CN107070972A - A kind of distributed document processing method and processing device - Google Patents
A kind of distributed document processing method and processing device Download PDFInfo
- Publication number
- CN107070972A CN107070972A CN201611265985.6A CN201611265985A CN107070972A CN 107070972 A CN107070972 A CN 107070972A CN 201611265985 A CN201611265985 A CN 201611265985A CN 107070972 A CN107070972 A CN 107070972A
- Authority
- CN
- China
- Prior art keywords
- server
- meta data
- data server
- file
- file system
- 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
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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/10—Network architectures or network communication protocols for network security for controlling access to devices or network resources
-
- 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
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Computing Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to technical field of data processing, a kind of distributed document processing method and processing device is disclosed, including：The file read-write request that meta data server receiving terminal is sent, the file read-write request is sent according to the share directory of the meta data server of the terminal carry；The meta data server sends the file read-write to the storage cluster of the file system and asked, the meta data server is located in identical file system cluster with the storage cluster according to the file system of the share directory carry.The present invention to solve different operating system in the prior art terminal can not incoming file system, or be linked into file system process it is complicated, the problem of taking longer.
Description
Technical field
The present invention relates to technical field of data processing, more particularly to a kind of distributed document processing method and processing device.
Background technology
As emerging Network storage technology, Ceph provide a dependable performance, favorable expandability, unified, distribution
Formula storage system, is a free open source software.Object storage service is only provided different from Swift, Ceph also provides block simultaneously
Store, the storage service such as file storage.
Ceph systems can be greatly classified into four parts：Client (client)；MDS (meta data server, Metadata
Server cluster), for caching and synchronous metadata；Osd (Object storage cluster, object storage collection
Group), for data storage；Mon (Cluster monitors, cluster monitoring device), performs function for monitoring.
For Ceph file system, user terminal is needed in locally-installed corresponding client software bag, and carry Ceph
File system, could conduct interviews to Ceph file system, and existing Ceph file system has two kinds of carry modes：Based on kernel
Carry and user's space carry are driven, corresponding software is respectively Ceph-fuse and mount.ceph.For both carry sides
Formula, before carry Ceph file system, be required to carry out cephx key authentications, and to the application of cluster monitoring device service IP and
Port.When listening port, service IP or cephx certification keys when cluster monitoring device change, client needs corresponding
Carry is changed, and because part operation system does not support the installation of software kit, such as windows systems or unix systems,
Then be mounted with this part operation system terminal can not incoming file system, or be linked into file system process it is considerably complicated numerous
Trivial, time loss is longer.
The content of the invention
The embodiment of the present invention provides a kind of distributed document processing method and processing device, to solve different behaviour in the prior art
File system can not be linked into by making the terminal of system, or be linked into the process complexity of file system, the problem of taking longer.
Distributed document processing method provided in an embodiment of the present invention includes：
The file read-write request that meta data server receiving terminal is sent, file read-write request be the terminal according to
What the share directory of the meta data server of carry was sent；
The meta data server is according to the file system of the share directory carry, to the storage collection of the file system
File read-write described in pocket transmission is asked, and the meta data server is located in identical file system cluster with the storage cluster.
Optionally, the meta data server includes host node server and slave node server；
The file read-write request that the meta data server receiving terminal is sent, including：
The file read-write request that the host node server receiving terminal is sent；
After the file read-write request that the meta data server receiving terminal is sent, in addition to：
The host node server is when it is determined that file read-write request is the read-write requests of multiple files, according to all
The workload of meta data server, determines the slave node server for handling the file read-write request；
The host node server asks the file read-write to be sent to the slave node clothes by internal transmission passage
Business device.
Optionally, the workload according to all meta data servers, determines the processing file read-write request
Slave node server, including：
For a meta data server, the meta data server calculates the meta data server according to below equation
Workload：
Gk=Nk*Tk
Wherein, NkThe quantity of the file read-write request stored in caching for K meta data servers, TkFor K metadata
Total time in server process caching needed for all file read-write requests, GkFor the workload of K meta data servers.
Optionally, the external storage service IP of the meta data server is same virtual IP address；The meta data server
Including host node server and slave node server；
Also include：The host node server enters the service access port of the host node server and the virtual IP address
Row binding.
Optionally, in addition to：
The meta data server receives the Status Change information of cluster monitoring device, and the cluster monitoring device belongs to the text
Part system cluster.
Optionally, in addition to：
Any piece of metadata server of the meta data server into the file system cluster sends detection messages, with
Determine the working condition of any piece of metadata server.
Optionally, any piece of metadata server of the meta data server into the file system cluster sends detection
Message, including：
The meta data server sends ICMP packets；Or,
The meta data server monitors the state of the service access port of any piece of metadata server；Or,
The meta data server monitors the service availability of any piece of metadata server.
Optionally, the share directory is provided by NFS NFS.
A kind of distributed document processing unit, including：
Receiving module, the file read-write request sent for receiving terminal, the file read-write request is according to the end
Hold the share directory transmission of the meta data server of carry；
Sending module, for the file system according to the share directory carry, to the storage cluster of the file system
The file read-write request is sent, the meta data server is located in identical file system cluster with the storage cluster.
Optionally, the meta data server includes host node server and slave node server；
Also include processing module, be used for：
When it is determined that file read-write request is the read-write requests of multiple files, according to the work of all meta data servers
Measure, determine the slave node server for handling the file read-write request；
File read-write request is sent to the slave node server by internal transmission passage.
Optionally, the processing module, specifically for：
For a meta data server, the workload of the meta data server is calculated according to below equation：
Gk=Nk*Tk
Wherein, NkThe quantity of the file read-write request stored in caching for K meta data servers, TkFor K metadata
Total time in server process caching needed for all file read-write requests, GkFor the workload of K meta data servers.
Optionally, the external storage service IP of the meta data server is same virtual IP address；The meta data server
Including host node server and slave node server；
Also include binding module, for the service access port of the host node server and the virtual IP address to be tied up
It is fixed.
