CN103428008B - The big data distributing method of facing multiple users group - Google Patents
The big data distributing method of facing multiple users group Download PDFInfo
- Publication number
- CN103428008B CN103428008B CN201310383301.2A CN201310383301A CN103428008B CN 103428008 B CN103428008 B CN 103428008B CN 201310383301 A CN201310383301 A CN 201310383301A CN 103428008 B CN103428008 B CN 103428008B
- Authority
- CN
- China
- Prior art keywords
- virtual server
- virtual
- server
- cluster
- big data
- 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
Landscapes
- Computer And Data Communications (AREA)
Abstract
The present invention relates to the big data distribution policy of a kind of facing multiple users group, including the multiple virtual servers operated on physical server and Virtual Machine Manager module, multiple virtual servers one Virtual Server Cluster of composition, main virtual server it is provided with in Virtual Server Cluster, main virtual server is obtained by selection in virtual server by Paxos algorithm, monitoring module it is additionally provided with in Virtual Server Cluster, the load state of the virtual server in monitoring cluster, and at any time by adding in cluster or removing virtual server.It is an advantage of the current invention that step is succinct, wherein calculation procedure is few, and fully compensate for pelletizes in response big data distribution time control spends in extensive, it is impossible to the problem accurately adjusting load balancing, has preferable using value.
Description
Technical field
The present invention relates to the big data distributing method of a kind of facing multiple users group.
Background technology
Along with the fast development of Internet and improving constantly of portfolio, network data access flow is rapid
Increase, the particularly access to data center, large enterprise and portal website etc..Meanwhile, server site by HTTP,
The application programs such as FTP, provide increasingly abundanter content and information for visitor, and server is gradually flooded by data.
In the computer network of magnanimity big data transmission, relative to the development of network technology, server process speed and
The growth of internal storage access speed is but well below the network bandwidth and the growth of application service, and the network bandwidth brings while increasing
The growth of number of users, also makes server resource consume serious, thus server becomes network bottleneck.Traditional unit mould
Formula, also tends to become network failure point.
All these demands that all application service is proposed high-performance and high reliability.The most most important two kinds of skills
Art is load-balancing technique and Intel Virtualization Technology.
Load balancing provides a kind of cheap, effective, transparent method extended network equipment and the bandwidth of server, increase
Handling capacity, Strengthens network data-handling capacity, improve motility and the availability of network.
Intel Virtualization Technology can expand the capacity of hardware, simplifies the re-configuration process of software.Along with multiple nucleus system in recent years,
Cluster, the grid even widespread deployment of cloud computing, Intel Virtualization Technology advantage on business is applied embodies day by day, not only reduces
IT cost, but also enhance security of system and reliability.
In the system towards multiple customer group, often different user group is different to the resource requirement of server,
And this demand is dynamically change, often there is certain user group's system overload or some system-computed in prior art
The situation that resource is idle, it is impossible to carry out reacting and adjusting according to user's request rapidly, reduce the reality of load-balancing technique
Effect, it is therefore necessary to develop a kind of new types of data distribution method that can make up existing load-balancing technique.
Summary of the invention
The present invention is directed to prior art according to different user, the dynamically change of resource requirement to be adjusted in real time, lead
The shortcoming that cause system service efficiency is low, it is provided that the big data distributing method of a kind of novel facing multiple users group.
For achieving the above object, the present invention can take following technical proposals:
The big data distributing method of facing multiple users group, including the multiple virtual servers operated on physical server with
And Virtual Machine Manager module, multiple virtual servers one Virtual Server Cluster of composition, it is provided with in Virtual Server Cluster
Main virtual server, main virtual server is obtained by selection in virtual server by Paxos algorithm;Specifically comprise the following steps that
1) on physical server, run multiple virtual server composition Virtual Server Cluster;
2) virtual server produces the loading statistics of self, synchronizes all void in the range of Virtual Server Cluster
Intend the loading statistics of server;
3) loading statistics of the virtual server in this cluster is monitored by Virtual Server Cluster;
4) above-mentioned steps 3) in, if the quantity of virtual server idle in Virtual Server Cluster exceedes threshold value, then
Randomly choose an idle virtual server, in Virtual Server Cluster, delete this virtual server, notify virtual machine pipe
Reason module reclaims the virtual server that is deleted in system free virtual server pools, the profit of the virtual server of described free time
Use rate utl=0, utilization rateWherein, Load is the system load of virtual server, and Capacity is virtual
The disposal ability of server;
5) above-mentioned steps 4) in, if utilization rate utl-Group of virtual server is more than threshold in Virtual Server Cluster
It is worth, and system free virtual server pools is not empty, then moved by the virtual server in a system free virtual server pools
Go out to Virtual Server Cluster, utilization rateWherein,For
The utilization rate of single virtual server, LoadiFor the system load of virtual server, CapacityiPlace for virtual server
Reason ability, G is the number of the virtual server in Virtual Server Cluster;
6) Virtual Server Cluster receive user data distribution request time, according to above-mentioned steps 4) gained Virtual Service
Utilization rate utl of device selects the virtual server that load is minimum, and data distribution request is sent to minimum virtual of this load
Server process.
As preferably, including multiple Virtual Server Cluster, the equal position of virtual server in each Virtual Server Cluster
On Same Physical server.
As preferably, also include monitoring module, above-mentioned steps 3) in, by the monitoring module load statistics to virtual server
Data are monitored, and submit to Virtual Machine Manager module.
As preferably, described monitoring module is arranged in virtual server;Also include step in detail below: above-mentioned steps
3), in, the loading statistics that monitoring is obtained by monitoring module enters between each virtual server in Virtual Server Cluster
Row synchronous transfer, carries out on-line checking each other between each monitoring module, described on-line checking is, within the detection cycle of 1s,
Monitoring module sends a connection request to busy virtual server, if continuous three detection cycles all do not meet with a response
Then this virtual server loses network connection;The synchronization biography of loading statistics is otherwise carried out with this busy virtual server
Defeated.
As preferably, also include step in detail below: when monitoring module loses after network is connected with main virtual server, logical
Crossing Paxos algorithm and reselect main virtual server, new main virtual server notice Virtual Machine Manager module reclaims old master
Virtual server.
As preferably, described Virtual Machine Manager module uses the virtual server that ltsh chain table structure organization is idle, Hash
The Key of chained list is the ID of the physical server running virtual server;Also include step in detail below: above-mentioned steps 4) in,
When virtual server is recycled to system free virtual server pools, Virtual Machine Manager module is according to this virtual server place
This virtual server is put in ltsh chain table by the ID of physical server;Above-mentioned steps 5) in, when idle virtual server moves
When going out system free virtual server pools, Virtual Machine Manager module preferential searching system free virtual server pools, with target
Virtual server in Virtual Server Cluster runs on the virtual server of Same Physical server and is allocated.
As preferably, also include peer-to-peer message distribution platform, the load of all virtual servers in synchronized clusters
Statistical data, the form that described peer-to-peer message distribution platform is broadcasted by message virtual server in cluster sends load
Statistical data.
Due to the fact that and have employed above technical scheme that there is significant technique effect:
The present invention, based on the dynamic adjustment to virtual server, compensate for carrying out between users the deficiency of load balancing,
Meanwhile, by the monitoring to loading statistics, user's request can be made and dynamically feeding back timely, compared to traditional quality
Ensure that service technology is more fine only for Control granularity for the service of certain physical equipment, more can meet current data distribution
The particular demands of business.Secondly, distribute reclaim mechanism by virtual machine, can more effectively manage resources of virtual machine, due to void
On the basis of plan machine distribution reclaim mechanism builds on loading statistics synchronous transfer, user is asked the anti-of required load
Should be the rapidest, accurate, compared to prior art, there is more high flexibility, more effectively utilize server resource, and can
Reduce the power consumption of data center.
Further, by ltsh chain table technology, and the Key of chained list is associated with the ID of physical server, it is ensured that
When virtual server is allocated, runs on the virtual server on same physical server and can be included into same cluster, carry
The efficiency of high Data Migration and the level of resources utilization.
In order to loading statistics is monitored the most accurately, each virtual server is provided with prison
Control module, monitoring module synchronized loading statistical data each other, also enters virtual server and main virtual server simultaneously
Row on-line checking, decreases due to the server failure impact on system entirety resource, improves the stability of system.
Additionally, monitoring module carries out data syn-chronization also by a peer-to-peer message distribution platform, further increase same
Step rate, improves the accuracy of distribution virtual server.
Accompanying drawing explanation
Fig. 1 is the main flow schematic diagram of the present invention.
Fig. 2 is the system structure schematic diagram of the present invention.
Fig. 3 is the data structure schematic diagram in Virtual Machine Manager module.
Detailed description of the invention
Below in conjunction with embodiment, the present invention is described in further detail.
Embodiment 1
The big data distributing method of facing multiple users group, its system architecture diagram is as in figure 2 it is shown, the Virtual Machine Manager of system
Module 2 is responsible for distribution and the recovery of virtual server 1, and traffic load statistics between responsible Virtual Server Cluster 3
Data.Virtual server 1, according to the application demand of different customer groups, is divided into several virtual server groups by system, virtual
Server group, as Virtual Server Cluster 3, has Group1 and Group2.Have a number of in each virtual server group
Virtual server 1, and a main virtual server 1 produced based on Paxos algorithm.Group1 has VS1~VS4,4 void
Intend server 1, and a main virtual server 1.In group, each virtual server 1 has a monitoring module 5, and load is same
Loading statistics in step group, and maintain the on-line checking between virtual server 1.
As it is shown on figure 3, Virtual Machine Manager module 2 safeguards idle virtual clothes by the data structure of a ltsh chain table
Business device 1.Each virtual server 1 carries out Hash according to No. ID of its affiliated physical server.This have the advantage that
Distributing virtual server 1 when, it is ensured that the virtual server 1 that the application of same customer group is used concentrates on as far as possible
In a small amount of physical server, such that it is able to the network traffics of communication in minimizing group, and by Hash table, search efficiency is compared
Also improve a lot in common list structure.
On the basis of the framework of apparatus above, the big data distributing method of the facing multiple users group of the present invention, such as Fig. 1 institute
Showing, it comprises the following steps:
Each physical server node is divided into multiple virtual server 1, and one group of virtual server 1 forms a cluster
It is responsible for processing the request of a class customer group distributed tasks;
Each virtual server 1 has allocated certain resource, if user is not specified, then uses the acquiescence of system
Value.One group, towards in the Virtual Server Cluster 3 of particular group, has a main virtual server 1, when this kind of user sends
During request, request first passes around main virtual server 1.Main virtual server 1 is by the association between virtual server 1 in cluster
Business elects, and after being chosen to be main virtual server 1, this virtual server 1 externally provides service.
In cluster, by on-line checking, other virtual servers 1 determine that main virtual server 1 the most normally works.Once
Main virtual server 1 is made mistakes or off-line, then renegotiated by above-mentioned Paxos algorithm and elect the virtual clothes of new master
Business device 1, the newest main virtual server 1 notifies Virtual Machine Manager module 2, reclaims the virtual server 1 made mistakes.
Each virtual server 1 safeguards own load statistical data.Often in group Virtual Server Cluster 3, by equity
The loading statistics of all virtual servers 1 in formula message distribution platform synchronized clusters group;
If the load of virtual server A is LoadA, system processing power is CapacityA, then the profit of virtual server A
Use rate utlAFor: utlA=LoadA/CapacityA.In group, the utilization rate of virtual server is:
Wherein G is group member's number.
The groundwork of monitoring programme is as follows: often have a monitoring module 5, monitoring void in group Virtual Server Cluster 3
Intend in server cluster 3 groups, the load of each virtual server 1;Utl-Group data in periodic collection group, and to being
System Virtual Machine Manager module 2 reports;If there being member to lose the connection between monitoring module 5, then notice virtual machine pipe in group
Reason module 2 reclaims the resource of this virtual server 1.
If idle server count exceedes certain threshold value in monitoring module 5 detects cluster, this threshold value is default
Value, then randomly select an idle server in cluster, then delete this server in cluster, and notify virtual machine pipe
Reason module 2 reclaims in this virtual server 1 to system free virtual server pools 4;Wherein idle virtual server 1 refers to
Do not provide a user with service at present, i.e. utl=0%.
If the load sum of all virtual servers 1 is more than threshold value in cluster being detected for monitoring module 5, this threshold value is
Setting value, and system free virtual server pools 4 is not empty, then and one virtual server 1 of application is taken by system free virtual
Business device pond 4 adds in cluster;
In Virtual Machine Manager module 2, by the virtual machine server 1 that ltsh chain table data structure organization is idle.Hash table
Structure in Key be the ID of physical server belonging to virtual machine server 1, i.e. Virtual Machine Manager module 2 is according to virtual clothes
The described physical server of business device 1 carries out ltsh chain table organization and administration to virtual server 1.
When user discharges virtual machine server 1, Virtual Machine Manager module 2 is according to virtual server 1 institute reclaimed
This virtual server 1 is put in Hash table by the physical server belonged to.
When certain Virtual Server Cluster 3 files an application request, Virtual Machine Manager module 2 first looks for and this cluster
Interior virtual server 1 has the free virtual server 1 of same physical server and is allocated.
The purpose of this organizational form is to make each virtual server 1 in Virtual Server Cluster 3 concentrate on several thing
Reason server, so that the communication between virtual server 1 faster, and can save physical server in cluster
Resource.
One group of Virtual Server Cluster 3 is when processing the request of a class user, according to each virtual server 1 in group
Loading statistics, selects to load minimum virtual server 1 and processes this request;
When server is filed a request by user, this request arrives first at the main virtual server in Virtual Server Cluster 3
1, main virtual server 1, according to the load of each virtual server 1 in group, selects a virtual server 1 to provide the user
Service.
In Virtual Machine Manager module 2, main virtual server 1 selects cluster interior-deficiency to intend server 1 and provides the user service
Algorithm be: in Virtual Server Cluster 3, according to the present load of virtual server 1, virtual server 1 is divided into two groups,
Virtual server 1 in first group currently provides service, and loads and reached the threshold value that user sets.Second group
The threshold value that sets of the load not up to user of virtual server 1.When user asks to arrive, main virtual server 1 is from the
In two groups, the virtual server 1 selecting load maximum provides the user service.As such, it is possible to make to call request process as far as possible
Concentrate, so that virtual server 1 quantity in Virtual Server Cluster 3 lacking as far as possible so that Virtual Machine Manager module 2
More free virtual server 1 can be had.
In concrete application example, the cycle of the on-line checking of monitoring module 5 is 1s, if continuous three on-line checking please
Connection is asked all not meet with a response, then it is assumed that this virtual server 1 loses network and connects.Time initial, a Virtual Server Cluster
Virtual server 1 number of 3 is 5, if the number of the most busy virtual machine server 1 accounts for the virtual clothes in cluster
More than the 80% of business device 1 number, then apply for new virtual server 1 resource to system virtual machine management module 2.If worked as
Free virtual server 1 number in front Virtual Server Cluster 3 exceedes 40% of virtual server 1 sum in cluster, then lead to
Know that system virtual machine management module 2 reclaims the virtual server 1 of free time.Monitoring module 5 was every 3 seconds, and i.e. 3 are detected the cycles
Loading statistics in synchronized clusters.
In a word, the foregoing is only presently preferred embodiments of the present invention, all equalizations made according to scope of the present invention patent
Change and modification, all should belong to the covering scope of patent of the present invention.
Claims (7)
1. the big data distributing method of a facing multiple users group, it is characterized in that, including the multiple virtual servers (1) operated on physical server (6) and Virtual Machine Manager module (2), multiple virtual servers (1) one Virtual Server Cluster (3) of composition, being provided with main virtual server (1) in Virtual Server Cluster (3), main virtual server (1) is obtained by selection in virtual server (1) by Paxos algorithm;Specifically comprise the following steps that
1) Virtual Server Cluster (3) is formed at the upper multiple virtual servers (1) that run of physical server (6);
2) virtual server (1) produces the loading statistics of self, synchronizes the loading statistics of all virtual servers (1) in the range of Virtual Server Cluster (3);
3) loading statistics of the virtual server (1) in this cluster is monitored by Virtual Server Cluster (3);
4) above-mentioned steps 3) in, if the quantity of virtual server (1) idle in Virtual Server Cluster (3) exceedes threshold value, then randomly choose an idle virtual server (1), this virtual server (1) is deleted in Virtual Server Cluster (3), it is interior to system free virtual server pools (4) that notice Virtual Machine Manager module (2) reclaims the virtual server (1) being deleted, utilization rate utl=0 of the virtual server (1) of described free time, utilization rateWherein, Load is the system load of virtual server (1), and Capacity is the disposal ability of virtual server (1);
5) above-mentioned steps 4) in, if utilization rate utl-Group of Virtual Server Cluster (3) interior virtual server (1) is more than threshold value, and system free virtual server pools (4) is not empty, then the virtual server (1) in system free virtual server pools (4) is moved in Virtual Server Cluster (3), utilization rateWherein,For the utilization rate of single virtual server (1), LoadiFor the system load of virtual server (1), CapacityiFor the disposal ability of virtual server (1), G is the number of the virtual server (1) in Virtual Server Cluster (3);
6) Virtual Server Cluster (3) receive user data distribution request time, according to above-mentioned steps 4) utilization rate utl of gained virtual server (1) selects the minimum virtual server (1) of load, and data distribution request is sent to the virtual server (1) that this load is minimum and process.
The big data distributing method of facing multiple users group the most according to claim 1, it is characterized in that, including multiple Virtual Server Cluster (3), the virtual server (1) in each Virtual Server Cluster (3) is respectively positioned on Same Physical server (6).
The big data distributing method of facing multiple users group the most according to claim 1, it is characterized in that, also include monitoring module (5), above-mentioned steps 3) in, by monitoring module (5), the loading statistics of virtual server (1) is monitored, and submits to Virtual Machine Manager module (2).
The big data distributing method of facing multiple users group the most according to claim 3, it is characterised in that described monitoring module (5) is arranged in virtual server (1);Also include step in detail below: above-mentioned steps 3) in, the loading statistics that monitoring is obtained by monitoring module (5) carries out synchronous transfer between each virtual server (1) in Virtual Server Cluster (3), on-line checking is carried out each other between each monitoring module (5), described on-line checking is, within the detection cycle of 1s, monitoring module (5) sends a connection request to busy virtual server (1), if continuous three detection cycles all do not meet with a response, this virtual server (1) loses network and connects;Otherwise carry out the synchronous transfer of loading statistics with this busy virtual server (1).
The big data distributing method of facing multiple users group the most according to claim 4, it is characterized in that, also include step in detail below: when monitoring module (5) loses after network is connected with main virtual server (1), reselecting main virtual server (1) by Paxos algorithm, new main virtual server (1) notice Virtual Machine Manager module (2) reclaims old main virtual server (1).
The big data distributing method of facing multiple users group the most according to claim 3, it is characterized in that, described Virtual Machine Manager module (2) uses the virtual server (1) that ltsh chain table structure organization is idle, and the Key of ltsh chain table is the ID of the physical server (6) running virtual server (1);Also include step in detail below: above-mentioned steps 4) in, when virtual server (1) is recycled to system free virtual server pools (4), this virtual server (1) is put in ltsh chain table by Virtual Machine Manager module (2) according to the ID of the physical server (6) at this virtual server (1) place;Above-mentioned steps 5) in, when idle virtual server (1) removal system free virtual server pools (4), preferential searching system free virtual server pools (4) of Virtual Machine Manager module (2), the virtual server (1) running on Same Physical server (6) with the virtual server (1) in destination virtual server cluster (3) is allocated.
7. according to the big data distributing method of the arbitrary described facing multiple users group of claim 4,5,6, it is characterized in that, also include peer-to-peer message distribution platform, the loading statistics of all virtual servers (1) in synchronized clusters, the form that described peer-to-peer message distribution platform is broadcasted by message virtual server (1) in cluster sends loading statistics.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310383301.2A CN103428008B (en) | 2013-08-28 | 2013-08-28 | The big data distributing method of facing multiple users group |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310383301.2A CN103428008B (en) | 2013-08-28 | 2013-08-28 | The big data distributing method of facing multiple users group |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103428008A CN103428008A (en) | 2013-12-04 |
CN103428008B true CN103428008B (en) | 2016-08-10 |
Family
ID=49652214
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310383301.2A Active CN103428008B (en) | 2013-08-28 | 2013-08-28 | The big data distributing method of facing multiple users group |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103428008B (en) |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104168310A (en) * | 2014-07-24 | 2014-11-26 | 深圳天珑无线科技有限公司 | Method and apparatus for scheduling cloud end vertual server |
CN104092782A (en) * | 2014-07-31 | 2014-10-08 | 武汉云雅科技有限公司 | Intelligent cloud cluster regulating and controlling method and system based on cloud computing |
CN105100271A (en) * | 2015-08-31 | 2015-11-25 | 南京势行软件开发有限公司 | System for publishing distributed flexible extension information and control method thereof |
CN107404511B (en) * | 2016-03-31 | 2020-11-06 | 阿里巴巴集团控股有限公司 | Method and device for replacing servers in cluster |
CN106603631B (en) * | 2016-11-17 | 2020-07-03 | 北京奇艺世纪科技有限公司 | Distributed message system and capacity expansion method |
CN106330576B (en) * | 2016-11-18 | 2019-10-25 | 北京红马传媒文化发展有限公司 | The method of containerization micro services automatic telescopic and migration scheduling, system and equipment |
CN108462737B (en) * | 2018-01-29 | 2021-02-02 | 哈尔滨工业大学深圳研究生院 | Batch processing and pipeline-based hierarchical data consistency protocol optimization method |
US11102060B2 (en) * | 2018-01-31 | 2021-08-24 | Hewlett Packard Enterprise Development Lp | Identification of a soft failure at a member |
CN110032438B (en) * | 2019-04-24 | 2021-11-26 | 北京高途云集教育科技有限公司 | Delayed task execution method and device and electronic equipment |
CN110633182B (en) * | 2019-07-09 | 2023-10-27 | 北京有竹居网络技术有限公司 | System, method and device for monitoring server stability |
CN110660389A (en) * | 2019-09-11 | 2020-01-07 | 北京小米移动软件有限公司 | Voice response method, device, system and equipment |
CN110650217B (en) * | 2019-11-07 | 2021-01-05 | 江苏康融科技有限公司 | Load balancing method for download server |
CN111935306B (en) * | 2020-09-11 | 2020-12-29 | 腾讯科技(深圳)有限公司 | Node scheduling method and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102035662A (en) * | 2009-09-27 | 2011-04-27 | 中国移动通信集团公司 | Virtual server management system, method and device |
CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
CN102130938A (en) * | 2010-12-03 | 2011-07-20 | 中国科学院软件研究所 | Resource supply method oriented to Web application host platform |
CN102891896A (en) * | 2012-10-18 | 2013-01-23 | 江苏物联网研究发展中心 | Cloud computing infrastructure system for mass storage of Internet of things |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5378946B2 (en) * | 2009-10-26 | 2013-12-25 | 株式会社日立製作所 | Server management apparatus and server management method |
-
2013
- 2013-08-28 CN CN201310383301.2A patent/CN103428008B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102035662A (en) * | 2009-09-27 | 2011-04-27 | 中国移动通信集团公司 | Virtual server management system, method and device |
CN102130938A (en) * | 2010-12-03 | 2011-07-20 | 中国科学院软件研究所 | Resource supply method oriented to Web application host platform |
CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
CN102891896A (en) * | 2012-10-18 | 2013-01-23 | 江苏物联网研究发展中心 | Cloud computing infrastructure system for mass storage of Internet of things |
Also Published As
Publication number | Publication date |
---|---|
CN103428008A (en) | 2013-12-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103428008B (en) | The big data distributing method of facing multiple users group | |
CN102724277B (en) | The method of live migration of virtual machine and deployment, server and group system | |
CN104102543B (en) | The method and apparatus of adjustment of load in a kind of cloud computing environment | |
CN102681895B (en) | Dynamic self-migrating cloud service method | |
CN104461740A (en) | Cross-domain colony computing resource gathering and distributing method | |
CN104657215A (en) | Virtualization energy-saving system in Cloud computing | |
TW201535266A (en) | Resource adjustment methods and systems for virtual machines | |
CN102111337A (en) | Method and system for task scheduling | |
CN103684916A (en) | Method and system for intelligent monitoring and analyzing under cloud computing | |
CN104038540A (en) | Method and system for automatically selecting application proxy server | |
CN105516347A (en) | Method and device for load balance allocation of streaming media server | |
CN109088941A (en) | A method of based on intelligent scheduling cloud resource under ddos attack | |
WO2023109068A1 (en) | Automatic virtual machine migration decision-making method based on user experience in multi-cloud environment | |
CN106131227A (en) | Balancing method of loads, meta data server system and load balance system | |
CN105607943A (en) | Dynamic deployment mechanism of virtual machine under cloud environment | |
CN107295358B (en) | A kind of 3D Streaming Media storage method under cloud environment | |
CN106681839A (en) | Elasticity calculation dynamic allocation method | |
CN102480502A (en) | I/O load equilibrium method and I/O server | |
Fu et al. | Network traffic based virtual machine migration in cloud computing environment | |
CN103297511B (en) | The dispatching method of the client/server under high dynamic environments and system | |
CN106936926A (en) | Access the method and system of back end | |
CN108306912A (en) | Virtual network function management method and its device, network function virtualization system | |
Charapko et al. | Adapting to access locality via live data migration in globally distributed datastores | |
CN109120424B (en) | Bandwidth scheduling method and device | |
CN107203256A (en) | Energy-conservation distribution method and device under a kind of network function virtualization scene |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |