CN103428008A - Big data distribution strategy oriented to multiple user groups - Google Patents
Big data distribution strategy oriented to multiple user groups Download PDFInfo
- Publication number
- CN103428008A CN103428008A CN2013103833012A CN201310383301A CN103428008A CN 103428008 A CN103428008 A CN 103428008A CN 2013103833012 A CN2013103833012 A CN 2013103833012A CN 201310383301 A CN201310383301 A CN 201310383301A CN 103428008 A CN103428008 A CN 103428008A
- Authority
- CN
- China
- Prior art keywords
- virtual server
- virtual
- server
- cluster
- load
- 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
- 238000012544 monitoring process Methods 0.000 claims abstract description 29
- 230000004044 response Effects 0.000 claims abstract description 5
- 238000001514 detection method Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 10
- 230000001360 synchronised effect Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 230000008901 benefit Effects 0.000 abstract description 3
- 230000001105 regulatory effect Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000011084 recovery Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000000275 quality assurance Methods 0.000 description 1
Images
Landscapes
- Computer And Data Communications (AREA)
Abstract
The invention relates to a big data distribution strategy oriented to multiple user groups. The big data distribution strategy relates to a plurality of virtual servers and a virtual machine management module, wherein the virtual servers operate on a physical server and constitute a virtual server cluster, main virtual servers are arranged in the virtual server cluster, the main virtual servers are selected from the virtual servers in the Paxos algorithm, the virtual server cluster is further provided with monitoring modules, and the monitoring modules are used for monitoring load conditions of the virtual servers in the virtual server cluster and adding the virtual servers into the cluster or removing the virtual servers from the cluster at any time. The big data distribution strategy has the advantages of having concise steps and few calculation steps, completely solving the problems that load balance cannot be regulated accurately because control granularity is too coarse during response to big data distribution, and having good application value.
Description
Technical Field
The invention relates to a big data distribution strategy facing a multi-user group.
Background
With the rapid development of the Internet and the continuous increase of the business volume, the data access flow based on the network is rapidly increased, especially the access to data centers, large-scale enterprises, portal websites and the like. Meanwhile, the server website provides more and more abundant content and information for visitors by means of application programs such as HTTP and FTP, and the server is gradually submerged by data.
In a computer network for transmitting massive large data, compared with the development of network technology, the increase of the processing speed and the memory access speed of a server is far lower than the increase of network bandwidth and application service, and the increase of the number of users brought by the increase of the network bandwidth also causes the resource consumption of the server to be serious, so that the server becomes a network bottleneck. The traditional single machine mode also tends to become a network failure point.
All this puts demands on high performance and high reliability of the application services. The two most important techniques are load balancing and virtualization.
Load balancing provides a cheap, effective and transparent method to expand the bandwidth of network devices and servers, increase throughput, enhance network data processing capacity, and improve network flexibility and availability.
Virtualization technology can enlarge the capacity of hardware and simplify the reconfiguration process of software. With the wide deployment of multi-core systems, clusters, grids and even cloud computing in recent years, the advantages of virtualization technology in commercial application are increasingly embodied, the IT cost is reduced, and the system safety and reliability are enhanced.
In a system facing various user groups, the resource requirements of different user groups on a server are different, and the requirements are dynamically changed, and in the prior art, the conditions that some user group systems are overloaded or some system computing resources are idle often exist, so that the user requirements cannot be quickly reacted and adjusted, the actual effect of the load balancing technology is reduced, and therefore a novel data distribution method capable of making up the existing load balancing technology is necessary to be developed.
Disclosure of Invention
Aiming at the defect that the prior art cannot adjust the dynamic change of resource requirements according to different users in real time, so that the use efficiency of the system is low, the invention provides a novel big data distribution strategy facing a multi-user group.
In order to achieve the purpose, the invention can adopt the following technical scheme:
the big data distribution strategy facing to the multi-user group comprises a plurality of virtual servers and a virtual machine management module, wherein the virtual servers run on a physical server, the virtual servers form a virtual server cluster, a main virtual server is arranged in the virtual server cluster, and the main virtual server is selected from the virtual servers through a Paxos algorithm; the method comprises the following specific steps:
1) running a plurality of virtual servers on a physical server to form a virtual server cluster;
2) the virtual servers generate load statistical data of the virtual servers, and the load statistical data of all the virtual servers are synchronized within the range of the virtual server cluster;
3) the virtual server cluster monitors the load statistical data of the virtual servers in the cluster;
4) in the step 3), if the number of the idle virtual servers in the virtual server clusterAnd if the quantity exceeds the threshold value, randomly selecting an idle virtual server, deleting the virtual server from the virtual server cluster, and informing the virtual machine management module to recycle the deleted virtual server into a system idle virtual server pool, wherein the utilization rate of the idle virtual server is utl =0, and the utilization rate of the idle virtual server is utl =0Wherein, Load is the system Load of the virtual server, and Capacity is the processing Capacity of the virtual server;
5) in the step 4), if the utilization ratio utl-Group of the virtual servers in the virtual server cluster is greater than the threshold value and the system idle virtual server pool is not empty, the virtual server in the system idle virtual server pool is moved out to the virtual server cluster, and the utilization ratio is higher than the threshold value
Wherein,for single virtual server utilization, LoadiCapacity is the system load of the virtual serveriG is the number of virtual servers within the virtual server cluster;
6) when the virtual server cluster receives a data distribution request of a user, the virtual server cluster selects the virtual server with the lowest load according to the utilization rate utl of the virtual server obtained in the step 4) and sends the data distribution request to the virtual server with the lowest load for processing.
Preferably, a plurality of virtual server clusters are included, and the virtual servers in each virtual server cluster are located on the same physical server.
Preferably, the system further comprises a monitoring module, and in the step 3), the monitoring module monitors the load statistical data of the virtual server and submits the load statistical data to the virtual machine management module.
Preferably, the monitoring module is arranged in the virtual server; the method also comprises the following specific steps: in the step 3), the monitoring module synchronously transmits the monitored load statistical data among the virtual servers in the virtual server cluster, and the monitoring modules perform online detection with each other, wherein the online detection is that in a detection period of 1s, the monitoring module sends a connection request to a non-idle virtual server, and if no response is obtained in three consecutive detection periods, the virtual server loses network connection; otherwise, synchronous transmission of the load statistical data is carried out with the non-idle virtual server.
Preferably, the method also comprises the following specific steps: and when the monitoring module loses network connection with the main virtual server, the main virtual server is reselected through a Paxos algorithm, and the new main virtual server informs the virtual machine management module to recover the old main virtual server.
Preferably, the virtual machine management module organizes an idle virtual server by adopting a hash chain table structure, and a Key of the hash chain table is an ID of a physical server running the virtual server; the method also comprises the following specific steps: in the step 4), when the virtual server is recovered to the system idle virtual server pool, the virtual machine management module puts the virtual server into the hash chain table according to the ID of the physical server where the virtual server is located; in the step 5), when the idle virtual server moves out of the system idle virtual server pool, the virtual machine management module preferentially searches for the virtual server of the system idle virtual server pool, and allocates the virtual server running in the same physical server as the virtual server in the target virtual server cluster.
Preferably, the system further comprises a peer-to-peer message distribution platform for synchronizing the load statistics data of all the virtual servers in the cluster, and the peer-to-peer message distribution platform sends the load statistics data to the virtual servers in the cluster in a message broadcasting mode.
Due to the adoption of the technical scheme, the invention has the remarkable technical effects that:
the invention makes up the deficiency of load balancing among users based on dynamic adjustment of the virtual server, and meanwhile, can make dynamic and timely feedback to the user request by monitoring the load statistical data, and compared with the traditional quality assurance service technology which only aims at the service of certain physical equipment, the invention has finer control granularity and can better meet the specific requirements of the current data distribution service. And secondly, the virtual machine resources can be managed more effectively through a virtual machine allocation and recovery mechanism, the virtual machine allocation and recovery mechanism is established on the basis of synchronous transmission of load statistical data, the response to the load required by a user request is quicker and more accurate, and compared with the prior art, the virtual machine allocation and recovery mechanism has higher flexibility, more effectively utilizes server resources and can reduce the power consumption of a data center.
Furthermore, by means of the Hash chain table technology, and the Key of the chain table is associated with the ID of the physical server, the virtual servers running on the same physical server can be classified into the same cluster when the virtual servers are distributed, and the efficiency of data migration and the resource utilization efficiency are improved.
In order to more accurately monitor the load statistical data, each virtual server is provided with a monitoring module, the monitoring modules synchronize the load statistical data with each other, and meanwhile, the virtual servers and the main virtual server are subjected to online detection, so that the influence of server faults on the whole resources of the system is reduced, and the stability of the system is improved.
In addition, the monitoring module also carries out data synchronization through a peer-to-peer message distribution platform, so that the synchronization rate is further improved, and the accuracy of virtual server distribution is improved.
Drawings
Fig. 1 is a schematic main flow diagram of the present invention.
Fig. 2 is a schematic diagram of the system architecture of the present invention.
Fig. 3 is a schematic diagram of a data structure in the virtual machine management module.
Detailed Description
The present invention will be described in further detail with reference to examples.
Example 1
The system architecture diagram of the big data distribution strategy facing a multi-user group is shown in fig. 2, and a virtual machine management module 2 of the system is responsible for managing allocation and recovery of virtual servers 1 and for transmitting load statistical data between virtual server clusters 3. The system divides the virtual server 1 into a plurality of virtual server groups according to the application requirements of different user groups, wherein the virtual server Group is used as a virtual server cluster 3 and comprises Group1 and Group 2. Each virtual server group has a certain number of virtual servers 1, and a master virtual server 1 generated based on Paxos algorithm. Group1 has VS 1-VS 4, 4 virtual servers 1, and one master virtual server 1. Each virtual server 1 in the group has a monitoring module 5 to load synchronize the load statistics in the group and maintain online detection between the virtual servers 1.
As shown in fig. 3, the virtual machine management module 2 maintains the free virtual server 1 through a hash chain table data structure. Each virtual server 1 performs hashing according to the ID number of the physical server to which it belongs. The advantage of this is that when allocating the virtual server 1, it can be ensured that the virtual server 1 used by the same user group application is concentrated in a small number of physical servers as much as possible, so that the network traffic of the intra-group communication can be reduced, and the lookup efficiency is greatly improved compared with that of a common linked list structure through the hash table.
On the basis of the architecture of the above device, the big data distribution strategy facing multi-user group of the present invention, as shown in fig. 1, includes the following steps:
each physical server node is divided into a plurality of virtual servers 1, and a group of virtual servers 1 form a cluster which is responsible for processing the requests of a class of user group distribution tasks;
each virtual server 1 has certain resources allocated and if not specified by the user, default values for the system are used. In a group of virtual server clusters 3 that are oriented to a particular group of users, there is a primary virtual server 1, and when such users send requests, the requests first pass through the primary virtual server 1. The primary virtual server 1 is generated by negotiation election between the virtual servers 1 in the cluster, and after the primary virtual server 1 is selected, the virtual server 1 provides external services.
The other virtual servers 1 in the cluster determine whether the main virtual server 1 works normally through online detection. Once the main virtual server 1 goes wrong or goes offline, a new main virtual server 1 is renegotiated and elected through the Paxos algorithm, and then the new main virtual server 1 notifies the virtual machine management module 2 to recover the virtual server 1 with the error.
Each virtual server 1 maintains its own load statistics. In each group of virtual server clusters 3, synchronizing the load statistical data of all the virtual servers 1 in the cluster group through a peer-to-peer message distribution platform;
let the Load of the virtual server A be LoadAThe system processing Capacity is CapacityAThen the utilization of virtual server A utlAComprises the following steps: utlA=LoadA/CapacityA. The utilization rate of the virtual servers in the group is as follows:
wherein G is the number of members in the group.
The main work of the monitoring program is as follows: each group of virtual server clusters 3 is provided with a monitoring module 5 for monitoring the load of each virtual server 1 in each group of virtual server clusters 3; utl-Group data in the Group are collected regularly and reported to the system virtual machine management module 2; if any member in the group loses the connection with the monitoring module 5, the virtual machine management module 2 is informed to reclaim the resources of the virtual server 1.
If the monitoring module 5 detects that the number of idle servers in the cluster exceeds a certain threshold value, wherein the threshold value is a preset value, an idle server is randomly selected in the cluster, then the server is deleted from the cluster, and the virtual machine management module 2 is informed to recycle the virtual server 1 to a system idle virtual server pool 4; where a free virtual server 1 means that no service is currently provided to the user, i.e. utl = 0%.
If the monitoring module 5 detects that the sum of the loads of all the virtual servers 1 in the cluster is greater than a threshold value, the threshold value is a set value, and the system idle virtual server pool 4 is not empty, applying for a virtual server 1 to be added into the cluster from the system idle virtual server pool 4;
in the virtual machine management module 2, the idle virtual machine servers 1 are organized by a hash chain table data structure. The Key in the structure of the hash table is the ID of the physical server to which the virtual machine server 1 belongs, that is, the virtual machine management module 2 organizes and manages the hash chain table of the virtual server 1 according to the physical server of the virtual server 1.
When the user releases the virtual machine server 1, the virtual machine management module 2 puts the virtual server 1 into the hash table according to the physical server to which the recovered virtual server 1 belongs.
When a certain virtual server cluster 3 makes an application request, the virtual machine management module 2 first searches for a free virtual server 1 having the same physical server as the virtual server 1 in the cluster to allocate.
The purpose of this organization is to centralize each virtual server 1 in the virtual server cluster 3 in several physical servers, so that communication between the virtual servers 1 within the cluster can be faster and resources of the physical servers can be saved.
When processing the request of a class of users, a group of virtual server clusters 3 selects the virtual server 1 with the lowest load to process the request according to the load statistical data of each virtual server 1 in the group;
when a user makes a request for a server, the request firstly reaches a main virtual server 1 in a virtual server cluster 3, and the main virtual server 1 selects one virtual server 1 to provide service for the user according to the load of each virtual server 1 in a group.
In the virtual machine management module 2, the algorithm for the main virtual server 1 to select the virtual server 1 in the cluster to provide service for the user is as follows: within the virtual server cluster 3, the virtual servers 1 are divided into two groups according to the current load of the virtual servers 1, the virtual servers 1 in the first group are currently providing services, and the load has reached a threshold set by the user. The load of the virtual servers 1 of the second group does not reach the threshold set by the user. When the user request arrives, the main virtual server 1 selects one virtual server 1 with the largest load from the second group to provide service for the user. In this way, the request processing can be centralized as much as possible, so that the number of the virtual servers 1 in the virtual server cluster 3 is as small as possible, and the virtual machine management module 2 can have more idle virtual servers 1.
In a specific application example, the period of online detection of the monitoring module 5 is 1s, and if no response is received in the online detection request connection for three consecutive times, the virtual server 1 is considered to lose network connection. Initially, the number of virtual servers 1 of one virtual server cluster 3 is 5, and if the number of the virtual machine servers 1 which are not idle currently accounts for more than 80% of the number of the virtual servers 1 in the cluster, new virtual server 1 resources are applied to the system virtual machine management module 2. And if the number of the idle virtual servers 1 in the current virtual server cluster 3 exceeds 40% of the total number of the virtual servers 1 in the cluster, informing the system virtual machine management module 2 to recover the idle virtual servers 1. The monitoring module 5 synchronizes the load statistics within the cluster every 3 seconds, i.e. 3 detection periods.
In summary, the above-mentioned embodiments are only preferred embodiments of the present invention, and all equivalent changes and modifications made in the claims of the present invention should be covered by the claims of the present invention.
Claims (7)
1. A big data distribution strategy facing a multi-user group is characterized by comprising a plurality of virtual servers (1) running on a physical server (6) and a virtual machine management module (2), wherein the virtual servers (1) form a virtual server cluster (3), a main virtual server (1) is arranged in the virtual server cluster (3), and the main virtual server (1) is selected from the virtual servers (1) through a Paxos algorithm; the method comprises the following specific steps:
1) running a plurality of virtual servers (1) on a physical server (6) to form a virtual server cluster (3);
2) the virtual server (1) generates load statistical data of the virtual server, and synchronizes the load statistical data of all the virtual servers (1) in the range of the virtual server cluster (3);
3) the virtual server cluster (3) monitors the load statistical data of the virtual servers (1) in the cluster;
4) in the step 3), if the number of the virtual servers (1) which are idle in the virtual server cluster (3) exceeds a threshold value, an idle virtual server (1) is randomly selected, the virtual server (1) is deleted from the virtual server cluster (3), and the virtual machine management module (2) is notified to recycle the deleted virtual server (1) into the system idle virtual server pool (4), wherein the utilization rate utl =0 of the idle virtual server (1) is obtained, and the utilization rate is utl =0Wherein, Load is the system Load of the virtual server (1), and Capacity is the processing Capacity of the virtual server (1);
5) in the step 4), if the utilization ratio utl-Group of the virtual server (1) in the virtual server cluster (3) is greater than the threshold value and the system idle virtual server pool (4) is not empty, the virtual server (1) in one system idle virtual server pool (4) is moved out to the virtual server cluster (3), and the utilization ratio is Wherein,
is a singleUtilization of virtual Server (1), LoadiCapacity is the system load of the virtual server (1)iG is the number of virtual servers (1) within the virtual server cluster (3);
6) when the virtual server cluster (3) receives a data distribution request of a user, the virtual server (1) with the lowest load is selected according to the utilization rate utl of the virtual server (1) obtained in the step 4), and the data distribution request is sent to the virtual server (1) with the lowest load for processing.
2. A big data distribution strategy towards multi-user group according to claim 1, characterized by comprising multiple virtual server clusters (3), the virtual servers (1) within each virtual server cluster (3) are located on the same physical server (6).
3. The big data distribution strategy facing multi-user group according to claim 1, further comprising a monitoring module (5), wherein in step 3), the monitoring module (5) monitors the load statistics data of the virtual server (1) and submits the load statistics data to the virtual machine management module (2).
4. Big data distribution strategy towards multi-user group according to claim 3, characterized in that the monitoring module (5) is located within the virtual server (1); the method also comprises the following specific steps: in the step 3), the monitoring module (5) synchronously transmits the monitored load statistical data among the virtual servers (1) in the virtual server cluster (3), and the monitoring modules (5) perform online detection with each other, wherein the online detection is that in a detection period of 1s, the monitoring module (5) sends a connection request to the non-idle virtual server (1), and if no response is obtained in three consecutive detection periods, the virtual server (1) loses network connection; otherwise, synchronous transmission of load statistical data is carried out with the non-idle virtual server (1).
5. The big data distribution strategy facing the multi-user group according to claim 4, further comprising the following specific steps: when the monitoring module (5) loses network connection with the main virtual server (1), the main virtual server (1) is reselected through a Paxos algorithm, and the new main virtual server (1) informs the virtual machine management module (2) to recycle the old main virtual server (1).
6. The big data distribution strategy facing the multi-user group according to claim 3, wherein the virtual machine management module (2) organizes the idle virtual servers (1) by adopting a hash chain table structure, and the Key of the hash chain table is the ID of the physical server (6) running the virtual server (1); the method also comprises the following specific steps: in the step 4), when the virtual server (1) is recycled to the system idle virtual server pool (4), the virtual machine management module (2) puts the virtual server (1) into the hash chain table according to the ID of the physical server (6) where the virtual server (1) is located; in the step 5), when the idle virtual server (1) moves out of the system idle virtual server pool (4), the virtual machine management module (2) preferentially searches the system idle virtual server pool (4) and allocates the virtual server (1) running in the same physical server (6) as the virtual server (1) in the target virtual server cluster (3).
7. The big data distribution strategy facing multiple user groups according to any one of claims 4, 5 and 6, further comprising a peer-to-peer message distribution platform for synchronizing the load statistics of all virtual servers (1) in the cluster, wherein the peer-to-peer message distribution platform sends the load statistics to the virtual servers (1) in the cluster by message broadcasting.
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 true CN103428008A (en) | 2013-12-04 |
| CN103428008B 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) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| 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 |
| CN106330576A (en) * | 2016-11-18 | 2017-01-11 | 北京红马传媒文化发展有限公司 | Automatic scaling and migration scheduling method, system and device for containerization micro-service |
| CN106603631A (en) * | 2016-11-17 | 2017-04-26 | 北京奇艺世纪科技有限公司 | Distributed message system and expansion method thereof |
| CN107404511A (en) * | 2016-03-31 | 2017-11-28 | 阿里巴巴集团控股有限公司 | The replacement method and equipment of server in cluster |
| CN108462737A (en) * | 2018-01-29 | 2018-08-28 | 哈尔滨工业大学深圳研究生院 | Individual-layer data consistency protocol optimization method based on batch processing and assembly line |
| CN110032438A (en) * | 2019-04-24 | 2019-07-19 | 北京高途云集教育科技有限公司 | Delayed tasks execute method, apparatus and electronic equipment |
| CN110096407A (en) * | 2018-01-31 | 2019-08-06 | 慧与发展有限责任合伙企业 | Identify the soft fault at member |
| CN110633182A (en) * | 2019-07-09 | 2019-12-31 | 北京无限光场科技有限公司 | System, method and apparatus for monitoring server stability |
| CN110650217A (en) * | 2019-11-07 | 2020-01-03 | 黄颖 | Load balancing method for download server |
| 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 |
| CN111935306B (en) * | 2020-09-11 | 2020-12-29 | 腾讯科技（深圳）有限公司 | Node scheduling method and device |
| CN118316887A (en) * | 2024-06-07 | 2024-07-09 | 湖南云客信息技术有限公司 | Flow surge elimination method and device for mass-sending information |
Citations (5)
| 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 |
| US20110099403A1 (en) * | 2009-10-26 | 2011-04-28 | Hitachi, Ltd. | Server management apparatus and server management method |
| 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 |
-
2013
- 2013-08-28 CN CN201310383301.2A patent/CN103428008B/en active Active
Patent Citations (5)
| 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 |
| US20110099403A1 (en) * | 2009-10-26 | 2011-04-28 | Hitachi, Ltd. | Server management apparatus and server management method |
| 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 |
Cited By (21)
| 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 |
| WO2016011826A1 (en) * | 2014-07-24 | 2016-01-28 | 深圳天珑无线科技有限公司 | Cloud virtual server scheduling method and device |
| 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 |
| CN107404511A (en) * | 2016-03-31 | 2017-11-28 | 阿里巴巴集团控股有限公司 | The replacement method and equipment of server in cluster |
| CN106603631A (en) * | 2016-11-17 | 2017-04-26 | 北京奇艺世纪科技有限公司 | Distributed message system and expansion method thereof |
| 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 |
| CN106330576A (en) * | 2016-11-18 | 2017-01-11 | 北京红马传媒文化发展有限公司 | Automatic scaling and migration scheduling method, system and device for containerization micro-service |
| CN108462737A (en) * | 2018-01-29 | 2018-08-28 | 哈尔滨工业大学深圳研究生院 | Individual-layer data consistency protocol optimization method based on batch processing and assembly line |
| CN110096407B (en) * | 2018-01-31 | 2023-10-13 | 慧与发展有限责任合伙企业 | Identifying soft faults at members |
| CN110096407A (en) * | 2018-01-31 | 2019-08-06 | 慧与发展有限责任合伙企业 | Identify the soft fault at member |
| CN110032438A (en) * | 2019-04-24 | 2019-07-19 | 北京高途云集教育科技有限公司 | Delayed tasks execute method, apparatus and electronic equipment |
| CN110633182A (en) * | 2019-07-09 | 2019-12-31 | 北京无限光场科技有限公司 | System, method and apparatus for monitoring server stability |
| 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 |
| CN110650217A (en) * | 2019-11-07 | 2020-01-03 | 黄颖 | Load balancing method for download server |
| 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 |
| CN118316887A (en) * | 2024-06-07 | 2024-07-09 | 湖南云客信息技术有限公司 | Flow surge elimination method and device for mass-sending information |
| CN118316887B (en) * | 2024-06-07 | 2024-09-17 | 湖南云客信息技术有限公司 | Flow surge elimination method and device for mass-sending information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103428008B (en) | 2016-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103428008B (en) | The big data distributing method of facing multiple users group | |
| CN105491138B (en) | Distributed load scheduling method based on load rate graded triggering | |
| KR20190038760A (en) | Load balancing optimization method and device based on cloud monitoring | |
| CN103812949B (en) | A kind of task scheduling towards real-time cloud platform and resource allocation methods and system | |
| CN103401947A (en) | Method and device for allocating tasks to multiple servers | |
| CN103259832A (en) | Cluster resource control method for achieving dynamic load balance, fault diagnosis and failover | |
| CN104836819A (en) | Dynamic load balancing method and system, and monitoring and dispatching device | |
| WO2013163865A1 (en) | Virtual machine hot migration and deployment method, server and cluster system | |
| WO2014024863A1 (en) | Load distribution method taking into account each node in multi-level hierarchy | |
| CN104038540A (en) | Method and system for automatically selecting application proxy server | |
| CN105242956A (en) | Virtual function service chain deployment system and deployment method therefor | |
| CN102694868A (en) | Cluster system implementation and task dynamic distribution method | |
| WO2023109068A1 (en) | Automatic virtual machine migration decision-making method based on user experience in multi-cloud environment | |
| US20120233313A1 (en) | Shared scaling server system | |
| CN105979273A (en) | Cloud monitor and cloud operation of intelligent commercial TVs based on big data and cloud computation | |
| CN108282526B (en) | Dynamic allocation method and system for servers between double clusters | |
| CN104092620A (en) | Method and device for achieving adjustment of network bandwidth | |
| CN105553872A (en) | Multipath data traffic load equalizing method | |
| CN108881379B (en) | Method and device for data synchronization between server clusters | |
| CN111858033A (en) | Load balancing method based on cluster and multiple processes | |
| CN103957251A (en) | Method and system for achieving server load balancing | |
| CN103631804B (en) | Electronic map cuts drawing method and processing system | |
| CN111418187A (en) | Scalable statistics and analysis mechanism in cloud networks | |
| CN114168312A (en) | Distributed cluster load balancing method and device and storage medium | |
| US11531572B2 (en) | Cross-cluster host reassignment |
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 |