Optionally, the receiving module, is additionally operable to：
The Status Change information of cluster monitoring device is received, the cluster monitoring device belongs to the file system cluster.
Optionally, the sending module, is additionally operable to：
Any piece of metadata server into the file system cluster sends detection messages, to determine any first number
According to the working condition of server.
Optionally, the sending module, specifically for：
Send ICMP packets；Or,
Monitor the state of the access port of any piece of metadata server；Or,
Monitor the service availability of any piece of metadata server.
Optionally, the share directory is provided by NFS NFS.
In the embodiment of the present invention, the share directory that carry meta data server is provided in terminal, terminal needs into style of writing
During the access of part, according to share directory, file read-write request is sent to meta data server.Shared mesh on meta data server
Carry file system is recorded, after the file read-write request that meta data server receiving terminal is sent, according to carry in share directory
File system, to file system storage cluster send file read-write request, storage cluster can according to this document read-write please
Seek the corresponding read-write operation of execution.The embodiment of the present invention assigns the access ability of terminal-pair file system by two-stage carry, and
Terminal need not install corresponding file system client software.Simultaneously as meta data server and storage cluster be located at it is same
Individual file system cluster, then meta data server and storage cluster have carry on identical cluster attribute, meta data server
During file system, the process without carrying out key authentication, it is to avoid the cumbersome configuration operation of terminal and software installation and early stage
Preparation.The terminal of different operating system, so as to access file system, can also be eliminated by direct carry shared file
Terminal is linked into the complex process of file system, saves time overhead and the resource input of manpower and materials.In addition, when monitoring end
When mouth, external storage service IP or cephx certification keys change, because the carry to file system is in metadata clothes
It is engaged on device, therefore, terminal need not change carry, and terminal is once configured, you can continuation uses the service of file system.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, makes required in being described below to embodiment
Accompanying drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this
For the those of ordinary skill in field, without having to pay creative labor, it can also be obtained according to these accompanying drawings
His accompanying drawing.
Fig. 1 is a kind of flow chart of distributed document processing method in the embodiment of the present invention；
Fig. 2 be the embodiment of the present invention in schematic diagram from host node server to slave node server distribution of document；
The schematic diagram for the system framework that Fig. 3 is handled for distributed document in the specific embodiment of the present invention；
Fig. 4 is a kind of structural representation of distributed document processing unit in the embodiment of the present invention.
Embodiment
In order that the object, technical solutions and advantages of the present invention are clearer, below in conjunction with accompanying drawing the present invention is made into
One step it is described in detail, it is clear that described embodiment is only embodiment of the invention a part of, rather than whole implementation
Example.Based on the embodiment in the present invention, what those of ordinary skill in the art were obtained under the premise of creative work is not made
All other embodiment, belongs to the scope of protection of the invention.
Two kinds of carry modes of existing Ceph file system：Based on kernel-driven carry and user's space carry, hang
When carrying file system, it is required for loading cephx certifications key and then seeks service by accessing Ceph mon listening port.
Consider following two situations：
1st, mon calamities are under switch mode, and new mon nodes are added, or IP, the listening port of mon nodes need to change
When, it is necessary to artificially change the service access IP and port parameter of client, to guarantee correctly to access the progress of ceph file system
Data read-write operation.
2nd, typically in client carry ceph file system, terminal needs to install ceph-fuse software kits and phase
The dependence bag of pass, and obtain cephx certifications key and could complete.And may not have in installation source built in part operation system
Ceph-fuse client softwares and related dependant bag, or operating system do not support ceph-fuse software in itself, for example
Windows and aix, and some server nodes because sensitive application has been installed and requirement can not arbitrarily update software source, dependence
Bag, ceph-fuse is installed in terminal will involve a large amount of early-stage preparations operations.And for mount.ceph carry modes, behaviour
Make the problem of system is not supported and seem more prominent.Mount.ceph needs the support of operating system nucleus, but not all
The operating system of version all supports this carry mode.Part operation system for supporting kernel-driven carry, due to operation system
The kernel version of system is too low, it is necessary to upgrade corresponding kernel to support ceph carries, this not only process it is cumbersome, can also be quick to deploying
The Ceph file system servers of sense business bring greater risk.
As can be seen here, the deployment that above-mentioned two situations are serviced for production environment, calamity for service recovery under pattern etc. all
Corresponding time overhead can be increased, this is obviously unsatisfactory for the ageing requirement of business.And with the expansion of cluster scale, corresponding behaviour
Then need the resource input of more manpower and materials.Accordingly, it would be desirable to which a kind of be not used in terminal installing the more general of client
Easily scheme, to use the massive store space that Ceph is provided.
In order to solve the above problems, the embodiments of the invention provide a kind of distributed document processing method, as shown in figure 1,
Comprise the following steps：
The file read-write request that step 101, meta data server receiving terminal are sent, the file read-write request is basis
What the share directory of the meta data server of the terminal carry was sent.
Step 102, the meta data server are according to the file system of the share directory carry, to the file system
Storage cluster send file read-write request, the meta data server is located at identical file system with the storage cluster
In cluster.
In the embodiment of the present invention, the share directory that carry meta data server is provided in terminal, terminal needs into style of writing
During the access of part, according to share directory, file read-write request is sent to meta data server.Shared mesh on meta data server
Carry file system is recorded, after the file read-write request that meta data server receiving terminal is sent, according to carry in share directory
File system, to file system storage cluster send file read-write request, storage cluster can according to this document read-write please
Seek the corresponding read-write operation of execution.The embodiment of the present invention assigns the access ability of terminal-pair file system by two-stage carry, and
Terminal need not install corresponding file system client software.Simultaneously as meta data server and storage cluster be located at it is same
Individual file system cluster, then meta data server and storage cluster have carry on identical cluster attribute, meta data server
During file system, the process without carrying out key authentication, it is to avoid the cumbersome configuration operation of terminal and software installation and early stage
Preparation.The terminal of different operating system, so as to access file system, can also be eliminated by direct carry shared file
Terminal is linked into the complex process of file system, saves time overhead and the resource input of manpower and materials.In addition, when monitoring end
When mouth, external storage service IP or cephx certification keys change, because the carry to file system is in metadata clothes
It is engaged on device, therefore, terminal need not change carry, and terminal is once configured, you can continuation uses the service of file system.
In the embodiment of the present invention, share directory is provided by shared-file system, and shared-file system is installed on metadata clothes
Share directory can be supplied to user terminal by business device, such meta data server using shared-file system.Meta data server
It is upper to support most of shared-file system, but in order to ensure to recommend to adopt in the user friendly of packing service, the embodiment of the present invention
With NFS (Network File System, NFS), it allows to pass through TCP/IP nets between the computer in network
Network shared resource.NFS can pass through TCP/IP (Transmission Control Protocol/Internet because of it
Protocol, transmission control protocol/Internet Protocol) computer in real-time performance network resource-sharing function and by
Industry extensive use.Using meta data server as NFS server end, configuration/etc/ in meta data server
Exportfs files, add the configuration sentence of NFS dependent-shared directories.Using user terminal as NFS client, it is whole in user
Carry NFS share directory in end.In NFS application, local NFS client can be pellucidly read and write positioned at distal end NFS
File on server, just as accessing local file.Compared to other network shared files systems, there is NFS Linux to write from memory
Recognize installation, kernel is directly supported, the features such as configuring simple.And shared-file system brings another benefit to be using NFS：With
Family terminal just only need to can be in carry meta data server by the NFS clients that are provided in kernel share directory, so as to visit
The file system of carry in share directory is asked, client configuration operation is simplified.
, can be by being sealed in meta data server by file system mounted in meta data server in the embodiment of the present invention
Ceph-fuse softwares are filled to realize, so can using meta data server as this document system client.On the other hand, by member
Share directory in data server is supplied to user terminal to carry out carry, because the embodiment of the present invention recommends NFS to support this
Function, therefore, also needs to encapsulate NFS, i.e. meta data server again as the NFS services in user terminal in meta data server
Device.File system is shared to user terminal, exempted from by the embodiment of the present invention by two layers of server-client architecture, two layers of carry
Monitoring of the cluster monitoring device to user terminal is gone, that is to say, that user terminal is without the key authentication by cluster monitoring device
The service to Ceph file system can be sought.
When the state of cluster monitoring device is changed, it, which becomes, further includes new cluster monitoring device addition, IP replacings, monitoring client
Mouth change etc., is that meta data server sends Status Change information, member in embodiments of the present invention to the client of file system
Data server receives the Status Change information of cluster monitoring device, and the cluster monitoring device belongs to the file system cluster.
That is, cluster monitoring device belongs to same file system cluster with meta data server.So, storage cluster, first number
Belonged to according to server and cluster monitoring device in same file system cluster, therefore, when internal state changes, metadata
Client on server makes corresponding renewal, and operation is updated without user terminal.
Metadata (Metadata), also known as broker data, relaying data, to describe data (the data about of data
Data), the information of data attribute is mainly described, for supporting as indicated storage location, historical data, resource lookup, file
The functions such as record.A kind of metadata electronic type catalogue at last, describes and has collected the interior perhaps characteristic of data, and then reach assistance
The purpose of data retrieval.Meta data server is that metadata is stored and managed, for file system, in data storage
When, file to be stored is divided into multiple block structures by system, then each block structure is stored respectively, meta data server
In i.e. store the information such as the mark, storage location, storage time of each block structure, so, data write and read mistake
Cheng Zhong, meta data server may indicate that the storage location of data, support data search etc..
In the embodiment of the present invention, Ceph-fuse and NFS in meta data server can also be packaged in other modules or service
In device, but it is optimal case to be packaged in meta data server.This is due to that Ceph-fuse and NFS are packaged in into other modules, is treated
The data of storage are also needed by meta data server, and the information such as the data storage of data are stored, equally, read storage
Data also need the instruction by meta data server, so that it is determined that the storage location of data.Therefore, Ceph-fuse and NFS is sealed
Loaded in meta data server, the shortest path of data flow can be made, Internet resources are saved.
In order to ensure that the meta data server in the High Availabitity of file system cluster, the embodiment of the present invention takes including host node
Business device and slave node server.Wherein, host node server can be one, and slave node server is multiple.Then first number
The file read-write request sent according to server receiving terminal, including：
The file read-write request that the host node server receiving terminal is sent；
After the file read-write request that the meta data server receiving terminal is sent, in addition to：
The host node server is when it is determined that file read-write request is the read-write requests of multiple files, according to all
The workload of meta data server, determines the slave node server for handling the file read-write request；
The host node server asks the file read-write to be sent to the slave node clothes by internal transmission passage
Business device.
Specifically, user terminal sends file read-write request to file system, due to including multiple members in file system
Data server, typically receives this document read-write requests by host node server, and is sent to text after handling file read-write request
The storage cluster of part system bottom.In order to mitigate host node server file storage addressing pressure, when host node server connects
When the file read-write request received is the read-write on multiple files, then need to carry out host node server load balancing, it is main
Node server can by software by file distributing give each slave node server.So, by using each slave node server
File system access ability, reached the parallel memorizing of file-level, improved file read-write efficiency and the profit of hardware resource
With rate.Specific practice can be that host node, for each file, takes after multiple files are received according to all metadata
The workload of business device, determines the slave node server of processing this document, and the read-write requests of this document are passed through into internal proprietary
The file forwarding network segment is sent to the slave node server.
The selection of the slave node server of file read-write request is handled, to select the meta data server that workload is minimum.
The workload according to all meta data servers, determines the slave node server for handling the file read-write request, bag
Include：
For a meta data server, the meta data server calculates the meta data server according to below equation
Workload：
Gk=Nk*Tk... ... ... formula 1
Wherein, NkThe quantity of the file read-write request stored in caching for K meta data servers, TkFor K metadata
Total time in server process caching needed for all file read-write requests, GkFor the workload of K meta data servers.
Specifically, host node server receives multiple files, and each file is placed on into caching queues up, right one by one
File is handled.When handling each file, the workload of each meta data server is calculated according to formula 1, from all members
Workload is selected in data server minimum.For example, as shown in Fig. 2 the file read-write that host node server is connected to please
Asking includes 3 files, the Queue sequence in caching, labeled as file a, file b, file c.File a is handled first, it is main
Node server calculates the workload of each meta data server, i.e., read file pending in the caching of meta data server
It is multiplied the time required to the file read-write request pending with handling these of the quantity of write request, draws the meta data server
Workload, this workload namely meta data server processing file a before remaining workload.From all first numbers
It is minimum according to workload is found out in server, file a is sent to the meta data server.It should be noted that the metadata
Server can be slave node server, it is also possible to host node server, if host node server, then the host node is serviced
Device is directly handled, without transmission process.File a is sent to after the meta data server of processing, file b is investigated, likewise, main
Node server calculates each meta data server remaining workload before processing file b, therefrom selects workload minimum
Meta data server, to handle file b.Same processing is also done for file c.File process task is distributed by formula 1, is protected
It is most not busy to have demonstrate,proved the server chosen, and load balancing and parallel processing is better achieved, the treatment effeciency of file is improved.
The embodiment of the present invention also ensures whole Metadata Service except meeting load balancing by multiple slave node servers
The High Availabitity of device cluster.Wherein, any piece of metadata server hair of the meta data server into the file system cluster
Detection messages are sent, to determine the working condition of any piece of metadata server.
In metadata server cluster, high-availability cluster software is installed on the meta data server of part.Each metadata
Server sends detection report by high-availability cluster software to remaining meta data server for deploying high-availability cluster software
Text, to detect the working condition of each meta data server.For host node server, if detecting any slave node server
Working condition be abnormal condition, then by the slave node server reject file system cluster.For slave node server, if
The working condition for detecting host node server is abnormal condition, then according to sequence set in advance, will be located at primary
Slave node server is as host node server, and high-availability cluster software is primary for section to this by all service bulk migrations
On point server, therefore, all slave node servers are consistent with the software encapsulated on host node server so that each standby section
Point server can be provided to be serviced with host node server identical.
It is preferred that any piece of metadata server of the meta data server into the file system cluster sends detection
Message, including：
The meta data server sends ICMP packets；Or,
The meta data server monitors the state of the access port of any piece of metadata server；Or,
The meta data server monitors the service availability of any piece of metadata server.
High-availability cluster software in the embodiment of the present invention is recommended to use keepalived softwares.Keepalived softwares are comprehensive
The advantage of most of High Availabitity softwares is closed, it is equal that the software can provide more preferable scalability, high availability and load for cluster
Weighing apparatus ability.The High Availabitity realization mechanism of keepalived softwares mainly acts on the layer 3 network layer of server OS seven layer models,
4th layer of transport layer and the 7th layer of application layer.For Internet, periodically ICMP (Internet Control are sent to other side's server
Message Protocol, network Internet Control Message Protocol) packet, if ICMP packets send failure, show other side take
The network of business device is obstructed, it is impossible to successfully receives data, need to be rejected server cluster.For transport layer, keepalived leads to
The state of monitoring other side's server TCP port is crossed to judge the state of other side's server transport layer, if port is closed mode,
Show the non-abnormal condition of transport layer, then rejected server cluster.For application layer, by checking that other side's server is provided
The availability of service comes whether determining server is working properly, and whether whether such as other side's server can normally log in, can normally locate
Manage file to be read and write etc..No matter which layer, keepalived are worked in once finding that some meta data server loses in cluster
Effect, will be rejected cluster.When the former data server in abnormal condition is main node server, keepalived
Can be by whole services migrating to slave node server, slave node server is provided and the service of host node server identical.Then
As the meta data server for host node delay machine when, its service provided will integrally switch to slave node, it is ensured that service
The continuity of offer.
In the conversion of above-mentioned host node, in order to ensure the access for not influenceing user terminal, the meta data server
External storage service IP is same virtual IP address；The meta data server includes host node server and slave node server.This
Inventive embodiments also include：The host node server is by the service access port of the host node server and the virtual IP address
Bound.
That is, the IP that whole file system provides service to user terminal is virtual IP address (VIP).All metadata clothes
Business device has the virtual IP address, and wherein host node server is bound service access port and virtual IP address by software, uses
Family terminal can be directly connected to the service access port of host node server by virtual IP address.When host node server is delayed machine
Afterwards, not only all service is moved in slave node server, and the slave node server is also by software by service access end
Mouth is bound with virtual IP address, and now the service access port of host node server and virtual IP address are unbinding.So, user is whole
End is attached to the service access port of slave node server by virtual IP address, because the IP that user terminal is used is virtual IP address,
No matter how the server inside file system changes for this, and its virtual IP address is constant, then user terminal need not follow changing for server
Change is updated, and is further ensured that continuity of the user using service.So, due to provide service virtual IP address be it is fixed,
Any modification made in file system has no effect on any non-of access of the user terminal to file system, i.e. file system
The modification of nfs server is all " transparent " for user terminal.So the scheme of the embodiment of the present invention is for user terminal
Speech, can accomplish the once configuration of carry, continuously available.
In addition, user terminal can configure/etc/fstab information, the service access port pair provided by file system
Share directory carries out the automation carry of automation carry, i.e. ceph systems, to meet the automatically dispose requirement of user terminal.
In order to be more clearly understood that the present invention, scheme is described in detail with specific embodiment below, Fig. 3 is shown
System framework in the specific embodiment of the present invention.
As shown in figure 3, including user terminal and server end.
