CN105812468A - High-availability storage method based on SCST - Google Patents
High-availability storage method based on SCST Download PDFInfo
- Publication number
- CN105812468A CN105812468A CN201610145514.5A CN201610145514A CN105812468A CN 105812468 A CN105812468 A CN 105812468A CN 201610145514 A CN201610145514 A CN 201610145514A CN 105812468 A CN105812468 A CN 105812468A
- Authority
- CN
- China
- Prior art keywords
- data
- storage
- scst
- method based
- node
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of data storage, and especially relates to a high-availability storage method based on SCST. According to the method, through adoption of a VRRP protocol, load balance is realized at a front end; and a single node fault is avoided. When multiple servers are operated, a same virtual IP address is shared together; and however there is a master server and backup servers. When a fault appears in the master server, the IP can be taken over by the backup servers, thus providing service. The availability is improved.
Description
Technical field
The present invention relates to technical field of data storage, particularly to a kind of storage high availability method based on SCST.
Background technology
The another kind of realization that the scsi target subsystem that SCST, Linux are general is one is LinuxSCSI target sub system.It provides between linux kernel and processes and the unification of scsi target driver, accordant interface, provides linux kernel to be connected with storage rear end handle simultaneously, connects storage rear end that is true or that simulate and target drives program.
SCST allows to create complicated storage device, it is provided that advanced function, for instance replicates, automatically simplify configuration, data de-duplication technology, high availability, automated back-up etc..
The application of another classification is VTL (VTL) and other are based on the backup solution of disk.
Under prior art, being based on fault-toleranr technique for main failure handling parts, fault-toleranr technique is to utilize redundant hardware detection operating result, just carries out isolating trouble unit thus reducing the technology to customer impact when occurring abnormal.Owing to achieving the isolation of fault, thus improve the intensity of system and the repeatability of application.But fault-toleranr technique more focuses on the technology of hardware, its range of application receives the restriction of hardware condition.Simultaneously along with the reduction of hardware price, adopting general hardware platform, utilize the High Availabitity technology that software realizes to instead of fault-toleranr technique gradually, application also expands gradually.
Generally, for storage server cluster, High Availabitity key problem in technology technology includes LVS(load equalizer), heartbeat detection technology (HeartBeat, Corosync etc.) and application server (MySQL, NFS, Apache etc.).Can dispose and realize multiple application, but bottom storage service can realize virtualization storage based on SCST software, provides multiple interfaces to upper platform, it is achieved the service request in different application.
Pacemaker is a cluster resource manager.It utilizes message that the cluster basic component (OpenAIS, heartbeat or corosync) that you like provides and member management ability detects and recovers from the fault of node or resource class, to realize the maximum availability of Cluster Service (also known as resource).
LoadBalance, load balancing is based upon on existing network infrastructure, it provides the bandwidth of a kind of cheap effectively transparent method extended network equipment and server, increases handling capacity, Strengthens network data-handling capacity, the motility improving network and availability.
OpenStack be one by NASA(US National Aeronautics and Space Administration) and Rackspace R & D Cooperation initiating, with the free software of Apache licence mandate and open source code project.
OpenStack is a cloud computing management platform project increased income, several main assemblies combined specific works.OpenStack supports almost all kinds of cloud environment, project objective be to provide implement simple, can extend on a large scale, enrich, cloud computing management platform that standard is unified.Namely OpenStack services the solution of (IaaS) by the service of the various complementations facility that provides the foundation, and each service provides API integrated to carry out.
Summary of the invention
In order to solve problem of the prior art, the invention provides a kind of storage high availability method based on SCST, it may be used for virtualization storage, it is achieved that the dynamic distribution of resource, provides the interface load balancing for management end simultaneously.
The technical solution adopted in the present invention is as follows:
A kind of storage high availability method based on SCST, comprises the following steps:
A, client data first pass through network and transfer data to reception server；
The client data arrived is carried out load balance process according to arranging LoadBalance rule by B, storage server, selects a reception server less with load in access network, and is distributed file system form by the format conversion of data；
C, utilize Peacemaker software, carry out message management, Redundancy Management according to the rule arranged, and after the machine of delaying, carry out failover and recovery；
D, then data consumers end data write in dummy node, and dummy node is the memory node that memory virtual goes out；
E, utilize SCST software that the data of write dummy node are converted into scsi command, and scsi command is committed to hard disk carries out the write of client data.
The Data Frontend of different clients data employs LoadBalance software and different requests has been divided, it is achieved the load balancing of request.
A kind of storage high availability method based on SCST of the present invention, is converted into dummy storage node by SCST software by bottom physical disk；Some storage servers are converged as unified memory resource pool by dummy storage node by Pacemaker software；Then the distribution of resource is carried out at resource pool, by the storage resource format of distribution is turned to different file system；Adapt to the storage demand of different pieces of information.
Method also includes: realize the method that primary storage deletes superfluous system, it is realize the single backup of data in primary storage that described primary storage deletes superfluous system, ensure that data are while being normally written equipment, the part repeated is deleted, also to ensure the integrity of data in data.
Completing after primary storage deletes superfluous system, by reading in dummy storage node, configure simultaneously and read by different main frames, it is achieved many copies storage of data.
Technical scheme provided by the invention has the benefit that
The present invention is by the load balancing of VRRP protocol realization front end, it is to avoid Single Point of Faliure, when multiple servers runs, and common employing same virtual ip address, but there is master server and backup server.Can by standby server adapter IP when main service is broken down time, it is provided that service, improve availability.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme in the embodiment of the present invention, below the accompanying drawing used required during embodiment is described is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the enforcement system entire block diagram of a kind of storage high availability method based on SCST of the present invention；
Fig. 2 is control end and the client corresponding relation figure of a kind of storage high availability method based on SCST of the present invention;
Fig. 3 is the method flow diagram of a kind of storage high availability method based on SCST of the present invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, embodiment of the present invention is described further in detail.
A kind of storage high availability method based on SCST of the present embodiment, its specific implementation method is as follows:
1: at storage server, SCST software is installed
2: the dummy node mapped out by SCST is mapped to Client end (client)；
3: the software such as PaceMaker and LoadBalance on Client end is affixed one's name to, configure.Affix one's name to OpenStack client in Client end simultaneously.
4: dispose OpenStack control end at Control end (control end) and carry out resource distribution, scheduling and monitoring.
The present embodiment comprises following three partial contents:
1, overall system design
System overall schematic as it is shown in figure 1, wherein bottom Disk be physical disk, by SCST software, physical disk is converted into dummy storage node (VirtualDiskNode).This process also includes Intel Virtualization Technology, and the storage device of distinct interface, different structure is mapped as unified dummy storage node.It can be unified memory resource pool by the convergence of some storage servers that dummy storage node can pass through Pacemaker, facilitates using and dispatching of system.Then the distribution of resource is carried out at resource pool, by the storage resource format of distribution is turned to different file system (such as OCFS2 etc.), adapt to different storage demands, for instance MySQL needs data block little, the structure of high degree of dispersion, but Stream Media Application needs the structure of long data block.Data Frontend employs LoadBalance and different requests has been divided, the load balancing of request.
Control end and client corresponding relation as in figure 2 it is shown, control Pacemaker and LoadBalance by OpenStack to realize the configuration to same address pool and load equalizer, realize the configuration to SCST by Web.Then control end and can check the running state information of each software, United Dispatching.
2, virtualization storage
When storing server and adding one piece of disk, mean that disk enters in memory resource pool (diskgroup), this disk is divided into block (such as 1MB) according to identical size by storage software automatically, and the diskgroup of establishment is exactly a unified memory space, and each fritter is called a storage grain.
This mode disperseing recombinant from physical disk, the truth of bottom space can be masked, provide a unified storage visual angle for upper layer software (applications), memory space is become a unified resource pool or super hard drive space, then resource pool is split.
3, many copies storage
In order to realize many copies memory technology, it is necessary to realize primary storage and delete superfluous system, primary storage realizes the single backup of data, it is ensured that the part repeated in data, while being normally written equipment, is deleted, also to be ensured the integrity of data by data.
Delete superfluous system completing primary storage, it is possible to by reading in dummy storage node, configure simultaneously and read the many copies storage achieving data by different main frames.So can avoid single memory node Problem of Failure, can read from multiple ground connection when reading data simultaneously simultaneously, be greatly improved the reading speed of data.Coordinate concordance hash algorithm, more can avoid the invalid write etc. of data.
In the present embodiment, manager revises the configuration file of SCST by Web, carries out rights management and capacity management etc. for different hard disks, and storage resource is distributed to different customer groups.Controlling Pacemaker and LoadBalance by OpenStack simultaneously and realize the configuration to same address pool and load equalizer, after completing following configuration, server end can provide the user storage service.
As it is shown on figure 3, be issued to storage server for one group of data from client below to complete storing process, explain in detail its process.Client data first passes through network (including fiber optic network, Inifiband network or Ethernet) and transfers data to reception server.The data arrived are carried out load balance process according to arranging LoadBalance rule by storage server, select a reception server less with load in access network, and are distributed file system form by the format conversion of data.Peacemaker carries out message management (by Web feedback management person), Redundancy Management (if data exist, then storage is not repeated) according to the rule arranged, and carries out failover and recovery after the machine of delaying.Then data can write in dummy node, and dummy node is the memory node that memory virtual goes out.The data of write dummy node are converted into scsi command by SCST, and scsi command is committed to hard disk carries out data write.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all within the spirit and principles in the present invention, any amendment of making, equivalent replacement, improvement etc., should be included within protection scope of the present invention.
Claims (4)
1., based on a storage high availability method of SCST, comprise the following steps:
A, client data first pass through network and transfer data to reception server；
The client data arrived is carried out load balance process according to arranging LoadBalance rule by B, storage server, selects a reception server less with load in access network, and is distributed file system form by the format conversion of data；
C, utilize Peacemaker software, carry out message management, Redundancy Management according to the rule arranged, and after the machine of delaying, carry out failover and recovery；
D, then data consumers end data write in dummy node, and dummy node is the memory node that memory virtual goes out；
E, utilize SCST software that the data of write dummy node are converted into scsi command, and scsi command is committed to hard disk carries out the write of client data.
2. a kind of storage high availability method based on SCST according to claim 1, it is characterised in that the Data Frontend of described different clients data employs LoadBalance software and different requests has been divided, it is achieved the load balancing of request.
3. a kind of storage high availability method based on SCST according to claim 1, it is characterized in that, described method also includes: realize the method that primary storage deletes superfluous system, it is realize the single backup of data in primary storage that described primary storage deletes superfluous system, ensure that data are while being normally written equipment, the part repeated in data is deleted, also to ensure the integrity of data.
4. a kind of storage high availability method based on SCST according to claim 3, it is characterized in that, completing after primary storage deletes superfluous system, by reading in dummy storage node, configure simultaneously and read by different main frames, it is achieved many copies storage of data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610145514.5A CN105812468A (en) | 2016-03-15 | 2016-03-15 | High-availability storage method based on SCST |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610145514.5A CN105812468A (en) | 2016-03-15 | 2016-03-15 | High-availability storage method based on SCST |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105812468A true CN105812468A (en) | 2016-07-27 |
Family
ID=56468355
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610145514.5A Pending CN105812468A (en) | 2016-03-15 | 2016-03-15 | High-availability storage method based on SCST |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105812468A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107453932A (en) * | 2017-09-29 | 2017-12-08 | 郑州云海信息技术有限公司 | A kind of distributed memory system management method and its device |
| CN107846318A (en) * | 2017-11-15 | 2018-03-27 | 郑州云海信息技术有限公司 | A kind of distributed type assemblies and distributed type assemblies management method |
| CN108600308A (en) * | 2018-03-20 | 2018-09-28 | 新华三技术有限公司 | Data uploading method, device, storage medium and server |
| CN108769199A (en) * | 2018-05-29 | 2018-11-06 | 郑州云海信息技术有限公司 | A kind of distributed file storage system host node management method and device |
| CN114090476A (en) * | 2021-10-20 | 2022-02-25 | 珠海翔翼航空技术有限公司 | SCSI hard disk simulation method, system and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101702115A (en) * | 2009-11-27 | 2010-05-05 | 浪潮电子信息产业股份有限公司 | Method for realizing block protocol interface based on cluster storage system |
| CN101808123A (en) * | 2010-03-09 | 2010-08-18 | 浪潮(北京)电子信息产业有限公司 | Method and device for accessing storage resources in storage system |
| CN103268252A (en) * | 2013-05-12 | 2013-08-28 | 南京载玄信息科技有限公司 | Virtualization platform system based on distributed storage and achieving method thereof |
-
2016
- 2016-03-15 CN CN201610145514.5A patent/CN105812468A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101702115A (en) * | 2009-11-27 | 2010-05-05 | 浪潮电子信息产业股份有限公司 | Method for realizing block protocol interface based on cluster storage system |
| CN101808123A (en) * | 2010-03-09 | 2010-08-18 | 浪潮(北京)电子信息产业有限公司 | Method and device for accessing storage resources in storage system |
| CN103268252A (en) * | 2013-05-12 | 2013-08-28 | 南京载玄信息科技有限公司 | Virtualization platform system based on distributed storage and achieving method thereof |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107453932A (en) * | 2017-09-29 | 2017-12-08 | 郑州云海信息技术有限公司 | A kind of distributed memory system management method and its device |
| CN107453932B (en) * | 2017-09-29 | 2020-12-01 | 苏州浪潮智能科技有限公司 | Distributed storage system management method and device |
| CN107846318A (en) * | 2017-11-15 | 2018-03-27 | 郑州云海信息技术有限公司 | A kind of distributed type assemblies and distributed type assemblies management method |
| CN108600308A (en) * | 2018-03-20 | 2018-09-28 | 新华三技术有限公司 | Data uploading method, device, storage medium and server |
| CN108769199A (en) * | 2018-05-29 | 2018-11-06 | 郑州云海信息技术有限公司 | A kind of distributed file storage system host node management method and device |
| CN114090476A (en) * | 2021-10-20 | 2022-02-25 | 珠海翔翼航空技术有限公司 | SCSI hard disk simulation method, system and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11579991B2 (en) | Dynamic allocation of compute resources at a recovery site | |
| US11922197B2 (en) | Virtual server agent load balancing | |
| US11314543B2 (en) | Architecture for implementing a virtualization environment and appliance | |
| US6571354B1 (en) | Method and apparatus for storage unit replacement according to array priority | |
| US6598174B1 (en) | Method and apparatus for storage unit replacement in non-redundant array | |
| US6640278B1 (en) | Method for configuration and management of storage resources in a storage network | |
| CN105812468A (en) | High-availability storage method based on SCST | |
| US10241712B1 (en) | Method and apparatus for automated orchestration of long distance protection of virtualized storage | |
| US20100293552A1 (en) | Altering Access to a Fibre Channel Fabric | |
| US11853587B2 (en) | Data storage system with configurable durability | |
| US20190187934A1 (en) | Random access memory (ram)-based computer systems, devices, and methods | |
| US8713218B2 (en) | Non-disruptive configuration of a virtualization controller in a data storage system | |
| US9304875B2 (en) | Dynamically tracking logical units moving between input/output ports of a storage area network target | |
| CN109547537A (en) | The method for realizing openstack High Availabitity based on SAN storage shared volume | |
| US20230136106A1 (en) | Space efficient distributed storage systems | |
| US9514013B2 (en) | Maintaining inactive copy relationships for secondary storages of active copy relationships having a common primary storage for use in case of a failure of the common primary storage | |
| CN105141459A (en) | Optimization method based on multipath technology | |
| KR20230091168A (en) | Techniques for creating configurations to electrically isolate fault domains in data centers | |
| US11983578B2 (en) | Provisioning a computing subsystem including disaggregated hardware resources that comply with a power domain requirement for a workload | |
| KR101673882B1 (en) | Storage system with virtualization using embedded disk and method of operation thereof | |
| JP2024504171A (en) | Operating system-based storage methods and systems | |
| US20220357853A1 (en) | Maintaining replication consistency during distribution instance changes | |
| US20240220101A1 (en) | Cluster management in large-scale storage systems | |
| US20220358018A1 (en) | Journal barrier consistency determination | |
| Hai et al. | Research of the Router Scheme for Virtual Storage |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160727 |
|
| WD01 | Invention patent application deemed withdrawn after publication |