CN201260187Y - Network topology structure for distributed dynamic load balance - Google Patents
Network topology structure for distributed dynamic load balance Download PDFInfo
- Publication number
- CN201260187Y CN201260187Y CNU2008200995467U CN200820099546U CN201260187Y CN 201260187 Y CN201260187 Y CN 201260187Y CN U2008200995467 U CNU2008200995467 U CN U2008200995467U CN 200820099546 U CN200820099546 U CN 200820099546U CN 201260187 Y CN201260187 Y CN 201260187Y
- Authority
- CN
- China
- Prior art keywords
- network
- information
- sub
- node
- topology structure
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Computer And Data Communications (AREA)
Abstract
The utility model discloses a distributed dynamic duty equalizer network topology structure. The network is formed by connecting at least one subsidiary network, and each subsidiary network comprises at least one information node. The utility model is characterized in that each subsidiary network also comprises one information center, each information node in each subsidiary network is connected with the information center to form a first closed hoop, and all information centers in the network are connected to form a second closed hoop. The topology structure not only realizes the efficient duty balance in the network, but also realizes the efficient duty balance of the integral network.
Description
Technical field
The utility model belongs to computer network architecture, relates in particular to a kind of distributed dynamic load balancing network topology structure.
Background technology
Distributed Computer System is by passing through network or communication line couples together and the computer of separate cooperation is again formed computer system more than one, the computer of forming this computer system is called information node, it also can be other-end equipment in a broad sense, as scanner, printer, work station, plotter, data processor etc.Distributed Calculation is a kind of computation schema that transmits based on message, has characteristics such as small investment, flexibility and reliability, can provide a kind of to the user and realize the effective way of high performance parallel computation with less cost, so be widely used in the industry-by-industry.
In the distributed system of a job, the load of information node is dynamic change, because each node processing ability there are differences, after system moves a period of time, the task that some node distributes a lot (being called heavy duty), and the other node is free time or task less (being called zero load or underloading), thereby cause the Processing tasks of each node seriously uneven, reduced the operational efficiency of system, even can cause the collapse of some node, take place for fear of this phenomenon, must take the load balancing measure.
Load balancing is that the task of distributing to each node is rescheduled, and by process migration, makes each node load roughly the same, can improve the operating efficiency and the performance of distributed parallel system so greatly.But present research mostly concentrates on balance policy in theory, to the research of implementation model seldom, and especially for the complicated balanced less discussion of being adopted of network topology structure.
The utility model content
The utility model purpose provides a kind of distributed dynamic load balancing network topology structure, by improving network topology structure, realizes each node high-efficient carrier equilibrium in the distributed network.
For achieving the above object, the utility model is explained a kind of distributed dynamic load balancing network topology structure, described network is formed by connecting by at least one sub-network, in each sub-network, include at least one information node, its key is: also include an information centre in each sub-network, information node in each sub-network and information centre connect into first closed hoop, and the information centre in the all-ones subnet network connects into second closed hoop.
Described information node is meant and carries out the information process unit that job task is handled, and computer normally in network broadly can be an other-end, as scanner, printer, work station, plotter, data processor etc.Information centre is the unit that writes down load message and schedule job Processing tasks in the sub-network.First closed hoop constitutes sub-network, and the division foundation of sub-network and scale are in the light of actual conditions and real needs.Each information centre realizes information exchange by second closed hoop.
In whole network, each sub-network couples together by information centre and other sub-networks, carry out information exchange, and any information node of each sub-network inside can not be connected with other sub-networks directly all, must be through the information centre of this sub-network.Information centre dynamically puts down in writing the loading condition and the relevant position of each information node in the sub-network of place, information centre with other sub-networks carries out information exchange simultaneously, each information node only need be known the information centre position, reports load information to information centre in real time and gets final product.
When certain information node load was heavier in the sub-network, information centre just gave the Task Distribution of this information node the lighter information node of load in the sub-network, realizes the load balancing of sub-network inside; When all information nodes of certain sub-network are all busy or not busy, can be through the information exchange of information centre, give other sub-networks with the load Task Distribution or receive the load task of other sub-networks, realize the load balancing of whole network.
Remarkable result of the present utility model is: a kind of distributed dynamic load balancing network topology structure is provided, change network topology structure into twin nuclei, not only can realize the load balancing in the sub-network, can also realize the load balancing of whole network, improve whole network operating efficiency.
Description of drawings
Fig. 1 is a structure principle chart of the present utility model.
Embodiment
Below in conjunction with embodiment the utility model is described further.
Embodiment 1:
As shown in Figure 1: plant distributed dynamic load balancing network topology structure, described network is formed by connecting by at least one sub-network, in each sub-network, include an information centre 2 and at least one information node 1, information node 1 in each sub-network and information centre 2 connect into first closed hoop 3, and the information centre 2 in the all-ones subnet network connects into second closed hoop 4.
Described information node 1 is meant and carries out the information process unit that job task is handled, and computer normally in network broadly can make other-end, as scanner, printer, work station, plotter, data processor etc.Information centre 2 is the unit that write down load message and schedule job Processing tasks in the sub-network.First closed 3 rings constitute sub-network, and each information centre 2 realizes information exchange by second closed hoop 4.
Its operation principle is as follows:
Information node 1 is when the processing operation task, in real time the load disposition is sent to the information centre 2 of place sub-network, information centre 2 is according to the loading condition of each information node 1, the Task Distribution of the information node 1 that load is heavier is given the lighter or idle information node 1 of load, realizes the load balancing of sub-network inside.
Each information centre 2 also realizes information exchange by second closed hoop 4 simultaneously, when the sub-network load at certain information centre 2 place is heavier or lighter, by information centre 2, just can give the lighter sub-network of other loads with the Task Distribution of sub-network, perhaps receive the task of other sub-networks, realize the load balancing of whole network.
The number of sub-network and divide foundation and scale in the light of actual conditions and real needs, be not limited to given example in the present embodiment.
Claims (1)
1, a kind of distributed dynamic load balancing network topology structure, described network is formed by connecting by at least one sub-network, in each sub-network, include at least one information node (1), it is characterized in that: in each sub-network, also include an information centre (2), information node in each sub-network (1) and information centre (2) connect into first closed hoop (3), and the information centre in the all-ones subnet network (2) connects into second closed hoop (4).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008200995467U CN201260187Y (en) | 2008-07-25 | 2008-07-25 | Network topology structure for distributed dynamic load balance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008200995467U CN201260187Y (en) | 2008-07-25 | 2008-07-25 | Network topology structure for distributed dynamic load balance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201260187Y true CN201260187Y (en) | 2009-06-17 |
Family
ID=40774426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNU2008200995467U Expired - Fee Related CN201260187Y (en) | 2008-07-25 | 2008-07-25 | Network topology structure for distributed dynamic load balance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201260187Y (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017067449A1 (en) * | 2015-10-19 | 2017-04-27 | 中兴通讯股份有限公司 | Deployment method and system for cloud platform |
-
2008
- 2008-07-25 CN CNU2008200995467U patent/CN201260187Y/en not_active Expired - Fee Related
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017067449A1 (en) * | 2015-10-19 | 2017-04-27 | 中兴通讯股份有限公司 | Deployment method and system for cloud platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102611735A (en) | Load balancing method and system of application services | |
| CN102333120B (en) | Flow storage system for load balance processing | |
| CN102546782A (en) | Distribution system and data operation method thereof | |
| CN102662639A (en) | Mapreduce-based multi-GPU (Graphic Processing Unit) cooperative computing method | |
| CN101262501A (en) | An inter-core communication method in multi-core system and one multi-core system | |
| CN102394903B (en) | A kind of active reconstruction calculating system constructing system | |
| CN102981890A (en) | Computing task and virtual machine deploying method within a virtual data center | |
| CN101753405A (en) | Cluster server memory management method and system | |
| CN202068449U (en) | Data exchange platform used for multistage data exchange | |
| CN102324076A (en) | Unification all-purpose card payment system based on cloud calculation and method thereof | |
| CN103473848A (en) | Network invoice checking frame and method based on high concurrency | |
| CN111260146A (en) | Method, device, equipment and medium for locating power system edge cloud data center | |
| CN201260187Y (en) | Network topology structure for distributed dynamic load balance | |
| CN110519354A (en) | A kind of distributed objects storage system and its method for processing business and storage medium | |
| CN102801750B (en) | Method for constructing cloud computing target system | |
| Guo | Aggregating uncertain incast transfers in BCube-like data centers | |
| CN105302649A (en) | Disaster recovery backup method and system | |
| CN102724290A (en) | Method, device and system for getting target customer group | |
| CN101778135A (en) | Server system of rural general message platform and operation method thereof | |
| Raj et al. | Novel Practices and Trends in Grid and Cloud Computing | |
| CN101420354A (en) | Multicast spreading method oriented to wide area network remote virtual environment | |
| CN103246262A (en) | Comprehensive service system for network data analysis | |
| CN103902574A (en) | Real-time data loading method and device based on data flow technology | |
| CN103152428A (en) | Method for carrying out service communication among nodes on cloud platform | |
| CN203204423U (en) | Comprehensive service system for network data analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090617 Termination date: 20110725 |