Wherein, NFS clients (NFS client) are installed in user terminal, it is shared in carry meta data server
Catalogue.
Server end includes Ceph file system clusters, and Ceph file system clusters include storage cluster, storage cluster
In have the data that file is resolved into fixed size in multiple memory cell, i.e. osd.0, osd.1 ... osd.n, memory cell
Block is stored.
Also include multiple meta data servers, i.e. server where mds modules in Ceph file system clusters, wherein one
Individual meta data server is the mds (active) in main node server, i.e. figure, and remaining meta data server is slave node service
Mds (standby) in device, i.e. figure.No matter host node server or slave node server, pacify in each meta data server
The software all same of dress, it is ensured that the High Availabitity of whole metadata server cluster.In addition, by software by virtual IP address (VIP)
Host node server is bound to, so, during user terminal access file system, its binding is connected to always by the virtual IP address
Port, so as to obtain the service of file system.
Mds modules, ceph-fuse softwares, keepalived are mounted in each meta data server in cluster soft
Part and nfs server end (NFS server) software.Certainly, part metadata is also included in whole Ceph file system cluster
Mds modules, ceph-fuse softwares, keepalived softwares and nfs server end software are not installed, then this part in server
Meta data server is not included in the range of the discussion of the embodiment of the present invention.Wherein, ceph-fuse causes meta data server to make
For the client of Ceph file system, by the file system mounted local folders in meta data server of Ceph.NFS
Server externally shares the local folders as share directory so that meta data server is used as NFS server end, peace
Share directory is filled, so that meta data server is file system mounted in share directory by Ceph.Keepalived has caused master
Node server is the meta data server that working condition is normal condition, it is ensured that the high availability of metadata server cluster
With operation duration.
The specific embodiment of the present invention is by ceph-fuse softwares, keepalived softwares and nfs server end (NFS
Server) software carries out service encapsulation, and so, meta data server can be provided out above-mentioned respective services.
Also include multiple cluster monitoring devices, i.e. mon.a, mon.b ... in Ceph file system clusters, cluster monitoring device exists
During the client access of Ceph file system, Cephx key authentications are carried out, and change in the IP of cluster monitoring device or port etc.
When, message is sent to client, so that the progress of client updates accordingly.But in one particular embodiment of the present invention,
Because the client of Ceph file system is meta data server, and meta data server and cluster monitoring device are in same Ceph
In file system cluster, therefore, the access of client wirelessly carries out Cephx key authentications, and Ceph file system cluster internals
State or device replacing, nor affect on the use service of user terminal.
In addition, in Fig. 3, public network are the publicly-owned nets that Ceph file system cluster externally provides read-write service
Section；Cluster network are transmission file, synchronous documents between each osd in Ceph file system clusters, send the special of heartbeat
The network segment；Trans network are that host node server reads and writes task to the distribution of document of secondary node server during service is encapsulated
The special network segment.Each network segment is separate, and the division of labor is clear and definite, it is ensured that the safe efficient running of service.
Based on identical technical concept, the embodiment of the present invention also provides a kind of distributed document processing unit, such as Fig. 4 institutes
Show, including：
Receiving module 401, the file read-write request sent for receiving terminal, the file read-write request is according to described
What the share directory of the meta data server of terminal carry was sent；
Sending module 402, for the file system according to the share directory carry, to the storage collection of the file system
File read-write described in pocket transmission is asked, and the meta data server is located in identical file system cluster with the storage cluster.
Optionally, the meta data server includes host node server and slave node server；
Also include processing module 403, be used for：
When it is determined that file read-write request is the read-write requests of multiple files, according to the work of all meta data servers
Measure, determine the slave node server for handling the file read-write request；
File read-write request is sent to the slave node server by internal transmission passage.
Optionally, the processing module 403, specifically for：
For a meta data server, the workload of the meta data server is calculated according to below equation：
Gk=Nk*Tk
Wherein, be K meta data servers caching in store file read-write request quantity, be K metadata clothes
It is engaged in file read-write requests all in device processing caching required total time, is the workload of K meta data servers.
Optionally, the external storage service IP of the meta data server is same virtual IP address；The meta data server
Including host node server and slave node server；
Also include binding module 404, for the service access port of the host node server and the virtual IP address to be entered
Row binding.
Optionally, the receiving module 401, is additionally operable to：
The Status Change information of cluster monitoring device is received, the cluster monitoring device belongs to the file system cluster.
Optionally, the sending module 402, is additionally operable to：
Any piece of metadata server into the file system cluster sends detection messages, to determine any first number
According to the working condition of server.
Optionally, the sending module 402, specifically for：
Send ICMP packets；Or,
Monitor the state of the service access port of any piece of metadata server；Or,
Monitor the service availability of any piece of metadata server.
Optionally, the share directory is provided by NFS NFS.
Generally speaking, any change that cluster monitoring device is done in Ceph file system all makees corresponding by meta data server
Operation is updated, focusing on for change operation has been reached, has simplified change operating procedure.And due to providing shared service
VIP is fixed, and any modification that service end is made has no effect on any non-of the access of terminal-pair file system, i.e. service end
NFS server modification is all " transparent " for terminal.So the embodiment of the present invention is for user terminal, it can accomplish
The once configuration of carry, it is continuously available.
In addition, the NFS clients connection packing service that the user terminal in the embodiment of the present invention is provided using system kernel
NFS server, i.e. meta data server completes the access to Ceph file system by servicing transfer, reached system
Load balancing, simplify storage service access step so that reduce user terminal deployment take, improve work effect
Rate.And user terminal can be by configuring, and the service access port provided using packing service carries out automatic to share directory
Change carry, meet the automatically dispose requirement of user terminal.
High Availabitity aspect of performance, the embodiment of the present invention realizes whole system " no Single Point of Faliure ".For file read-write clothes
It is engaged in for supplier's Ceph file system clusters, its stability dependency is designed in the High Availabitity of itself, i.e., when mon, mds, osd event
There is corresponding handover mechanism after barrier.And because data are triplicate memory modules, security, the integrality of file are obtained
Guarantee is arrived.And for NFS systems, each meta data server all deploys identical software and business, its high availability
Realization ensured by keepalived.As the mds for main node server delay machine when, its service provided is by overall switching
To slave node server, it is ensured that the continuity that service is provided.In addition, the network connection of whole system all uses double netcard
Binding mode, it is ensured that the high availability energy of network.So far, whole scheme realizes software, the work of hardware High Availabitity all standing
Operation mode.
The present invention is the flow with reference to method according to embodiments of the present invention, equipment (system) and computer program product
Figure and/or block diagram are described.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
, but those skilled in the art once know basic creation although preferred embodiments of the present invention have been described
Property concept, then can make other change and modification to these embodiments.So, appended claims are intended to be construed to include excellent
Select embodiment and fall into having altered and changing for the scope of the invention.
Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the present invention to the present invention
God and scope.So, if these modifications and variations of the present invention belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprising including these changes and modification.
Claims (16)
1. a kind of distributed document processing method, it is characterised in that including：
The file read-write request that meta data server receiving terminal is sent, the file read-write request is according to the terminal carry
The meta data server share directory send；
The meta data server is sent out according to the file system of the share directory carry to the storage cluster of the file system
The file read-write is sent to ask, the meta data server is located in identical file system cluster with the storage cluster.
2. the method as described in claim 1, it is characterised in that the meta data server includes host node server and standby section
Point server；
The file read-write request that the meta data server receiving terminal is sent, including：
The file read-write request that the host node server receiving terminal is sent；
After the file read-write request that the meta data server receiving terminal is sent, in addition to：
The host node server is when it is determined that file read-write request is the read-write requests of multiple files, according to all first numbers
According to the workload of server, the slave node server for handling the file read-write request is determined；
The host node server asks the file read-write to be sent to the slave node server by internal transmission passage.
3. method as claimed in claim 2, it is characterised in that the workload according to all meta data servers, it is determined that
Go out the slave node server for handling the file read-write request, including：
For a meta data server, the meta data server calculates the work of the meta data server according to below equation
Measure：
Gk=Nk*Tk
Wherein, NkThe quantity of the file read-write request stored in caching for K meta data servers, TkFor K Metadata Services
Total time in device processing caching needed for all file read-write requests, GkFor the workload of K meta data servers.
4. the method as described in claim 1, it is characterised in that the external storage service IP of the meta data server is same
Virtual IP address；The meta data server includes host node server and slave node server；
Also include：The host node server is tied up the service access port of the host node server and the virtual IP address
It is fixed.
5. the method as described in claim 1, it is characterised in that also include：
The meta data server receives the Status Change information of cluster monitoring device, and the cluster monitoring device belongs to the file system
System cluster.
6. the method as described in any one of claim 1 to 5, it is characterised in that also include：
Any piece of metadata server of the meta data server into the file system cluster sends detection messages, to determine
The working condition of any piece of metadata server.
7. method as claimed in claim 6, it is characterised in that the meta data server is into the file system cluster
Any piece of metadata server sends detection messages, including：
The meta data server sends ICMP packets；Or,
The meta data server monitors the state of the service access port of any piece of metadata server；Or,
The meta data server monitors the service availability of any piece of metadata server.
8. method as claimed in claim 6, it is characterised in that the share directory is provided by NFS NFS.
9. a kind of distributed document processing unit, it is characterised in that including：
Receiving module, the file read-write request sent for receiving terminal, the file read-write request is hung according to the terminal
What the share directory of the meta data server of load was sent；
Sending module, for the file system according to the share directory carry, sends to the storage cluster of the file system
The file read-write request, the meta data server is located in identical file system cluster with the storage cluster.
10. device as claimed in claim 9, it is characterised in that the meta data server includes host node server and standby
Node server；
Also include processing module, be used for：
When it is determined that file read-write request is the read-write requests of multiple files, according to the work of all meta data servers
Amount, determines the slave node server for handling the file read-write request；
File read-write request is sent to the slave node server by internal transmission passage.
11. device as claimed in claim 10, it is characterised in that the processing module, specifically for：
For a meta data server, the workload of the meta data server is calculated according to below equation：
Gk=Nk*Tk
Wherein, NkThe quantity of the file read-write request stored in caching for K meta data servers, TkFor K Metadata Services
Total time in device processing caching needed for all file read-write requests, GkFor the workload of K meta data servers.
12. device as claimed in claim 9, it is characterised in that the external storage service IP of the meta data server is same
One virtual IP address；The meta data server includes host node server and slave node server；
Also include binding module, for the service access port of the host node server and the virtual IP address to be bound.
13. device as claimed in claim 9, it is characterised in that the receiving module, is additionally operable to：
The Status Change information of cluster monitoring device is received, the cluster monitoring device belongs to the file system cluster.
14. the device as described in any one of claim 9 to 13, it is characterised in that the sending module, is additionally operable to：
Any piece of metadata server into the file system cluster sends detection messages, to determine that any piece of metadata takes
The working condition of business device.
15. device as claimed in claim 14, it is characterised in that the sending module, specifically for：
Send ICMP packets；Or,
Monitor the state of the service access port of any piece of metadata server；Or,
Monitor the service availability of any piece of metadata server.
16. device as claimed in claim 14, it is characterised in that the share directory is provided by NFS NFS.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611265985.6A CN107070972B (en) | 2016-12-30 | 2016-12-30 | Distributed file processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611265985.6A CN107070972B (en) | 2016-12-30 | 2016-12-30 | Distributed file processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107070972A true CN107070972A (en) | 2017-08-18 |
| CN107070972B CN107070972B (en) | 2020-02-07 |
Family
ID=59624357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611265985.6A Active CN107070972B (en) | 2016-12-30 | 2016-12-30 | Distributed file processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107070972B (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107547447A (en) * | 2017-08-31 | 2018-01-05 | 郑州云海信息技术有限公司 | Distributed Filesystem Network communication means, device and network communicating system |
| CN107769996A (en) * | 2017-10-24 | 2018-03-06 | 新华三云计算技术有限公司 | A kind of service end working state detecting method and device |
| CN108089818A (en) * | 2017-12-12 | 2018-05-29 | 腾讯科技（深圳）有限公司 | Data processing method, device and storage medium |
| 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 |
| CN108900606A (en) * | 2018-06-28 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of cross-system realizes the method, device and equipment of data sharing |
| CN108900602A (en) * | 2018-06-27 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of virtual linkage method for building up and relevant apparatus |
| CN109165206A (en) * | 2018-08-27 | 2019-01-08 | 中科曙光国际信息产业有限公司 | HDFS high availability implementation method based on container |
| CN109343801A (en) * | 2018-10-23 | 2019-02-15 | 深圳前海微众银行股份有限公司 | Date storage method, equipment and computer readable storage medium |
| CN109445709A (en) * | 2018-11-05 | 2019-03-08 | 郑州云海信息技术有限公司 | The management method and device of storage resource in virtualization system |
| CN109688214A (en) * | 2018-12-24 | 2019-04-26 | 杭州数梦工场科技有限公司 | Data transmission method, device and storage medium |
| CN109783438A (en) * | 2018-12-05 | 2019-05-21 | 南京华讯方舟通信设备有限公司 | Distributed NFS system and its construction method based on librados |
| CN109885552A (en) * | 2019-02-18 | 2019-06-14 | 天固信息安全系统(深圳)有限责任公司 | The metadata dynamic management approach and distributed file system of distributed file system |
| CN110865989A (en) * | 2019-11-22 | 2020-03-06 | 浪潮电子信息产业股份有限公司 | Business processing method for large-scale computing cluster |
| CN111225003A (en) * | 2018-11-23 | 2020-06-02 | 北京京东金融科技控股有限公司 | NFS node configuration method and device |
| CN111338902A (en) * | 2020-02-28 | 2020-06-26 | 上海商汤智能科技有限公司 | Data processing method, device and system |
| CN111367740A (en) * | 2020-02-29 | 2020-07-03 | 苏州浪潮智能科技有限公司 | BMC debugging system and method and computer readable storage medium |
| CN111488322A (en) * | 2020-04-08 | 2020-08-04 | 网易（杭州）网络有限公司 | File system service method and device and server equipment |
| CN111625513A (en) * | 2020-05-11 | 2020-09-04 | 紫光云技术有限公司 | Method for realizing file storage gateway |
| CN111930706A (en) * | 2020-07-08 | 2020-11-13 | 华东计算技术研究所（中国电子科技集团公司第三十二研究所） | Distributed network file storage system and method based on remote call |
| CN112328571A (en) * | 2020-11-25 | 2021-02-05 | 创新奇智(重庆)科技有限公司 | Object storage method and device, communication equipment and storage medium |
| CN113535094A (en) * | 2021-08-06 | 2021-10-22 | 上海德拓信息技术股份有限公司 | Cross-platform client implementation method based on distributed storage |
| CN115378930A (en) * | 2021-05-17 | 2022-11-22 | 成都鼎桥通信技术有限公司 | File distribution method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101789976A (en) * | 2010-01-27 | 2010-07-28 | 上海交通大学 | Embedded network storage system and method thereof |
| CN103858407A (en) * | 2013-12-02 | 2014-06-11 | 华为技术有限公司 | File processing method, device and system |
| CN104320401A (en) * | 2014-10-31 | 2015-01-28 | 北京思特奇信息技术股份有限公司 | Big data storage and access system and method based on distributed file system |
| CN104639658A (en) * | 2015-03-12 | 2015-05-20 | 浪潮集团有限公司 | Method for realizing object storage by mounting and accessing file system |
| WO2016110004A1 (en) * | 2015-01-05 | 2016-07-14 | 中兴通讯股份有限公司 | Management method and apparatus for disk storage space, and storage device |
-
2016
- 2016-12-30 CN CN201611265985.6A patent/CN107070972B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101789976A (en) * | 2010-01-27 | 2010-07-28 | 上海交通大学 | Embedded network storage system and method thereof |
| CN103858407A (en) * | 2013-12-02 | 2014-06-11 | 华为技术有限公司 | File processing method, device and system |
| CN104320401A (en) * | 2014-10-31 | 2015-01-28 | 北京思特奇信息技术股份有限公司 | Big data storage and access system and method based on distributed file system |
| WO2016110004A1 (en) * | 2015-01-05 | 2016-07-14 | 中兴通讯股份有限公司 | Management method and apparatus for disk storage space, and storage device |
| CN104639658A (en) * | 2015-03-12 | 2015-05-20 | 浪潮集团有限公司 | Method for realizing object storage by mounting and accessing file system |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107547447B (en) * | 2017-08-31 | 2021-06-29 | 郑州云海信息技术有限公司 | Network communication method and device of distributed file system and network communication system |
| CN107547447A (en) * | 2017-08-31 | 2018-01-05 | 郑州云海信息技术有限公司 | Distributed Filesystem Network communication means, device and network communicating system |
| CN107769996A (en) * | 2017-10-24 | 2018-03-06 | 新华三云计算技术有限公司 | A kind of service end working state detecting method and device |
| CN108089818A (en) * | 2017-12-12 | 2018-05-29 | 腾讯科技（深圳）有限公司 | Data processing method, device and storage medium |
| 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 |
| CN108900602A (en) * | 2018-06-27 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of virtual linkage method for building up and relevant apparatus |
| CN108900602B (en) * | 2018-06-27 | 2021-08-10 | 郑州云海信息技术有限公司 | Virtual link establishing method and related device |
| CN108900606A (en) * | 2018-06-28 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of cross-system realizes the method, device and equipment of data sharing |
| CN109165206A (en) * | 2018-08-27 | 2019-01-08 | 中科曙光国际信息产业有限公司 | HDFS high availability implementation method based on container |
| CN109165206B (en) * | 2018-08-27 | 2022-02-22 | 中科曙光国际信息产业有限公司 | High-availability implementation method for HDFS (Hadoop distributed File System) based on container |
| CN109343801A (en) * | 2018-10-23 | 2019-02-15 | 深圳前海微众银行股份有限公司 | Date storage method, equipment and computer readable storage medium |
| CN109343801B (en) * | 2018-10-23 | 2021-11-09 | 深圳前海微众银行股份有限公司 | Data storage method, device and computer readable storage medium |
| CN109445709A (en) * | 2018-11-05 | 2019-03-08 | 郑州云海信息技术有限公司 | The management method and device of storage resource in virtualization system |
| CN111225003B (en) * | 2018-11-23 | 2022-12-27 | 京东科技控股股份有限公司 | NFS node configuration method and device |
| CN111225003A (en) * | 2018-11-23 | 2020-06-02 | 北京京东金融科技控股有限公司 | NFS node configuration method and device |
| CN109783438A (en) * | 2018-12-05 | 2019-05-21 | 南京华讯方舟通信设备有限公司 | Distributed NFS system and its construction method based on librados |
| CN109688214A (en) * | 2018-12-24 | 2019-04-26 | 杭州数梦工场科技有限公司 | Data transmission method, device and storage medium |
| CN109885552B (en) * | 2019-02-18 | 2023-08-18 | 天固信息安全系统(深圳)有限责任公司 | Metadata dynamic management method of distributed file system and distributed file system |
| CN109885552A (en) * | 2019-02-18 | 2019-06-14 | 天固信息安全系统(深圳)有限责任公司 | The metadata dynamic management approach and distributed file system of distributed file system |
| CN110865989A (en) * | 2019-11-22 | 2020-03-06 | 浪潮电子信息产业股份有限公司 | Business processing method for large-scale computing cluster |
| CN111338902A (en) * | 2020-02-28 | 2020-06-26 | 上海商汤智能科技有限公司 | Data processing method, device and system |
| CN111338902B (en) * | 2020-02-28 | 2024-04-12 | 上海商汤智能科技有限公司 | Data processing method, device and system |
| CN111367740A (en) * | 2020-02-29 | 2020-07-03 | 苏州浪潮智能科技有限公司 | BMC debugging system and method and computer readable storage medium |
| CN111488322A (en) * | 2020-04-08 | 2020-08-04 | 网易（杭州）网络有限公司 | File system service method and device and server equipment |
| CN111488322B (en) * | 2020-04-08 | 2023-05-12 | 网易（杭州）网络有限公司 | File system service method and device and server equipment |
| CN111625513A (en) * | 2020-05-11 | 2020-09-04 | 紫光云技术有限公司 | Method for realizing file storage gateway |
| CN111930706B (en) * | 2020-07-08 | 2024-04-09 | 华东计算技术研究所（中国电子科技集团公司第三十二研究所） | Remote call-based distributed network file storage system and method |
| CN111930706A (en) * | 2020-07-08 | 2020-11-13 | 华东计算技术研究所（中国电子科技集团公司第三十二研究所） | Distributed network file storage system and method based on remote call |
| CN112328571A (en) * | 2020-11-25 | 2021-02-05 | 创新奇智(重庆)科技有限公司 | Object storage method and device, communication equipment and storage medium |
| CN115378930A (en) * | 2021-05-17 | 2022-11-22 | 成都鼎桥通信技术有限公司 | File distribution method and device |
| CN113535094A (en) * | 2021-08-06 | 2021-10-22 | 上海德拓信息技术股份有限公司 | Cross-platform client implementation method based on distributed storage |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107070972B (en) | 2020-02-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107070972A (en) | A kind of distributed document processing method and processing device | |
| CN114946166B (en) | Method and system for dynamic resource movement in heterogeneous computing environments including cloud edge locations | |
| US11392400B2 (en) | Enhanced migration of clusters based on data accessibility | |
| US10044550B2 (en) | Secure cloud management agent | |
| US8589543B2 (en) | Virtual data center monitoring | |
| CA2783452C (en) | Migrating virtual machines among networked servers upon detection of degrading network link operation | |
| US10936549B2 (en) | Cluster-wide container optimization and storage compression | |
| EP2944070B1 (en) | Service migration across cluster boundaries | |
| US20220353237A1 (en) | Systems and methods of database instance container deployment | |
| CN108369544A (en) | Deferred server recovery in a computing system | |
| US20160269439A1 (en) | Deploying a security appliance system in a high availability environment without extra network burden | |
| WO2015191649A1 (en) | Providing multiple synchronous serial console sessions using data buffering | |
| CN109144972A (en) | A kind of method and back end of Data Migration | |
| US8543680B2 (en) | Migrating device management between object managers | |
| CN105095103A (en) | Storage device management method and device used for cloud environment | |
| CN102346707A (en) | Server system and operation method thereof | |
| US20200310658A1 (en) | Machine learning for local caching of remote data in a clustered computing environment | |
| CN108363611A (en) | Method for managing security, device and the omnidirectional system of virtual machine | |
| WO2021041039A1 (en) | Computational instance batching and automation orchestration based on resource usage and availability | |
| CN112448833A (en) | Multi-management-domain communication method and device | |
| US11876689B1 (en) | Method and system for SLA/QoS adherence-based flex on demand in a multi-API virtual desktop infrastructure (VDI) environment | |
| CN114584515B (en) | Endpoint notification of storage area network congestion | |
| US11977909B2 (en) | Hardware placement and maintenance scheduling in high availability systems | |
| CN110647289A (en) | Satellite remote sensing cloud computing platform and system | |
| CN118282866B (en) | Multi-tenant isolation deployment method, system, equipment and medium based on container cluster |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |