CN107707654B - The management method of container mirror image data, apparatus and system - Google Patents
The management method of container mirror image data, apparatus and system Download PDFInfo
- Publication number
- CN107707654B CN107707654B CN201710922809.3A CN201710922809A CN107707654B CN 107707654 B CN107707654 B CN 107707654B CN 201710922809 A CN201710922809 A CN 201710922809A CN 107707654 B CN107707654 B CN 107707654B
- Authority
- CN
- China
- Prior art keywords
- server
- mirror image
- data management
- image data
- data
- 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.)
- Active
Links
- 238000007726 management method Methods 0.000 title claims abstract description 64
- 238000013523 data management Methods 0.000 claims abstract description 342
- 238000012360 testing method Methods 0.000 claims abstract description 161
- 238000012545 processing Methods 0.000 claims abstract description 59
- 230000004044 response Effects 0.000 claims abstract description 46
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000004891 communication Methods 0.000 claims description 114
- 230000005540 biological transmission Effects 0.000 claims description 45
- 238000010586 diagram Methods 0.000 description 14
- 235000013399 edible fruits Nutrition 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
-
- 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
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Facsimiles In General (AREA)
Abstract
The application provides a kind of management method of container mirror image data, apparatus and system, this method comprises: obtaining mirror image data scheduling request;The first test request is sent according to mirror image data scheduling request, while at least two data management servers;Obtain the first test result information of runtime server feedback;According to the first test result information, first object data management server is determined;Mirror image data scheduling request is sent to first object data management server, so that first object data management server carries out response processing to mirror image data scheduling request.By using at least two data management servers be mutually backups ground memory image data, improve the safety of mirror image data.It solves in the prior art, only one data management server, once the problem of mirror image data goes wrong, and the corresponding first category vessel of the mirror image data cannot be then created.
Description
Technical field
This application involves field of communication technology more particularly to a kind of management methods of container mirror image data, apparatus and system.
Background technique
With the rapid development of virtual technology, container cluster is increasingly becoming the important technology of big data processing, using container
When cluster carries out big data processing, it usually needs creation container, and the creation of container needs to dispatch mirror image data.
In the prior art, there is usually one data management servers for the management of mirror image data, when certain class container is corresponding
Mirror image data when something goes wrong, then when creating such container, since its corresponding mirror image data goes wrong, and cannot be dispatched
The mirror image data creates container, therefore can not restart operation container, and the safety of mirror image data cannot ensure.
Therefore, the technical issues of safety of mirror image data, which how is effectively ensured, becomes urgent need to resolve.
Summary of the invention
The application provides a kind of management method of container mirror image data, apparatus and system, there was only one to solve the prior art
A data management server, once mirror image data goes wrong, the corresponding first category vessel of the mirror image data cannot be then created
Defect.
The application first aspect provides a kind of management method of container mirror image data, comprising:
General pipeline server obtains mirror image data scheduling request;
According to the mirror image data scheduling request, while the first test is sent at least two data management servers and is asked
Ask so that the data management server to runtime server send the first test data packet, trigger the runtime server to
First test result information of general pipeline server feedback;
Obtain the first test result information of the runtime server feedback;
According to first test result information, first object data management server is determined;
The mirror image data scheduling request is sent to the first object data management server, so that first mesh
Mark data management server carries out response processing to the mirror image data scheduling request.
According to method as described above, optionally, the method also includes:
Obtain the storage solicited message for storing new mirror image data that runtime server is sent;
According to the storage solicited message, while the second test is sent at least two data management servers and is asked
Ask so that the data management server to runtime server send the second test data packet, trigger the runtime server to
Second test result information of general pipeline server feedback;
Obtain the second test result information of the runtime server feedback;
According to second test result information, the second target data management server is determined;
Sending the first communication to the runtime server proves that Xiang Suoshu the second target data management server sends second
Communication proves, so that the runtime server proves the first communication and new mirror image data to be stored is sent to described the
Two target data management servers, trigger the second target data management server proves and described according to first communication
Second communication proves, after being verified, carries out storage processing to the new mirror image data, and to except second target
Other data management servers except data management server send communication request.
According to method as described above, optionally, proved sending the first communication to the runtime server, Xiang Suoshu the
After two target data management servers send the second communication proof, the method also includes:
Pre-receiving request is sent to other data management servers in addition to the second target data management server, with
It requests that other described data management servers according to the pre-receiving, the second target data management server is sent
The communication request feedback response information.
According to method as described above, optionally, if the second target data management server do not receive it is described its
The response message of his data management server, then the method also includes:
Obtain the communication failure that the second target data management server is sent information and the new mirror image number
According to;
Timing sends third test request to other described data management servers;
When that can be communicated with other described data management servers, by the new mirror image data be sent to it is described other
Data management server, so that other described data management servers carry out storage processing to the new mirror image data.
Further aspect of the application provides a kind of management method of container mirror image data, comprising: data management server
Obtain the first test request that general pipeline server is sent;
The first test data packet is sent to runtime server according to first test request, to trigger the operation service
Device is to first test result information of general pipeline server feedback;
If the data management server is confirmed as first object data management server, the general pipeline service is received
The mirror image data scheduling request that device is sent;
Response processing is carried out according to the mirror image data scheduling request.
It is optionally, described that response processing is carried out according to the mirror image data scheduling request according to method as described above, packet
It includes:
According to the mirror image data scheduling request, corresponding target mirror image data is sent to runtime server, so that institute
It states runtime server and runs the target mirror image data to create container.
According to method as described above, optionally, the method also includes:
Obtain the second test request that general pipeline server is sent;
The second test data packet is sent to runtime server according to second test request, to trigger the operation service
Device is to second test result information of general pipeline server feedback;
If the data management server is confirmed as the second target data management server, the general pipeline service is received
The second communication that device is sent proves;
Obtain the first communication proof and new mirror image data that runtime server is sent;
It is proved according to the first communication and second communication proves to be verified;
If being verified, storage processing carried out to the new mirror image data, and to being not determined to the second number of targets
Communication request is sent according to other data management servers of management server, so that other described data management servers are according to total
The pre-receiving request that pipe server is sent, to the communication request feedback response information.
According to method as described above, optionally, if being carried out at storage being verified to the new mirror image data
Reason, and to after being not determined to other data management servers transmission communication request of the second target data management server,
The method also includes:
If getting the response message of other data management servers feedback, the new mirror image data is sent
To other described data management servers, to be backed up;
If not getting the response message, to the information of the general pipeline server feedback communication failure, and by institute
It states new mirror image data and is sent to the general pipeline server, so that the general pipeline server is arrived in test to be counted with described other
When according to management server communication, the new mirror image data is sent to other described data management servers, to be backed up.
The another aspect of the application provides a kind of managing device of container mirror image data, comprising: first obtains module, uses
In acquisition mirror image data scheduling request;
First sending module, for according to the mirror image data scheduling request, while at least two data management services
Device sends the first test request, so that the data management server sends the first test data packet, triggering to runtime server
The runtime server is to first test result information of general pipeline server feedback;
First receiving module, for obtaining the first test result information of the runtime server feedback;
First processing module, for determining first object data management server according to first test result information;
First sending module is also used to the mirror image data scheduling request being sent to the first object data pipe
Server is managed, so that the first object data management server carries out response processing to the mirror image data scheduling request.
According to device as described above, optionally, described first obtains module, is also used to obtain runtime server transmission
Store the storage solicited message of new mirror image data;
First sending module is also used to according to the storage solicited message, while at least two data pipes
It manages server and sends the second test request;So that the data management server sends the second test data to runtime server
Packet, triggers the runtime server to second test result information of general pipeline server feedback;
First receiving module is also used to obtain the second test result information of the runtime server feedback;
The first processing module is also used to determine the second target data management according to second test result information
Server;
First sending module, being also used to send the first communication to the runtime server proves, the second mesh of Xiang Suoshu
Mark data management server send second communication prove so that the runtime server by first communication prove and it is to be stored
New mirror image data is sent to the second target data management server, triggers the second target data management server root
It is proved according to first communication and second communication proves, after being verified, the new mirror image data is deposited
Storage processing, and communication request is sent to other data management servers in addition to the second target data management server.
According to device as described above, optionally, first sending module is also used to except second target data
Other data management servers outside management server send pre-receiving request so that other described data management servers according to
The pre-receiving request, the communication request feedback response information that the second target data management server is sent.
According to device as described above, optionally, if the second target data management server do not receive it is described its
The response message of his data management server, then first receiving module, is also used to obtain the second target data management
The information for the communication failure that server is sent and the new mirror image data;
First sending module is also used to timing to other described data management servers and sends third test request;
First sending module is also used to when that can communicate with other described data management servers, will be described new
Mirror image data be sent to other described data management servers so that other described data management servers are to the new mirror
As data carry out storage processing.
Another aspect of the application provides a kind of managing device of container mirror image data, comprising:
Second receiving module, for obtaining the first test request of general pipeline server transmission;
Second sending module, for sending the first test data packet to runtime server according to first test request,
To trigger the runtime server to first test result information of general pipeline server feedback;
Second receiving module, if being also used to the data management server is confirmed as first object data management clothes
Business device then receives the mirror image data scheduling request that the general pipeline server is sent;
Second processing module, for carrying out response processing according to the mirror image data scheduling request
According to device as described above, optionally, the Second processing module is specifically used for according to the mirror image data tune
Degree request, is sent to runtime server for corresponding target mirror image data, so that the runtime server runs the target mirror
As data are to create container.
According to device as described above, optionally, second receiving module is also used to obtain the transmission of general pipeline server
Second test request;
Second sending module is also used to send the second test number to runtime server according to second test request
According to packet, to trigger the runtime server to second test result information of general pipeline server feedback;
Second receiving module, if being also used to the data management server is confirmed as the second target data management clothes
Business device, then receiving the second communication that the general pipeline server is sent proves;
Second receiving module, be also used to obtain runtime server transmission first communicate proof and new mirror image number
According to;
The Second processing module, is also used to be proved according to the first communication and second communication proves to be verified;
The Second processing module carries out storage processing to the new mirror image data if being also used to be verified;
Second sending module is also used to other data pipes for being not determined to the second target data management server
It manages server and sends communication request, so that other described data management servers are asked according to the pre-receiving that general pipeline server is sent
It asks, to the communication request feedback response information.
According to device as described above, optionally, if getting the response letter of other data management servers feedback
Breath, then second sending module, is also used to, and the new mirror image data is sent to other described data management servers,
To be backed up;
If not getting the response message, second sending module is also used to anti-to the general pipeline server
The information of communication failure is presented, and the new mirror image data is sent to the general pipeline server, so that the general pipeline server
In test to when can communicate with other described data management servers, the new mirror image data is sent to other described numbers
According to management server, to be backed up.
An additional aspect of the present invention provides a kind of management system of container mirror image data, including general pipeline server and/or
At least two data management servers.
Wherein, the general pipeline server includes the managing device of corresponding container mirror image data as described above, the number
It include the managing device of corresponding container mirror image data as described above according to management server.
The management method of container mirror image data provided by the present application, apparatus and system, by using at least two data pipes
Reason server is stored with mirror image data simultaneously, and is responsible for the distribution of scheduler task by a general pipeline server, when general pipeline server
After getting mirror image data scheduling request, transmitted according to each data management server that runtime server returns to runtime server
First test result information of data determines and provides the first object data management server of dispatch service, realizes from least
Select the faster data management server of transmission speed for first object data management server in two data management servers,
Improve the dispatching efficiency of mirror image data, also, ought wherein mirror image data to be scheduled goes wrong in some data management server
When, it is also an option that other data management servers are scheduled, the safety of mirror image data is improved, solves the prior art
In, only one data management server, once mirror image data goes wrong, the corresponding first category vessel of the mirror image data then cannot
The problem of being created.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this Shen
Some embodiments please for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow diagram of the management method for the container mirror image data that one embodiment of the application provides;
Fig. 2 is the flow diagram of the management method for the container mirror image data that another embodiment of the application provides;
Fig. 3 is the flow diagram of the management method for the container mirror image data that the application another embodiment provides;
Fig. 4 is the flow diagram of the management method for the container mirror image data that the another embodiment of the application provides;
Fig. 5 is the structural schematic diagram of the managing device for the container mirror image data that one embodiment of the application provides;
Fig. 6 is the structural schematic diagram of the managing device for the container mirror image data that another embodiment of the application provides;
Fig. 7 is the structural schematic diagram of the management system for the container mirror image data that one embodiment of the application provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall in the protection scope of this application.
Noun involved in the application is explained first:
Container: it is by way of virtualizing operating systems come management code and application program, includes in each container
One whole user environment space exclusively enjoyed, and the variation in a container will not influence the running environment of other containers.
Mirror image data: referring to the statement or description for creating container, and container is the example of mirror image, when creating container,
Pull-up mirror image is needed to be created.
Data management server: it is stored with the server of mirror image data.
Runtime server: need to create the server of container.
Embodiment one
The present embodiment provides a kind of management methods of container mirror image data, for managing mirror image data.The present embodiment is held
Row main body is the managing device of container mirror image data, which can be set in the server, for the ease of distinguishing, the server
It is properly termed as general pipeline server.
As shown in Figure 1, the flow diagram of the management method for container mirror image data provided in this embodiment, this method packet
It includes:
Step 101, mirror image data scheduling request is obtained.
Step 102, according to mirror image data scheduling request, while the first test is sent at least two data management servers
Request, so that data management server sends the first test data packet to runtime server, triggering runtime server takes to general pipeline
Business device feeds back the first test result information.
Wherein, identical mirror image data is stored in each data management server, each data management server is mutually backups
Server.First test result information includes velocity information of the data management server to runtime server transmission data.
Step 103, the first test result information of runtime server feedback is obtained.
Step 104, according to the first test result information, first object data management server is determined.
Step 105, mirror image data scheduling request is sent to first object data management server, so that first object number
Response processing is carried out to mirror image data scheduling request according to management server.
Specifically, then being sent out to general pipeline server when runtime server or other container controlling equipments need to create container
It after sending mirror image data scheduling request, general pipeline server to get the mirror image data scheduling request, is first tested, is determined by which
Data management server responds the mirror image data scheduling request.
Specifically, general pipeline server is according to mirror image data scheduling request, while at least two data management servers
The first test request is sent, after each data management server receives the first test request, then according to the first test request to fortune
Row server sends the first test data packet, after runtime server receives the first test data packet, to general pipeline server feedback
First test result information, first test result information include that each data management server is sent to it test data packet
Transmission speed information, for example, receiving the temporal information etc. for the first test data packet that each data management server is sent.
After general pipeline server gets the first test result information of runtime server feedback, believed according to the first test result
Breath, determines first object data management server.For example, receiving time earliest data management server can then be determined as
First object data management server.After first object management server has been determined, then mirror image data scheduling request is sent
After giving first object data management server, first object data management server to receive the mirror image data scheduling request, then
Processing can be responded thereto.Specifically, first object data management server will be corresponded to according to mirror image data scheduling request
Target mirror image data be sent to runtime server so that runtime server operational objective mirror image data is to create container.
It should be noted that above-mentioned first test request is the survey for requesting to carry out data transmission to each data management server
Examination.Data management server to runtime server send the first test data packet be in order to test current data management server with
The rate that data are transmitted between runtime server.The particular content of first test data packet can be pre-set, can also
Being generated according to mirror image data scheduling request, herein with no restrictions.
The management method of container mirror image data provided in this embodiment, it is same by using at least two data management servers
When be stored with mirror image data, and be responsible for the distribution of scheduler task by a general pipeline server, when general pipeline server gets mirror image
After data scheduling request, the first of data are transmitted to runtime server according to each data management server that runtime server returns
Test result information determines and provides the first object data management server of dispatch service, realizes from least two data pipes
It selects the faster data management server of transmission speed for first object data management server in reason server, improves mirror image number
According to dispatching efficiency can also select also, when mirror image data to be scheduled in wherein some data management server goes wrong
It selects other data management servers to be scheduled, improves the safety of mirror image data, solve in the prior art, only one
Data management server, once mirror image data goes wrong, what the corresponding first category vessel of the mirror image data cannot be then created is asked
Topic.
Embodiment two
The present embodiment does further supplementary explanation to the management method for the container mirror image data that embodiment one provides.
As shown in Fig. 2, the flow diagram of the management method for container mirror image data provided in this embodiment.In above-mentioned reality
On the basis of applying example one, optionally, this method can also include:
Step 201, the storage solicited message for storing new mirror image data that runtime server is sent is obtained.
When runtime server has the demand for storing new mirror image data, then it can send and be stored newly to general pipeline server
The storage solicited message of mirror image data.General pipeline server then available storage solicited message.
Alternatively it is also possible to be that other are responsible for uploading new mirror image data, without run the server of mirror image data to
General pipeline server sends the storage solicited message for storing new mirror image data.
Step 202, according to storage solicited message, while the second test is sent at least two data management servers and is asked
It asks, so that data management server sends the second test data packet to runtime server, triggers runtime server to general pipeline service
Device feeds back the second test result information.
Step 203, the second test result information of runtime server feedback is obtained.
Step 204, according to the second test result information, the second target data management server is determined.
It should be noted that the concrete operations of step 202- step 204, consistent with step 102- step 104, herein no longer
It repeats.It can be identical as the first test request and the first test data packet about the second test request and the second test data packet
It pre-sets, is also possible to according to storage solicited message generation, herein with no restrictions.
Step 205, sending the first communication to runtime server proves, sends second to the second target data management server
Communication proves, so that runtime server proves the first communication and new mirror image data to be stored is sent to the second number of targets
According to management server, trigger the second target data management server proves to prove with the second communication according to the first communication, is tested
After card passes through, storage processing carried out to new mirror image data, and to other numbers in addition to the second target data management server
Communication request is sent according to management server.
After the second target data management server has been determined, general pipeline server can send first to runtime server
Communication proves that sending the second communication to the second target data management server proves that runtime server is receiving the first communication
After proof, the first communication can be proved and new mirror image data to be stored is sent to the second target data management service
Device, received first communication of the second target data management server prove then to be demonstrate,proved according to the first communication with after new mirror image data
Bright and the second communication proves, is verified, for example, the first communication proves and the second communication proves to be secret by the master of encryption respectively
Key and from code key, the second target data management server can obtain two code keys by decryption, and be matched, if matching at
Function is then verified, and is otherwise verified and is not passed through.It is to be appreciated that the first communication proves and the second communication proves to can also be it
His any verification mode that can be implemented, it is not limited here.After being verified, storage processing is carried out to new mirror image data,
And communication request is sent to other data management servers in addition to the second target data management server.
Optionally, while step 205 or later, can also include:
Step 206, pre-receiving is sent to other data management servers in addition to the second target data management server to ask
It asks, so that other data management servers are requested according to pre-receiving, the communication sent to the second target data management server is asked
Seek feedback response information.
Specifically, general pipeline server is sending the first communication proof to runtime server, taken to the second target data management
While device transmission the second communication proof of being engaged in or later, to other data managements clothes in addition to the second target data management server
Device transmission pre-receiving of being engaged in is requested, to notify other data management servers that the event of new mirror image data storage has occurred.Other
Data management server, can be to the second number of targets when receiving the communication request of the second target data management server transmission
According to the response message that management server feedback allows to communicate, the second target data management server is allowed to send out new mirror image data
Give other data management servers.
Optionally, if the second target data management server does not receive the response message of other data management servers,
Then after step 206, can also include:
Step 207, the information for the communication failure that the second target data management server of acquisition is sent and new mirror image number
According to.
Step 208, third test request periodically is sent to other data management servers.
Step 209, when that can communicate with other data management servers, new mirror image data is sent to other data
Management server, so that other data management servers carry out storage processing to new mirror image data.
Specifically, if the second target data management server does not receive the response message of other data management servers,
It then proves not communicated with other data management servers, at this point, the second target data management server then takes to general pipeline
The information of business device feedback communication failure, informs that the second target data of general pipeline server management server cannot be with other data managements
Server communication, therefore new mirror image data can not be sent to other data management servers, the second target data management clothes
New mirror image data then can be sent to general pipeline server by business device, carry out subsequent processing by general pipeline server.General pipeline server
After the information and the new mirror image data that receive the communication failure of the second target data management server transmission, then it can determine
When to other data management servers send third test request, when can receive operational management server feedback third survey
When test result information or when can receive the feedback information of other data management servers, it was demonstrated that general pipeline server can
Communicate with other data management servers, then new mirror image data can be sent to other data pipes by general pipeline server
Server is managed, so that other data management servers carry out storage processing to new mirror image data, to be backed up.
The management method of container mirror image data provided in this embodiment, it is same by using at least two data management servers
When memory image data, and be responsible for by a general pipeline server distribution of store tasks, asked when general pipeline server gets storage
After seeking information, each data management server returned according to runtime server is tied to the second test of runtime server transmission data
Fruit information, determine receive runtime server send new mirror image data the second target data management server, realize from
The faster data management server of transmission speed is selected to receive what runtime server was sent at least two data management servers
New mirror image data, guarantee data transmission efficiency, and backuped in other data management servers by internal data transfer, and
Other data management servers need not be sent to by runtime server again, further improve the efficiency of management, and improve mirror image
The safety of data solves in the prior art, only one data management server, should once mirror image data goes wrong
The problem of corresponding first category vessel of mirror image data cannot be then created.
Embodiment three
The present embodiment provides a kind of management methods of container mirror image data, for managing mirror image data.The present embodiment is held
Row main body is the managing device of container mirror image data, which can be set in the server, for the ease of distinguishing, the server
It is properly termed as data management server.
As shown in figure 3, the flow diagram of the management method for container mirror image data provided in this embodiment, this method packet
It includes:
Step 301, the first test request that general pipeline server is sent is obtained.
Step 302, the first test data packet is sent to runtime server according to the first test request, to trigger operation service
Device is to the first test result information of general pipeline server feedback.
First test result information includes velocity information of the data management server to runtime server transmission data.
Step 303, if data management server is confirmed as first object data management server, general pipeline service is received
The mirror image data scheduling request that device is sent.
Step 304, response processing is carried out according to mirror image data scheduling request.
Optionally, first object data management server is according to mirror image data scheduling request, by corresponding target mirror image number
According to runtime server is sent to, so that runtime server operational objective mirror image data is to create container.
It should be noted that above-mentioned first test request is the survey for requesting to carry out data transmission to each data management server
Examination.Data management server to runtime server send the first test data packet be in order to test current data management server with
The rate that data are transmitted between runtime server.The particular content of first test data packet can be pre-set, can also
Being generated according to mirror image data scheduling request, herein with no restrictions.
It should also be noted that, in the present embodiment the concrete operations of each step carried out in the above-described embodiments it is detailed
Illustrate, details are not described herein.
The management method of container mirror image data provided in this embodiment, it is same by using at least two data management servers
When be stored with mirror image data, and be responsible for the distribution of scheduler task by a general pipeline server, when general pipeline server gets mirror image
After data scheduling request, the first of data are transmitted to runtime server according to each data management server that runtime server returns
Test result information determines and provides the first object data management server of dispatch service, realizes from least two data pipes
It selects the faster data management server of transmission speed for first object data management server in reason server, improves mirror image number
According to dispatching efficiency can also select also, when mirror image data to be scheduled in wherein some data management server goes wrong
It selects other data management servers to be scheduled, improves the safety of mirror image data, solve in the prior art, only one
Data management server, once mirror image data goes wrong, what the corresponding first category vessel of the mirror image data cannot be then created is asked
Topic.
Example IV
The present embodiment does further supplementary explanation to the management method for the container mirror image data that embodiment three provides.
As shown in figure 4, the flow diagram of the management method for container mirror image data provided in this embodiment.In above-mentioned reality
On the basis of applying example three, optionally, this method can also include:
Step 401, the second test request that general pipeline server is sent is obtained.
Step 402, the second test data packet is sent to runtime server according to the second test request, to trigger operation service
Device is to the second test result information of general pipeline server feedback.
It should be noted that can be with the first test request and about the second test request and the second test data packet
One test data packet is mutually all to pre-set, and is also possible to according to storage solicited message generation, herein with no restrictions.
Step 403, if data management server is confirmed as the second target data management server, general pipeline service is received
The second communication that device is sent proves.
Step 404, the first communication proof and new mirror image data that runtime server is sent are obtained.
Step 405, it is proved according to the first communication and the second communication proves to be verified.
Step 406, if being verified, storage processing carried out to new mirror image data, and to being not determined to the second mesh
Other data management servers for marking data management server send communication request, so that other data management servers are according to total
The pre-receiving request that pipe server is sent, to communication request feedback response information.
Optionally, after step 406, this method can also include:
Step 407, if getting the response message of other data management servers feedback, new mirror image data is sent
Other data management servers are given, to be backed up.
Step 408, if not getting response message, to the information of general pipeline server feedback communication failure, and will be new
Mirror image data be sent to general pipeline server so that general pipeline server is in test to can communicate with other data management servers
When, new mirror image data is sent to other data management servers, to be backed up.
It should be noted that the concrete operations of each step have carried out specifically in the above-described embodiments in the present embodiment
Bright, details are not described herein.
The management method of container mirror image data provided in this embodiment, it is same by using at least two data management servers
When memory image data, and be responsible for by a general pipeline server distribution of store tasks, asked when general pipeline server gets storage
After seeking information, each data management server returned according to runtime server is tied to the second test of runtime server transmission data
Fruit information, determine receive runtime server send new mirror image data the second target data management server, realize from
The faster data management server of transmission speed is selected to receive what runtime server was sent at least two data management servers
New mirror image data, guarantee data transmission efficiency, and backuped in other data management servers by internal data transfer, and
Other data management servers need not be sent to by runtime server again, further improve the efficiency of management, and improve mirror image
The safety of data solves in the prior art, only one data management server, should once mirror image data goes wrong
The problem of corresponding first category vessel of mirror image data cannot be then created.
Embodiment five
The present embodiment provides a kind of managing devices of container mirror image data, for executing the container of the offer of above-described embodiment one
The management method of mirror image data.The managing device of the container mirror image data can be set in the server, should for the ease of distinguishing
Server can be described as general pipeline server.
As shown in figure 5, the structural schematic diagram of the managing device for container mirror image data provided in this embodiment.The container mirror
As the managing device 50 of data includes the first acquisition module 51, the first sending module 52, the first receiving module 53 and the first processing
Module 54.
Wherein, the first acquisition module 51 is for obtaining mirror image data scheduling request;First sending module 52 is used for according to mirror
The first test request is sent as data scheduling request, while at least two data management servers, so that data management service
Device sends the first test data packet to runtime server, and triggering runtime server is believed to the first test result of general pipeline server feedback
Breath;First receiving module 53 is used to obtain the first test result information of runtime server feedback;First processing module 54 is used for
According to the first test result information, first object data management server is determined;First sending module 52 is also used to mirror image number
It is sent to first object data management server according to scheduling request, so that first object data management server is to mirror image data tune
Degree request carries out response processing.
Device in this present embodiment is closed, wherein modules execute the concrete mode of operation in related this method
It is described in detail in embodiment, no detailed explanation will be given here.
According to the managing device of container mirror image data provided in this embodiment, by using at least two data management services
Device is stored with mirror image data simultaneously, and is responsible for the distribution of scheduler task by a general pipeline server, when general pipeline server is got
After mirror image data scheduling request, data are transmitted to runtime server according to each data management server that runtime server returns
First test result information determines and provides the first object data management server of dispatch service, realizes from least two numbers
According to selected in management server the faster data management server of transmission speed for first object data management server, improve mirror
As the dispatching efficiency of data, also, when mirror image data to be scheduled in wherein some data management server goes wrong, may be used also
To select other data management servers to be scheduled, the safety of mirror image data is improved, is solved in the prior art, only
One data management server, once mirror image data goes wrong, the corresponding first category vessel of the mirror image data cannot be then created
The problem of.
Embodiment six
The present embodiment does further supplementary explanation to the managing device for the container mirror image data that above-described embodiment five provides.
On the basis of above-described embodiment five, optionally, the first acquisition module 51 is also used to obtain runtime server transmission
The new mirror image data of storage storage solicited message;First sending module 52 is also used to according to storage solicited message, while to
At least two data management servers send the second test request, so that data management server sends second to runtime server
Test data packet triggers runtime server to the second test result information of general pipeline server feedback;First receiving module 53 is also used
In the second test result information for obtaining runtime server feedback;First processing module 54 is also used to be believed according to the second test result
Breath, determines the second target data management server;First sending module 52 is also used to send the first communication card to runtime server
It is bright, to the second target data management server send second communication prove so that runtime server by first communication prove and
New mirror image data to be stored is sent to the second target data management server, triggers the second target data management server root
It is proved according to the first communication and the second communication proves, after being verified, storage processing carried out to new mirror image data, and to removing
Other data management servers except second target data management server send communication request.
Optionally, the first sending module 52 is also used to other data managements in addition to the second target data management server
Server sends pre-receiving request, so that other data management servers are requested according to pre-receiving, to the second target data management
The communication request feedback response information that server is sent.
Optionally, if the second target data management server does not receive the response message of other data management servers,
Then the first receiving module 53 is also used to obtain the information of the communication failure of the second target data management server transmission, and newly
Mirror image data;First sending module 52 is also used to timing and sends third test request to other data management servers;First hair
It send module 52 to be also used to when that can communicate with other data management servers, new mirror image data is sent to other data pipes
Server is managed, so that other data management servers carry out storage processing to new mirror image data.
Device in this present embodiment is closed, wherein modules execute the concrete mode of operation in related this method
It is described in detail in embodiment, no detailed explanation will be given here.
It is same by using at least two data management servers according to the managing device of the container mirror image data of the present embodiment
When memory image data, and be responsible for by a general pipeline server distribution of store tasks, asked when general pipeline server gets storage
After seeking information, each data management server returned according to runtime server is tied to the second test of runtime server transmission data
Fruit information, determine receive runtime server send new mirror image data the second target data management server, realize from
The faster data management server of transmission speed is selected to receive what runtime server was sent at least two data management servers
New mirror image data, guarantee data transmission efficiency, and backuped in other data management servers by internal data transfer, and
Other data management servers need not be sent to by runtime server again, further improve the efficiency of management, and improve mirror image
The safety of data solves in the prior art, only one data management server, should once mirror image data goes wrong
The problem of corresponding first category vessel of mirror image data cannot be then created.
Embodiment seven
The present embodiment provides a kind of managing devices of container mirror image data, for executing the container of the offer of above-described embodiment three
The management method of mirror image data.The managing device of the container mirror image data can be set in the server, should for the ease of distinguishing
Server can be described as data management server.
As shown in fig. 6, the structural schematic diagram of the managing device for container mirror image data provided in this embodiment.The container mirror
As the managing device 70 of data includes the second receiving module 71, the second sending module 72 and Second processing module 73.
Wherein, the second receiving module 71 is used to obtain the first test request of general pipeline server transmission;Second sending module
72, for sending the first test data packet to runtime server according to the first test request, are taken with triggering runtime server to general pipeline
Business device feeds back the first test result information;If the second receiving module 71 is also used to data management server and is confirmed as first object
Data management server then receives the mirror image data scheduling request of general pipeline server transmission;Second processing module 73 is used for basis
Mirror image data scheduling request carries out response processing.
Device in this present embodiment is closed, wherein modules execute the concrete mode of operation in related this method
It is described in detail in embodiment, no detailed explanation will be given here.
According to the managing device of container mirror image data provided in this embodiment, by using at least two data management services
Device is stored with mirror image data simultaneously, and is responsible for the distribution of scheduler task by a general pipeline server, when general pipeline server is got
After mirror image data scheduling request, data are transmitted to runtime server according to each data management server that runtime server returns
First test result information determines and provides the first object data management server of dispatch service, realizes from least two numbers
According to selected in management server the faster data management server of transmission speed for first object data management server, improve mirror
As the dispatching efficiency of data, also, when mirror image data to be scheduled in wherein some data management server goes wrong, may be used also
To select other data management servers to be scheduled, the safety of mirror image data is improved, is solved in the prior art, only
One data management server, once mirror image data goes wrong, the corresponding first category vessel of the mirror image data cannot be then created
The problem of.
Embodiment eight
The present embodiment does further supplementary explanation to the managing device for the container mirror image data that above-described embodiment seven provides.
On the basis of above-described embodiment seven, optionally, Second processing module 73 is specifically used for being dispatched according to mirror image data
Request, is sent to runtime server for corresponding target mirror image data, so that runtime server operational objective mirror image data is to create
Build container.
On the basis of above-described embodiment seven, optionally, the second receiving module 71 is also used to obtain the transmission of general pipeline server
The second test request;Second sending module 72 is also used to send the second test number to runtime server according to the second test request
According to packet, to trigger runtime server to the second test result information of general pipeline server feedback;If the second receiving module 71 is also used to
Data management server is confirmed as the second target data management server, then receives the second communication card of general pipeline server transmission
It is bright;The first communication that second receiving module 71 is also used to obtain runtime server transmission proves and new mirror image data;Second
Processing module 73 is also used to be proved according to the first communication and the second communication proves to be verified;If Second processing module 73 is also used to
It is verified, then storage processing is carried out to new mirror image data;Second sending module 72 is also used to being not determined to the second mesh
Other data management servers for marking data management server send communication request, so that other data management servers are according to total
The pre-receiving request that pipe server is sent, to communication request feedback response information.
Optionally, if getting the response message of other data management servers feedback, the second sending module 72 is also used
In new mirror image data is then sent to other data management servers, to be backed up;If not getting response message,
Second sending module 72 is also used to the information to general pipeline server feedback communication failure, and new mirror image data is sent to general pipeline
Server sends out new mirror image data so that general pipeline server is being tested to when can communicate with other data management servers
Other data management servers are given, to be backed up.
Device in this present embodiment is closed, wherein modules execute the concrete mode of operation in related this method
It is described in detail in embodiment, no detailed explanation will be given here.
It is same by using at least two data management servers according to the managing device of the container mirror image data of the present embodiment
When memory image data, and be responsible for by a general pipeline server distribution of store tasks, asked when general pipeline server gets storage
After seeking information, each data management server returned according to runtime server is tied to the second test of runtime server transmission data
Fruit information, determine receive runtime server send new mirror image data the second target data management server, realize from
The faster data management server of transmission speed is selected to receive what runtime server was sent at least two data management servers
New mirror image data, guarantee data transmission efficiency, and backuped in other data management servers by internal data transfer, and
Other data management servers need not be sent to by runtime server again, further improve the efficiency of management, and improve mirror image
The safety of data solves in the prior art, only one data management server, should once mirror image data goes wrong
The problem of corresponding first category vessel of mirror image data cannot be then created.
Embodiment nine
The present embodiment provides a kind of management systems of container mirror image data, the appearance provided for executing any of the above-described embodiment
The management method of device mirror image data.
As shown in fig. 7, the structural schematic diagram of the management system for container mirror image data provided in this embodiment.The container mirror
As the management system 90 of data includes general pipeline server 91 or at least two data management servers 92 or the container mirror image number
According to management system 90 include general pipeline server 91 and at least two data management servers 92, specifically can be two, three
Or it is multiple.Can according to the actual situation, considering cost is configured, herein with no restrictions.
Wherein, general pipeline server 91 includes the management dress of the container mirror image data of above-described embodiment five or the offer of embodiment six
50 are set, data management server 92 includes the managing device of the container mirror image data of above-described embodiment seven or the offer of embodiment eight
70。
Optionally, the management system of the container mirror image data can also include runtime server.
Device in this present embodiment is closed, wherein modules execute the concrete mode of operation in related this method
It is described in detail in embodiment, no detailed explanation will be given here.
According to the management system of container mirror image data provided in this embodiment, by using at least two data management services
Device is stored with mirror image data simultaneously, and is responsible for the distribution of scheduler task by a general pipeline server, realizes from least two numbers
According to selected in management server the faster data management server of transmission speed for first object data management server, improve mirror
As the dispatching efficiency of data, also, when mirror image data to be scheduled in wherein some data management server goes wrong, may be used also
To select other data management servers to be scheduled, the safety of mirror image data is improved, is solved in the prior art, only
One data management server, once mirror image data goes wrong, the corresponding first category vessel of the mirror image data cannot be then created
The problem of.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed;And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
The various media that can store program code such as disk.
Finally, it should be noted that above embodiments are only to illustrate the technical solution of the application, rather than its limitations;Although
The application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features;
And these are modified or replaceed, the range of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution.
Claims (13)
1. a kind of management method of container mirror image data characterized by comprising
General pipeline server obtains mirror image data scheduling request;
The first test request is sent according to the mirror image data scheduling request, while at least two data management servers, with
So that the data management server is sent the first test data packet to runtime server, triggers the runtime server to described total
The first test result information of pipe server feedback;
Obtain the first test result information of the runtime server feedback;
According to first test result information, first object data management server is determined;
The mirror image data scheduling request is sent to the first object data management server, so that the first object number
Response processing is carried out to the mirror image data scheduling request according to management server;
Obtain the storage solicited message for storing new mirror image data that runtime server is sent;
The second test request is sent according to the storage solicited message, while at least two data management servers, with
So that the data management server is sent the second test data packet to runtime server, triggers the runtime server to described total
The second test result information of pipe server feedback;
Obtain the second test result information of the runtime server feedback;
According to second test result information, the second target data management server is determined;
Sending the first communication to the runtime server proves, Xiang Suoshu the second target data management server sends the second communication
It proves, so that the runtime server proves the first communication and new mirror image data to be stored is sent to second mesh
Data management server is marked, trigger the second target data management server proves and described second according to first communication
Communication proves, after being verified, carries out storage processing to the new mirror image data, and to except second target data
Other data management servers except management server send communication request.
2. the method according to claim 1, wherein to the runtime server send first communication prove,
After sending the second communication proof to the second target data management server, the method also includes:
Pre-receiving request is sent to other data management servers in addition to the second target data management server, so that institute
It states other data management servers to be requested according to the pre-receiving, to described in the second target data management server transmission
Communication request feedback response information.
3. according to the method described in claim 2, it is characterized in that, if the second target data management server does not receive
The response message of other data management servers, then the method also includes:
Obtain the communication failure that the second target data management server is sent information and the new mirror image data;
Timing sends third test request to other described data management servers;
When that can communicate with other described data management servers, the new mirror image data is sent to other described data
Management server, so that other described data management servers carry out storage processing to the new mirror image data.
4. a kind of management method of container mirror image data characterized by comprising
Data management server obtains the first test request that general pipeline server is sent;
According to first test request to runtime server send the first test data packet, with trigger the runtime server to
First test result information of general pipeline server feedback;
If the data management server is confirmed as first object data management server, the general pipeline server hair is received
The mirror image data scheduling request sent;
Response processing is carried out according to the mirror image data scheduling request;
Obtain the second test request that general pipeline server is sent;
According to second test request to runtime server send the second test data packet, with trigger the runtime server to
Second test result information of general pipeline server feedback;
If the data management server is confirmed as the second target data management server, the general pipeline server hair is received
The second communication sent proves;
Obtain the first communication proof and new mirror image data that runtime server is sent;
It is proved according to the first communication and second communication proves to be verified;
If being verified, storage processing carried out to the new mirror image data, and to being not determined to the second target data pipe
Other data management servers for managing server send communication request, so that other described data management servers take according to general pipeline
The pre-receiving that device is sent of being engaged in is requested, to the communication request feedback response information.
5. according to the method described in claim 4, it is characterized in that, described responded according to the mirror image data scheduling request
Processing, comprising:
According to the mirror image data scheduling request, corresponding target mirror image data is sent to runtime server, so that the fortune
Row server runs the target mirror image data to create container.
If 6. according to the method described in claim 4, it is characterized in that, being verified, to the new mirror image data into
Row storage processing, and send communication to other data management servers for being not determined to the second target data management server and ask
After asking, the method also includes:
If getting the response message of other data management servers feedback, the new mirror image data is sent to institute
Other data management servers are stated, to be backed up;
If not getting the response message, to the information of the general pipeline server feedback communication failure, and will be described new
Mirror image data be sent to the general pipeline server so that the general pipeline server is in test to can be with other described data pipes
When managing server communication, the new mirror image data is sent to other described data management servers, to be backed up.
7. a kind of managing device of container mirror image data characterized by comprising
First obtains module, for obtaining mirror image data scheduling request;
First sending module, for being sent out according to the mirror image data scheduling request, while at least two data management servers
The first test request is sent, so that the data management server sends the first test data packet to runtime server, described in triggering
Runtime server is to the first test result information of general pipeline server feedback;
First receiving module, for obtaining the first test result information of the runtime server feedback;
First processing module, for determining first object data management server according to first test result information;
First sending module is also used to for the mirror image data scheduling request being sent to the first object data management clothes
Business device, so that the first object data management server carries out response processing to the mirror image data scheduling request;
Described first obtains module, is also used to obtain the storage request letter for storing new mirror image data of runtime server transmission
Breath;
First sending module is also used to according to the storage solicited message, while being taken at least two data managements
Business device sends the second test request;So that the data management server sends the second test data packet, touching to runtime server
Send out runtime server described to second test result information of general pipeline server feedback;
First receiving module is also used to obtain the second test result information of the runtime server feedback;
The first processing module is also used to determine the second target data management service according to second test result information
Device;
First sending module, being also used to send the first communication to the runtime server proves, the second number of targets of Xiang Suoshu
Sending the second communication according to management server proves, so that the runtime server proves the first communication and to be stored new
Mirror image data is sent to the second target data management server, triggers the second target data management server according to institute
Stating the first communication proves to prove with second communication, after being verified, carries out at storage to the new mirror image data
Reason, and communication request is sent to other data management servers in addition to the second target data management server.
8. device according to claim 7, which is characterized in that first sending module is also used to except described second
Other data management servers outside target data management server send pre-receiving request, so that other described data managements take
Business device is requested according to the pre-receiving, the communication request feedback response sent to the second target data management server
Information.
9. device according to claim 8, which is characterized in that if the second target data management server does not receive
The response message of other data management servers, then first receiving module, is also used to obtain second number of targets
Information and the new mirror image data according to the communication failure of management server transmission;
First sending module is also used to timing to other described data management servers and sends third test request;
First sending module is also used to when that can communicate with other described data management servers, by the new mirror
As data be sent to described in other data management servers so that other described data management servers are to the new mirror image number
According to carrying out storage processing.
10. a kind of managing device of container mirror image data characterized by comprising
Second receiving module, for obtaining the first test request of general pipeline server transmission;
Second sending module, for sending the first test data packet to runtime server according to first test request, with touching
Send out runtime server described to first test result information of general pipeline server feedback;
Second receiving module, if being also used to data management server is confirmed as first object data management server,
Receive the mirror image data scheduling request that the general pipeline server is sent;
Second processing module, for carrying out response processing according to the mirror image data scheduling request;
Second receiving module is also used to obtain the second test request of general pipeline server transmission;
Second sending module is also used to send the second test data to runtime server according to second test request
Packet, to trigger the runtime server to second test result information of general pipeline server feedback;
Second receiving module, if being also used to the data management server is confirmed as the second target data management service
Device, then receiving the second communication that the general pipeline server is sent proves;
Second receiving module, be also used to obtain runtime server transmission first communicate proof and new mirror image data;
The Second processing module, is also used to be proved according to the first communication and second communication proves to be verified;
The Second processing module carries out storage processing to the new mirror image data if being also used to be verified;
Second sending module is also used to take to other data managements for being not determined to the second target data management server
Business device sends communication request, so that other described data management servers are requested according to the pre-receiving that general pipeline server is sent, it is right
The communication request feedback response information.
11. device according to claim 10, which is characterized in that the Second processing module is specifically used for according to
Corresponding target mirror image data is sent to runtime server by mirror image data scheduling request, so that the runtime server is run
The target mirror image data is to create container.
12. device according to claim 10, which is characterized in that if getting other data management servers feedback
Response message, then second sending module, is also used to, and the new mirror image data is sent to other described data pipes
Server is managed, to be backed up;
If not getting the response message, second sending module is also used to logical to the general pipeline server feedback
Believe the information of failure, and the new mirror image data is sent to the general pipeline server, so that the general pipeline server is being surveyed
When trying to communicate with other described data management servers, the new mirror image data is sent to other described data pipes
Server is managed, to be backed up.
13. a kind of management system of container mirror image data characterized by comprising general pipeline server and at least two data pipes
Manage server;
Wherein, the general pipeline server includes such as the described in any item devices of claim 7-9, the data management server packet
It includes such as the described in any item devices of claim 10-12.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710922809.3A CN107707654B (en) | 2017-09-30 | 2017-09-30 | The management method of container mirror image data, apparatus and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710922809.3A CN107707654B (en) | 2017-09-30 | 2017-09-30 | The management method of container mirror image data, apparatus and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107707654A CN107707654A (en) | 2018-02-16 |
CN107707654B true CN107707654B (en) | 2019-05-14 |
Family
ID=61184429
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710922809.3A Active CN107707654B (en) | 2017-09-30 | 2017-09-30 | The management method of container mirror image data, apparatus and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107707654B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112988321A (en) * | 2019-12-13 | 2021-06-18 | 北京金山云网络技术有限公司 | Container mirror image data management method and device, electronic equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102132255A (en) * | 2008-05-29 | 2011-07-20 | 思杰系统有限公司 | Systems and methods for load balancing via a plurality of virtual servers upon failover using metrics from a backup virtual server |
CN102891894A (en) * | 2012-10-17 | 2013-01-23 | 中国工商银行股份有限公司 | Caching method used for server cluster, cache server and cache system |
CN103684871A (en) * | 2013-12-25 | 2014-03-26 | 乐视网信息技术(北京)股份有限公司 | Method for monitoring operation and maintenance asset state, method and system for updating operation and maintenance configuration information |
CN105554121A (en) * | 2015-12-18 | 2016-05-04 | 深圳中兴网信科技有限公司 | Method and system for realizing load equalization of distributed cache system |
CN106411585A (en) * | 2016-09-21 | 2017-02-15 | 深圳市智物联网络有限公司 | Server switching method, adapter, server and scheduling device |
CN106850759A (en) * | 2016-12-31 | 2017-06-13 | 广州勤加缘科技实业有限公司 | MySQL database clustering methods and its processing system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10282246B2 (en) * | 2014-01-14 | 2019-05-07 | Sk Techx Co., Ltd. | Application error detection method for cloud streaming service, and apparatus and system therefor |
-
2017
- 2017-09-30 CN CN201710922809.3A patent/CN107707654B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102132255A (en) * | 2008-05-29 | 2011-07-20 | 思杰系统有限公司 | Systems and methods for load balancing via a plurality of virtual servers upon failover using metrics from a backup virtual server |
CN102891894A (en) * | 2012-10-17 | 2013-01-23 | 中国工商银行股份有限公司 | Caching method used for server cluster, cache server and cache system |
CN103684871A (en) * | 2013-12-25 | 2014-03-26 | 乐视网信息技术(北京)股份有限公司 | Method for monitoring operation and maintenance asset state, method and system for updating operation and maintenance configuration information |
CN105554121A (en) * | 2015-12-18 | 2016-05-04 | 深圳中兴网信科技有限公司 | Method and system for realizing load equalization of distributed cache system |
CN106411585A (en) * | 2016-09-21 | 2017-02-15 | 深圳市智物联网络有限公司 | Server switching method, adapter, server and scheduling device |
CN106850759A (en) * | 2016-12-31 | 2017-06-13 | 广州勤加缘科技实业有限公司 | MySQL database clustering methods and its processing system |
Also Published As
Publication number | Publication date |
---|---|
CN107707654A (en) | 2018-02-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102319172B1 (en) | Platform providing system based on cloud computing and platform provision method using the same | |
EP3618352B1 (en) | Virtual machine management | |
CN111931949A (en) | Communication in a federated learning environment | |
US9634848B2 (en) | Message broadcasting in a clustered computing environment | |
CN104932941B (en) | A kind of distributed message processing system and device and method therein | |
US8307054B2 (en) | Systems and/or methods for location transparent routing and execution of processes | |
CA3077225A1 (en) | Methods for improved web application testing using remote headless browsers and devices thereof | |
CN103312801A (en) | Application devices, interaction method and system for data between application devices and server | |
CN105979007A (en) | Acceleration resource processing method and device and network function virtualization system | |
CN105897805A (en) | Method and device for cross-layer scheduling of resources of data center with multi-layer architecture | |
US10691501B1 (en) | Command invocations for target computing resources | |
KR20190028210A (en) | Cloud service method and system for deployment of artificial intelligence application using container | |
EP4302189A1 (en) | Serverless runtime container allocation | |
CN113672352A (en) | Method and device for deploying federated learning task based on container | |
US11586470B2 (en) | Scalable workflow engine with a stateless orchestrator | |
CN110012050A (en) | Message Processing, storage method, apparatus and system | |
US10585730B1 (en) | Enqueue-related processing based on timing out of an attempted enqueue | |
CN107707654B (en) | The management method of container mirror image data, apparatus and system | |
CN115349117B (en) | Multi-level cache grid system for multi-tenant, serverless environments | |
US20150067097A1 (en) | Managing data distribution to networked client computing devices | |
CN111126604B (en) | Model training method, device, server and storage medium | |
CN114598700B (en) | Communication method and communication system | |
CN114721743B (en) | Task execution method and device and electronic equipment | |
CN110413398A (en) | Method for scheduling task, device, computer equipment and storage medium | |
JP2018514023A (en) | Content deployment, scaling and telemetry |
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 |