distribution isLoad is That is next stage calculation server SiThe request number that should be assigned；It is counting
Calculate server SiThe upper request number n currently distributed by web monitor software detectioni, measure present load Li, setting is often
Platform calculation server customer in response end request number in same time t is identical, wherein t<Tmax, therefore, the request of distribution
Several variations is consistent with load variation, as shown in formula (V)：
It is obtained by formula (II),
In load balancing state,
Therefore,
Formula (VI) is obtained in substitution formula (V)：
It is rightRounding is to get next stage calculation server SiThe request number that should be assigned；
D, it calculates each calculation server and distributes weights：Distribution weights refer to the request number that should be assigned of calculation server with
The ratio for the request number that KVM virtualization cluster should reach, calculates calculation server SiDistribution weightsCalculation formula is such as
Shown in formula (VII)：
The distribution weights for seeking each calculation server find the minimum corresponding calculation server of distribution weights.
2. load-balancing method according to claim 1, which is characterized in that the web monitor software refers to httpwatch.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510351675.5A CN104994145B (en) | 2015-06-23 | 2015-06-23 | A kind of load-balancing method based on KVM virtualization cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510351675.5A CN104994145B (en) | 2015-06-23 | 2015-06-23 | A kind of load-balancing method based on KVM virtualization cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104994145A CN104994145A (en) | 2015-10-21 |
| CN104994145B true CN104994145B (en) | 2018-08-10 |
Family
ID=54305896
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510351675.5A Active CN104994145B (en) | 2015-06-23 | 2015-06-23 | A kind of load-balancing method based on KVM virtualization cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104994145B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6544872B2 (en) * | 2015-12-30 | 2019-07-17 | ホアウェイ・テクノロジーズ・カンパニー・リミテッド | Load balancing computer device, system and method |
| CN105704233B (en) * | 2016-03-19 | 2019-04-16 | 浙江大学 | A kind of channel distribution method towards Initiative Defense |
| CN106257424B (en) * | 2016-06-16 | 2019-03-22 | 山东大学 | A method of the distributed data base system based on KVM cloud platform realizes automatic telescopic load balancing |
| CN106250781B (en) * | 2016-07-31 | 2019-05-07 | 山东大学 | A kind of user information guard method based on virtual machine increment mirror image |
| CN108234670B (en) * | 2018-01-31 | 2021-01-15 | 北京中安智达科技有限公司 | Zookeeper and netty-based distributed remote calling method |
| CN108199926A (en) * | 2018-01-31 | 2018-06-22 | 维沃移动通信有限公司 | A kind of building method, relevant device and the system of mobile terminal virtual system |
| CN108471457A (en) * | 2018-06-16 | 2018-08-31 | 温州职业技术学院 | Based on distributed node dynamic memory load-balancing method |
| CN109495542B (en) * | 2018-10-16 | 2022-08-02 | 深圳壹账通智能科技有限公司 | Load distribution method based on performance monitoring and terminal equipment |
| CN110012065A (en) * | 2019-02-25 | 2019-07-12 | 贵州格物数据有限公司 | A kind of scheduling of resource platform and method based on virtual technology |
| CN112866334A (en) * | 2020-12-29 | 2021-05-28 | 武汉烽火富华电气有限责任公司 | Video streaming media load balancing method based on dynamic load feedback |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103338228A (en) * | 2013-05-30 | 2013-10-02 | 江苏大学 | Cloud calculating load balancing scheduling algorithm based on double-weighted least-connection algorithm |
| CN103713956A (en) * | 2014-01-06 | 2014-04-09 | 山东大学 | Method for intelligent weighing load balance in cloud computing virtualized management environment |
| CN104202254A (en) * | 2014-08-14 | 2014-12-10 | 江苏省邮电规划设计院有限责任公司 | An intelligent load balancing method based on a cloud calculation platform server |
| CN104580538A (en) * | 2015-02-12 | 2015-04-29 | 山东大学 | Method for improving load balance efficiency of Nginx server |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110078303A1 (en) * | 2009-09-30 | 2011-03-31 | Alcatel-Lucent Usa Inc. | Dynamic load balancing and scaling of allocated cloud resources in an enterprise network |
| WO2013029051A2 (en) * | 2011-08-25 | 2013-02-28 | to fuVIRTUSTREAM, INC. | Systems and methods of host-aware resource management involving cluster-based resource pools |
-
2015
- 2015-06-23 CN CN201510351675.5A patent/CN104994145B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103338228A (en) * | 2013-05-30 | 2013-10-02 | 江苏大学 | Cloud calculating load balancing scheduling algorithm based on double-weighted least-connection algorithm |
| CN103713956A (en) * | 2014-01-06 | 2014-04-09 | 山东大学 | Method for intelligent weighing load balance in cloud computing virtualized management environment |
| CN104202254A (en) * | 2014-08-14 | 2014-12-10 | 江苏省邮电规划设计院有限责任公司 | An intelligent load balancing method based on a cloud calculation platform server |
| CN104580538A (en) * | 2015-02-12 | 2015-04-29 | 山东大学 | Method for improving load balance efficiency of Nginx server |
Non-Patent Citations (3)
| Title |
|---|
| Load Balancing Optimization in Cloud Computing: Applying Endocrine-particale Swarm Optimization;Shahrzad Aslanzadeh et al;《Electro/Information Technology(EIT),2015 IEEE International Conference on》;IEEE;20150523;全文 * |
| 一种面向云服务平台的自适应负载均衡方法;张涛;《中国优秀硕士学位论文全文数据库》;20140531;全文 * |
| 云环境下虚拟机集群***动态负载均衡机制;李立耀;《计算机应用》;20141110;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104994145A (en) | 2015-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104994145B (en) | A kind of load-balancing method based on KVM virtualization cluster | |
| US20200287961A1 (en) | Balancing resources in distributed computing environments | |
| CN105279027B (en) | A kind of virtual machine deployment method and device | |
| CN108322548B (en) | Industrial process data analysis platform based on cloud computing | |
| Caron et al. | Auto-scaling, load balancing and monitoring in commercial and open-source clouds | |
| CN105703940A (en) | Multistage dispatching distributed parallel computing-oriented monitoring system and monitoring method | |
| CN107707612B (en) | Method and device for evaluating resource utilization rate of load balancing cluster | |
| Dargie | Estimation of the cost of VM migration | |
| CN110798517B (en) | Decentralized cluster load balancing method and system, mobile terminal and storage medium | |
| WO2021136137A1 (en) | Resource scheduling method and apparatus, and related device | |
| CN105511953B (en) | Virtual machine load evaluation system, method and service node under cloud environment | |
| CN112506643A (en) | Load balancing method and device of distributed system and electronic equipment | |
| CN103428008A (en) | Big data distribution strategy oriented to multiple user groups | |
| WO2018200162A1 (en) | Anti-flapping system for autoscaling resources in cloud networks | |
| EP3433740A1 (en) | Control device for estimation of power consumption and energy efficiency of application containers | |
| CN102339233A (en) | Cloud computing centralized management platform | |
| Guo et al. | A container scheduling strategy based on neighborhood division in micro service | |
| WO2014084819A1 (en) | Determining at least one cost factor associated with using the cloud to provide an application component | |
| CN115334082A (en) | Load balancing method, device, computer equipment, storage medium and product | |
| CN113590403A (en) | Pressure testing method, device, system, electronic equipment, storage medium and product | |
| CN110597619A (en) | Load balancing method, load balancing device and load balancing server | |
| Weng et al. | Automatic performance tuning for the virtualized cluster system | |
| Tiwari et al. | Dynamic weighted virtual machine live migration mechanism to manages load balancing in cloud computing | |
| CN109981794B (en) | Processing method and device based on block chain node point network and electronic equipment | |
| CN111124669A (en) | Operation method, system, terminal and storage medium of distributed SaaS software |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |