CN101741905A - Rapid deployment method for cluster - Google Patents
Rapid deployment method for cluster Download PDFInfo
- Publication number
- CN101741905A CN101741905A CN200910238672A CN200910238672A CN101741905A CN 101741905 A CN101741905 A CN 101741905A CN 200910238672 A CN200910238672 A CN 200910238672A CN 200910238672 A CN200910238672 A CN 200910238672A CN 101741905 A CN101741905 A CN 101741905A
- Authority
- CN
- China
- Prior art keywords
- cluster
- sub
- nodes
- isomorphism
- service
- 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.)
- Pending
Links
Images
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a rapid deployment method for a cluster. The method comprises the following steps: S102, setting a central node for providing a TRACKER service and a downloading service; S104, configuring different operating systems, services, a software and the like for a plurality of nodes except the central node in the cluster, wherein the nodes only provide the downloading service; and S106, setting a preset transmission mode among the nodes. By using the rapid deployment method for the cluster, the flexibility and performance bottleneck of the deployment of the operating system of the heterogeneous cluster are eliminated, and the efficiency of the deployment of the operating system in cluster is improved greatly.
Description
Technical field
The present invention relates to the cluster management technical field, more specifically, relate to a kind of quick deployment method of cluster.
Background technology
Cluster is exactly two or many computers or node co-operation in a group.Compare with the computer that works independently, cluster can provide higher availability and expandability.Each node in the cluster all has the resource (processor, I/O, internal memory, operating system, memory) of oneself usually, and oneself user is collected responsible.
Single computer in the group system is commonly referred to node, connects by local area network (LAN) usually.Cluster computer is commonly used to improve the computational speed and/or the reliability of single computer.Cluster computer is than single computer generally speaking, and is much higher such as the work station or the supercomputer ratio of performance to price.
The cluster classification cluster is divided into two kinds of isomorphism and isomeries, and their difference is: whether the architecture of forming between the computer of group system is identical.
At present, the allocating operating system mode in the cluster mainly is the server towards isomorphism, if the machine of different model is arranged in the cluster, then is difficult to carry out unified deployment.In addition, if cluster scale is bigger, then the network bandwidth, hard disk throughput rate just become the bottleneck of whole system, have influenced the quick deployment of whole cluster greatly.
Therefore, need a kind of operating system that can solve isomeric group to be difficult to unified plan, and can effectively improve the technical scheme that group system is disposed efficient.
Summary of the invention
At the problem that prior art exists, make the present invention.
According to embodiments of the invention, a kind of quick deployment method of cluster is provided, this method may further comprise the steps: S102, a Centroid is set, Centroid provides TRACKER service and download service; S104 carries out the difference configuration of operating system, service, software etc. respectively to a plurality of nodes except that Centroid in the cluster, and a plurality of nodes only provide download service; And S106, between a plurality of nodes, set preset transmission mode.
Particularly, preset transmission mode is point-to-point mode or NFS mode.
Further, a plurality of nodes can be subdivided into a plurality of sub-cluster,, carry out network design respectively then at different sub-cluster.Wherein, can a plurality of nodes be subdivided into a plurality of sub-cluster, and a plurality of sub-cluster can be isomorphism sub-cluster or isomery sub-cluster according to machine models, device type or purposes.
For in the isomorphism sub-cluster each, with one of them node as template processing machine, to template processing machine installing operating system, configuration service and install software, be a mirror image with the backup of the entire system of template processing machine then, and described mirror image is distributed to other nodes in this isomorphism sub-cluster.In addition, for the isomery sub-cluster, adopt the build-in services device identical to carry out network installation in the CD mirroring mode with the isomorphism sub-cluster.
On the other hand,, then in Centroid, preserve a plurality of mirror images that comprise template processing machine mirror image and CD mirroring, at the different different mirror images of isomorphism sub-cluster distribution if a plurality of sub-cluster all is the isomorphism sub-cluster.
In technical scheme of the present invention, remove after the Centroid, other all nodes all are equality, be simultaneously server be again client, each node also provides upload service for other nodes when downloading.Therefore, solve the flexibility of isomeric group allocating operating system and the problem of performance bottleneck, improved the deployment efficient of operating system in the whole cluster greatly.
Other features and advantages of the present invention will be set forth in the following description, and, partly from specification, become apparent, perhaps understand by implementing the present invention.Purpose of the present invention and other advantages can realize and obtain by specifically noted structure in the specification of being write, claims and accompanying drawing.
Description of drawings
Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used to explain the present invention, do not constitute improper qualification of the present invention.In the accompanying drawings:
Fig. 1 is the flow chart that illustrates according to the quick deployment method of the cluster of the embodiment of the invention;
Fig. 2 illustrates the schematic diagram that adopts point-to-point deployment way in an embodiment of the present invention;
Fig. 3 illustrates the scheme of installation that in an embodiment of the present invention cluster is divided into isomorphism sub-cluster and isomery sub-cluster; And
Fig. 4 illustrates the scheme of installation that in an embodiment of the present invention cluster is divided into a plurality of isomorphism sub-cluster.
Embodiment
Describe embodiments of the invention in detail below in conjunction with accompanying drawing.
Fig. 1 is the flow chart that illustrates according to the quick deployment method of the cluster of the embodiment of the invention.
With reference to Fig. 1, may further comprise the steps according to the method for the embodiment of the invention: S102, a Centroid is set, Centroid provides TRACKER service and download service; S104 carries out the difference configuration of operating system, service, software etc. respectively to a plurality of nodes except that Centroid in the cluster, and a plurality of nodes only provide download service; And S106, between a plurality of nodes, set preset transmission mode.
Fig. 2 illustrates the schematic diagram that adopts point-to-point deployment way in an embodiment of the present invention.
Suppose that the A node is an intermediate node, other node B, C ... expression.The A node provides TRACKER service, and other all nodes and this communication for services such as B, C, D are informed progress and the state of oneself and obtained the information of other node.Suppose that the B node downloaded file 001, need to download 002, and C, D have downloaded 002, need to download 001 that then Tracker " tells " B node file 002 on C, D, the while, Tracker " told " C, D node file 001 on B.So B is from C, D file in download sheet 002, and simultaneously with oneself downloaded files sheet 001 offer C, D and download.Like this, B, C, D be file in download both, again while " uploading " file.
Intermediate node is not controlled other node when network installation, it only provides service, and TRACKER both is provided service, and download service also is provided, and other node then only provides download service.The characteristic of point-to-point transmission that Here it is.Server does not link to each other with one of them node among the figure, and this is a special case in the actual installation process, that is, this node does not constantly need to download and install file from intermediate node.Intermediate node is not controlled other node, and other node always removes to download and install file on one's own initiative.
In the figure, isomeric group has been realized point-to-point mode, but might not need to use point-to-point mode to connect, point-to-point just wherein a kind of transmission means.If selected point-to-point mode to connect, then except the tracker node, all nodes all are equality.If do not have choice point, then can connect, for example NFS (network file system abbreviates NFS as) mode by alternate manner to point mode.
Further, a plurality of nodes can be subdivided into a plurality of sub-cluster,, carry out network design respectively then at different sub-cluster.
Generally, the mode that cluster is segmented comprises: divide according to machine models; Divide according to device types such as hard disks; And according to purposes division (the necessary model unanimity or the device type unanimity of same purposes, such as the A620 type machine, the installation purposes is divided into the web service and ftp serves two kinds).
Fig. 3 illustrates the scheme of installation that in an embodiment of the present invention cluster is divided into isomorphism sub-cluster and isomery sub-cluster, and Fig. 4 illustrates the scheme of installation that in an embodiment of the present invention cluster is divided into a plurality of isomorphism sub-cluster.
With reference to Fig. 3,, install in the following manner for the sub-cluster of isomorphism: look for a machine as template processing machine earlier, installing operating system, and dispose the service of required configuration, and required software is installed, the entire system backup with this node is a mirror image then.Because the node isomorphism that in this sub-cluster is, promptly, all nodes can have identical configuration (have only configuration differences such as IP, and these different configurations can be provided with when mounted as required at different nodes), and can be distributed to the mirror image of this template processing machine other node and install this moment.
If it is not (for example, the IO node in the general cluster etc.) of isomorphism that minority node and most of node are arranged in the cluster, then can use the mode of CD mirroring to carry out network installation by configuration.Notice that this moment, the build-in services device and the above-mentioned isomorphism cluster build-in services device of CD mirroring were same node.
With reference to Fig. 4, if this part isomeric group quantity is more, also it can be considered as an independent isomorphism sub-cluster, can find out a node as template processing machine according to above-mentioned installation method, use the template processing machine mirror image to install.
If isomeric group is divided into a plurality of isomorphism sub-cluster, then only need a node (Centroid) that all sub-cluster are carried out master control, this node is preserved a plurality of mirror images (existing template processing machine mirror image also has CD mirroring), can be at the different different mirror images of sub-cluster distribution.
In technical scheme of the present invention, remove after the Centroid, other all nodes all are equality, be simultaneously server be again client, each node also provides upload service for other nodes when downloading.Therefore, solve the flexibility of isomeric group allocating operating system and the problem of performance bottleneck, improved the deployment efficient of operating system in the whole cluster greatly.
The above is the preferred embodiments of the present invention only, is not limited to the present invention, and for a person skilled in the art, the present invention can have various changes and variation.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (9)
1. the quick deployment method of a cluster is characterized in that, said method comprising the steps of:
S102 is provided with a Centroid, and described Centroid provides TRACKER service and download service;
S104 carries out the difference configuration of operating system, service, software etc. respectively to a plurality of nodes except that described Centroid in the described cluster, and described a plurality of nodes only provide download service;
And
S106 sets preset transmission mode between described a plurality of nodes.
2. method according to claim 1 is characterized in that, described preset transmission mode is point-to-point mode.
3. method according to claim 1 is characterized in that, described preset transmission mode is the NFS mode.
4. method according to claim 1 is characterized in that, described a plurality of nodes are subdivided into a plurality of sub-cluster, then at different sub-cluster, carries out network design respectively.
5. method according to claim 4 is characterized in that, according to machine models, device type or purposes described a plurality of nodes is subdivided into a plurality of sub-cluster.
6. method according to claim 4 is characterized in that, described a plurality of sub-cluster are isomorphism sub-cluster or isomery sub-cluster.
7. method according to claim 6, it is characterized in that, for in the described isomorphism sub-cluster each, with one of them node as template processing machine, to described template processing machine installing operating system, configuration service and install software, be a mirror image with the backup of the entire system of described template processing machine then, and described mirror image is distributed to other nodes in this isomorphism sub-cluster.
8. method according to claim 7 is characterized in that, for described isomery sub-cluster, adopts the build-in services device identical with described isomorphism sub-cluster to carry out network installation in the CD mirroring mode.
9. method according to claim 6, it is characterized in that, if described a plurality of sub-cluster all is the isomorphism sub-cluster, then in described Centroid, preserve a plurality of mirror images that comprise template processing machine mirror image and CD mirroring, at the different different mirror images of isomorphism sub-cluster distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910238672A CN101741905A (en) | 2009-11-30 | 2009-11-30 | Rapid deployment method for cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910238672A CN101741905A (en) | 2009-11-30 | 2009-11-30 | Rapid deployment method for cluster |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101741905A true CN101741905A (en) | 2010-06-16 |
Family
ID=42464788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910238672A Pending CN101741905A (en) | 2009-11-30 | 2009-11-30 | Rapid deployment method for cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101741905A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147809A (en) * | 2011-03-22 | 2011-08-10 | 曙光信息产业股份有限公司 | Parallel file system and management method thereof |
| CN102436501A (en) * | 2011-12-02 | 2012-05-02 | 曙光信息产业(北京)有限公司 | Parallel file managing system based on web |
| CN102455936A (en) * | 2010-11-25 | 2012-05-16 | 中标软件有限公司 | Trunk quick allocation method |
| CN102497408A (en) * | 2011-12-08 | 2012-06-13 | 曙光信息产业(北京)有限公司 | Semiautomatic batch deployment method for heterogeneous cluster operating system |
| CN102520989A (en) * | 2011-12-01 | 2012-06-27 | 浪潮电子信息产业股份有限公司 | Intelligent colony system deployment method based on U disk |
| CN102736946A (en) * | 2011-04-11 | 2012-10-17 | 阿里巴巴集团控股有限公司 | Batch application node deploying method and device |
| WO2016000137A1 (en) * | 2014-06-30 | 2016-01-07 | 北京新媒传信科技有限公司 | Method, client and system for multi-site automatic update |
| CN109962941A (en) * | 2017-12-14 | 2019-07-02 | 华为技术有限公司 | Communication means, device and server |
| CN110784546A (en) * | 2019-10-31 | 2020-02-11 | 浙江大华技术股份有限公司 | Distributed cluster deployment method, server and storage device |
-
2009
- 2009-11-30 CN CN200910238672A patent/CN101741905A/en active Pending
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102455936A (en) * | 2010-11-25 | 2012-05-16 | 中标软件有限公司 | Trunk quick allocation method |
| CN102147809A (en) * | 2011-03-22 | 2011-08-10 | 曙光信息产业股份有限公司 | Parallel file system and management method thereof |
| CN102736946A (en) * | 2011-04-11 | 2012-10-17 | 阿里巴巴集团控股有限公司 | Batch application node deploying method and device |
| CN102520989A (en) * | 2011-12-01 | 2012-06-27 | 浪潮电子信息产业股份有限公司 | Intelligent colony system deployment method based on U disk |
| CN102436501A (en) * | 2011-12-02 | 2012-05-02 | 曙光信息产业(北京)有限公司 | Parallel file managing system based on web |
| CN102497408A (en) * | 2011-12-08 | 2012-06-13 | 曙光信息产业(北京)有限公司 | Semiautomatic batch deployment method for heterogeneous cluster operating system |
| CN102497408B (en) * | 2011-12-08 | 2015-08-19 | 曙光信息产业(北京)有限公司 | A kind of method of isomeric group operating system Semiautomatic batch deployment |
| WO2016000137A1 (en) * | 2014-06-30 | 2016-01-07 | 北京新媒传信科技有限公司 | Method, client and system for multi-site automatic update |
| CN109962941A (en) * | 2017-12-14 | 2019-07-02 | 华为技术有限公司 | Communication means, device and server |
| CN109962941B (en) * | 2017-12-14 | 2021-10-26 | 华为技术有限公司 | Communication method, device and server |
| CN110784546A (en) * | 2019-10-31 | 2020-02-11 | 浙江大华技术股份有限公司 | Distributed cluster deployment method, server and storage device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101741905A (en) | Rapid deployment method for cluster | |
| CN102209087B (en) | Method and system for MapReduce data transmission in data center having SAN | |
| US7370083B2 (en) | System and method for providing virtual network attached storage using excess distributed storage capacity | |
| US9886260B2 (en) | Managing software version upgrades in a multiple computer system environment | |
| JP5914245B2 (en) | Load balancing method considering each node of multiple layers | |
| US20050060608A1 (en) | Maximizing processor utilization and minimizing network bandwidth requirements in throughput compute clusters | |
| CN102413172B (en) | Parallel data sharing method based on cluster technology and apparatus thereof | |
| US20040098729A1 (en) | System and method for reducing user-application interactions to archivable form | |
| US10127077B2 (en) | Event distribution pattern for use with a distributed data grid | |
| JP2004518335A (en) | How to process large amounts of data | |
| WO2002048866A8 (en) | Method and system for management of multiple network resources | |
| CN107454171B (en) | Message service system and implementation method thereof | |
| CN102497408B (en) | A kind of method of isomeric group operating system Semiautomatic batch deployment | |
| CN112351051A (en) | Cloud service processing method and device, cloud server, cloud service processing system and storage medium | |
| CN102983996A (en) | Dynamic allocation method and system for high-availability cluster resource management | |
| US11494130B2 (en) | Operation data accessing device and accessing method thereof | |
| CN102932455A (en) | Construction method based on cloud computing render farms | |
| CN104123183A (en) | Cluster assignment dispatching method and device | |
| CN108259569A (en) | It is a kind of based on IPSAN share storage without acting on behalf of continuous data protection method | |
| CN101853185B (en) | Blade server and service dispatching method thereof | |
| CN104158833A (en) | Method for constructing intelligent desktop system | |
| CN112350855B (en) | Configuration-based cloud center management method | |
| CN107566509B (en) | Information publishing system capable of bearing large-batch terminals | |
| CN113778334A (en) | Distributed storage system based on electronic files | |
| CN113630317A (en) | Data transmission method and device, nonvolatile storage medium and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent for invention or patent application | ||
| CB02 | Change of applicant information |
Address after: 100193 Beijing, Haidian District, northeast Wang West Road, building 8, No. 36 Applicant after: Dawning Information Industry (Beijing) Co., Ltd. Address before: 100084 No. 6 South Road, Zhongguancun Academy of Sciences, Beijing, Haidian District Applicant before: Dawning Information Industry (Beijing) Co., Ltd. |
|
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20100616 |