CN112799818A - Cloud fusion task migration method and system based on check point description file - Google Patents
Cloud fusion task migration method and system based on check point description file Download PDFInfo
- Publication number
- CN112799818A CN112799818A CN202110149647.0A CN202110149647A CN112799818A CN 112799818 A CN112799818 A CN 112799818A CN 202110149647 A CN202110149647 A CN 202110149647A CN 112799818 A CN112799818 A CN 112799818A
- Authority
- CN
- China
- Prior art keywords
- task
- cloud
- file
- description
- checkpoint
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5021—Priority
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/508—Monitor
Abstract
The invention provides a cloud fusion task migration method and system based on a checkpoint description file, which comprises the following steps: the system comprises a check point description file component, mobile internet equipment and a cloud component; the description of the checkpoint description file component to the functions and requirements of a task running on the mobile internet of things device is manually filled in by a user or generated through a template; the mobile Internet of things equipment carries out actual operation of the task, and when the cloud issues the task and the check point description file thereof, the mobile Internet of things equipment executes the task and generates a data backup file and a state collection file; the cloud part is connected with all the mobile Internet of things devices; and the cloud part is responsible for monitoring the actual running condition after the task is issued. The invention effectively describes the key information needed to be used in the execution and migration of the task by using the checkpoint description file, and a user can customize the service quality requirement, the task scheduling and the backup strategy of the task by filling the checkpoint description file.
Description
Technical Field
The invention relates to the technical field of mobile edge computing, in particular to a cloud fusion task migration method and system based on a check point description file, and particularly relates to a cloud fusion task migration method based on a check point description file and oriented to a mobile edge computing scene.
Background
With the continuous development of the mobile internet of things technology, the performance of mobile internet of things equipment is continuously improved, and a lot of computing and sensing equipment can be carried on the current mobile equipment at a lower cost, so that people put forward higher requirements on the mobile internet of things equipment. In the traditional cloud computing paradigm, the internet of things equipment is only responsible for acquiring sensor data, and all acquired sensor data need to be sent to the cloud end and processed at the cloud end, so that the pressures of cloud end network transmission, data storage and data computing are increased, the computing capacity of the mobile internet of things equipment is wasted, and the service capacity of the mobile internet of things equipment is limited. In the current 'cloud side end' fusion system, the cloud end can issue a task to the Internet of things equipment, and the Internet of things equipment migrates the task of data processing and calculation to the edge node, so that the pressure of the cloud end is reduced, the delay is reduced, and the service quality is improved.
However, migrating the task to the edge node for calculation also brings additional problems, because the performance of the edge calculation device is far inferior to that of the server cluster in the cloud, the quality of service is difficult to guarantee; the stability of the edge node is not reliable enough, similar to a common computer, the situation of long-time downtime or software crash can occur, the downtime probability of the cloud server is very small, and meanwhile, the complete disaster recovery service is provided. For the problems, a plurality of strategies for task migration and replacement are proposed, and some of the strategies are based on prediction to carry out load balancing in advance so as to prevent the edge nodes from being over stressed; some of the tasks are migrated to other devices to be executed when the service quality is reduced or the tasks are wrong, so as to ensure that the tasks are successfully completed.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a cloud fusion task migration method and system based on a checkpoint description file.
The invention provides a cloud fusion task migration system based on a checkpoint description file, which comprises: the system comprises a check point description file component, mobile internet equipment and a cloud component;
the description of the checkpoint description file component to the functions and requirements of a task running on the mobile internet of things device is manually filled in by a user or generated through a template;
the mobile Internet of things equipment carries out actual operation of the task, and after the cloud issues the task and the check point description file thereof to the specific equipment, the mobile Internet of things equipment executes the task and generates a data backup file and a state collection file;
the cloud part is connected with all the mobile Internet of things devices through a network;
the cloud part is mainly responsible for monitoring the actual running condition after the task is issued;
when the cloud detects that the task cannot be completed on time, the cloud part can actively perform task migration scheduling so as to ensure the completion rate of the task.
Preferably, the checkpoint tracing the file component comprises:
quality of service description: the user's requirements for the task are mainly task execution time requirements;
task load description: describing the load type of the task processing;
preferably, the checkpoint profiling the file component further comprises:
description of backup strategy: describes at which moments the task should backup the data in what manner.
Preferably, the mobile internet of things device includes:
a task execution module: according to the service quality description content, scheduling matched resources at the equipment end to execute the task;
a state collection module: and according to the task load description, collecting the load content processing state designated in the task execution, and generating a task state file.
Preferably, the mobile internet of things device further comprises:
the data backup module: and saving the task execution state according to the data backup strategy, generating a data backup file, and sending the data backup file to the cloud.
Preferably, the cloud component comprises:
a task migration module: a state collection module of a receiving Internet of things equipment end acquires current task state information, and migration scheduling of tasks is carried out according to the task state information;
a calculation module: when the task migration module requires to migrate part of task loads to the cloud, the computing module is responsible for processing the loads.
Preferably, the cloud component further comprises:
a task recovery module: and receiving state backup data uploaded by a data backup module of the equipment end of the Internet of things, and when the task needs to be restarted after being migrated, the task recovery module is responsible for recovering the execution state of the task according to the backup data.
A cloud fusion task migration system method based on a checkpoint description file adopts a cloud fusion task migration system based on a checkpoint description file, and comprises the following steps:
step S1: acquiring control information according to current and historical load processing rates, acquiring the current and historical load processing rates, and predicting the execution rate of the task in the period from the next moment to the time limit of task completion by using a long-short term memory neural network;
step S2: accumulating the predicted execution rates, if the ratio of the predicted execution rates to the residual load capacity is lower than a threshold value, marking the task as incapable of being completed in time, and acquiring information of a result of marking incapable of being completed in time;
step S3: according to the information that the marking result cannot be completed in time, if the load of the task can be processed in parallel, transferring part of the load to a cloud end for execution;
and if the load of the task cannot be processed in parallel, recovering the task execution state at the cloud end by using the state backup data of the task, and re-issuing the recovered task to the cloud end or other Internet of things equipment ends.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the invention, the checkpoint description file is utilized to effectively describe the key information required to be used in the execution and migration of the task, and a user can customize the service quality requirement, task scheduling and backup strategy of the task in a mode of filling the checkpoint description file without modifying a service program running on the mobile Internet of things equipment terminal;
2. the long-short term memory neural network model can better process data contents related to historical information, can accurately identify tasks which cannot be completed on time, and can migrate the tasks to the cloud or other nodes in advance so as to ensure the smooth completion of the tasks;
3. the invention backups the program running state, and effectively performs task rollback and redistribution when the task fails, thereby reducing the cost of task restart and coping with the high task failure rate possibly caused by high mobility of the mobile Internet of things equipment.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
fig. 1 is a schematic diagram of a technical solution architecture of the present invention.
Fig. 2 is a schematic diagram of a specific process of the task migration policy in the present invention.
FIG. 3 is a schematic diagram illustrating a task migration strategy according to the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
As shown in fig. 1, the technical scheme of the invention is mainly divided into three parts:
the checkpoint description file: a file describing a task execution policy;
a mobile internet device: a carrier of task execution;
cloud: and the system is responsible for monitoring and migrating tasks to ensure the quality of service.
Specifically, the checkpoint description file includes the following parts:
quality of service description: this paragraph describes the user's requirements for the task, where the requirements refer primarily to the time limit for task completion, and may also include requirements for task security, task priority, and other customized requirements for the geographic environment in which the task is executed. The content is generated by filling in by the user before the task is issued.
Task load description: the content describes the type of load content to be processed in a program corresponding to the task, and the load type includes load content which can be processed in parallel, such as image processing, matrix operation, voice recognition and the like, and also includes load content which needs to be executed in series, such as serial numerical calculation, screen display, remote sensing task and the like. Specifically, the load type needs to include a load content name, whether the load can be processed in parallel, and interface information obtained by the load amount and the load margin. This portion of content is provided by the programmer at the time of task programming.
Description of backup strategy: the time for performing state backup and the method for performing backup of a program corresponding to a task are described. The backup time is usually a time sequence with fixed time intervals, namely, the program state is backed up at fixed time intervals, the sequence is generated according to the task execution time requirement and the task importance degree description in the user service quality description, for important tasks and contents with long running time, the backup frequency of the tasks is more, and for tasks with short running time and unimportant tasks, the backup frequency of the tasks is less. The backup method describes a method for state backup, and comprises the following steps: uploading the calculation results of the calculated part in the load; selecting and storing important state parameters in a program; completely copying and compressing all memories occupied by the process; and generating a snapshot for the container where the task is located.
The mobile Internet of things equipment is provided with a plurality of application programs, the programs are registered in the cloud in a service mode, after a user submits a task in the cloud, the cloud selects and calls appropriate mobile Internet of things equipment to execute the task, meanwhile, a check point description file is submitted to the equipment, and the equipment executes the task according to the called task content and the check point description file. The mobile internet of things device is specifically three modules, including:
a task execution module: the module is responsible for executing a task program, when a task request arrives, the requested task is completed by using a program written in the device storage in advance, and the module selects the content of program execution by reading the service quality description in the checkpoint description file: the program with high safety requirement is independently calculated by using a computing device meeting the safety requirement; providing non-shareable computing resources for higher priority programs. The module provides an interface query function for other modules to query the task execution state.
A state collection module: the module acquires an interface related to a load state in the task execution module by reading task load description in the checkpoint description file, queries the interface and the task state interface when the cloud requests the running state of the task from the equipment end, and combines the load type, the load allowance, the load capacity, the current program running time and the program state into a task state file to be sent to the cloud. The program state specifically refers to normal operation or error interruption of the program.
The data backup module: the module refers to the backup strategy description in the checkpoint description file and processes the specified data content according to the specified processing method, which is already described in the backup strategy description part. And the generated data backup file is sent to a task recovery module at the cloud end through a network.
The cloud end main function is through collecting the state of task execution to task migration action is made in view of the above, specifically divide into three modules:
a task recovery module: and when the task unloading module requires to execute task recovery, a new task is recovered through the stored task execution state file and is handed over to the computing module, and the computing module is responsible for the subsequent execution of the task.
A calculation module: and the recovery module is responsible for executing the program, and when the task after restarting and recovery is sent to the computing module by the recovery module, the computing module can independently complete the task or send the task to other available mobile Internet of things devices. When the task migration module requires to migrate part of loads of the mobile internet of things device end to the cloud end for execution, the computing module is responsible for processing the loads.
A task migration module: and analyzing the program running state sent by the state collection module to provide a task migration decision.
The specific flow of task migration decision is shown in fig. 2, and the content of each step will be explained in detail here:
acquiring an operation state: and for each running task, sending a request to a mobile Internet of things equipment end state collection module at a fixed time interval from the beginning of running to acquire all states related to the running of the task.
And (4) viewing the process state: by checking the process state item in the running state data, if the process state item displays program interruption, a task restarting flow needs to be started, and if the process state item is normally executed, whether the service quality requirement can be met needs to be checked.
Checking whether a backup exists: the task recovery module is inquired whether the task has a backup or not, if the backup exists, the task recovery module is required to perform task state recovery, otherwise, a same task is newly established.
And (4) restoring the program state: the task migration module requests the task recovery module to recover the task state, and the task migration module returns a task which is recovered to the specified running state.
And (4) new task establishment: and a new task is created according to the description of the original task.
And (3) re-issuing the task: and issuing the task generated in the task new establishment or task state recovery stage to proper equipment, if the cloud has enough capacity to execute the task, preferably issuing the task to the cloud, and otherwise, searching other available mobile edge computing equipment for issuing the task.
Load handling speed was predicted using long-short term memory neural networks: the step mainly checks whether the program can be completed in time, the invention uses a method of a long-short term memory neural network to predict the task execution rate, and the specific flow is as follows:
1. firstly, the cloud end needs to perform data statistics on a processing rate change curve of each load on the internet of things equipment in advance, and a long-term and short-term memory network is used for training an execution rate model for each load according to the data. The structure used by the model is an 'encoding-decoding' model framework, and the trained model is f and is based on the load processing rate [ v ] from the starting time to the t-1 moment0…vt-1]Prediction vtBy all execution rates vt,…]During training, only the part before the actual completion time of the task is intercepted to calculate the loss function.
2. Secondly, after the task runs for a period of time and the cloud accumulates m moments of real-time load processing rate data, the data are used as input and transmitted into a model f, and a pre-measured result [ v ] is obtainedm+1,…,ve]. Wherein v iseRepresents the time teRunning rate of time teIs the execution deadline requirement in the task quality of service requirement.
3. Thirdly, accumulating the execution rate predicted value between the time limit of the task execution at the moment, if the ratio of the execution rate predicted value to the residual load is lower than a threshold value, the task needs to be migrated, otherwise, the task does not need to be migrated, and the residual load is recorded as lremainThe threshold is e, and the specific formula is as follows:
checking the load type: the task migration method is a first step of a task migration process, whether task loads can be processed in parallel or not needs to be checked, if the task loads can be processed in parallel, part of the task loads are unloaded to a computing module of a cloud end, and the cloud end and a device end execute tasks at the same time so as to meet execution time requirements. For the task load which cannot be migrated, the task flow needs to be restarted, and the step of checking whether a backup stage exists is skipped.
Migrating partial load: and migrating a part of task load to the cloud according to the task execution state. The specific load calculation method needing migration is as follows:
recording the residual load as lremainLoad amount which can be processed before task completion time limit and is obtained by prediction in the task execution rate prediction step
Is marked aspre. Calculating a predicted credibility factor epsilon, wherein the credibility factor is used for describing the matching degree of the current task execution rate and the model, and the calculation method is to use the first n moments [ v ] of the m moments0,…,vn]To predict the time [ v ] aftern,…,vn]And calculating a loss function of the prediction result and the actual running speed, dividing the loss function into a plurality of grades according to the ratio of the loss function and the error value during model training, wherein the loss function is 1 when the reliability is higher, and the worse the reliability is, the closer to 0. The equivalent of the extra load per unit load due to migration is recorded as llossThe value is statistically obtained in advance on the internet of things device, and represents that the computing resource occupied by transmitting each unit load is equivalent to the computing resource occupied by processing how many loads. According to the information, calculating the migration quantity lsendThe formula is as follows:
according to the invention, the checkpoint description file is utilized to effectively describe the key information required to be used in the execution and migration of the task, and a user can customize the service quality requirement, task scheduling and backup strategy of the task in a mode of filling the checkpoint description file without modifying a service program running on the mobile Internet of things equipment terminal;
the long-short term memory neural network model can better process data contents related to historical information, can accurately identify tasks which cannot be completed on time, and can migrate the tasks to the cloud or other nodes in advance so as to ensure the smooth completion of the tasks;
the invention backups the program running state, and effectively performs task rollback and redistribution when the task fails, thereby reducing the cost of task restart and coping with the high task failure rate possibly caused by high mobility of the mobile Internet of things equipment.
In the description of the present application, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience in describing the present application and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present application.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims.
Claims (8)
1. A cloud fusion task migration system based on a checkpoint description file is characterized by comprising: the system comprises a check point description file component, mobile internet equipment and a cloud component;
the description of the checkpoint description file component to the functions and requirements of a task running on the mobile internet of things device is manually filled in by a user or generated through a template;
the mobile Internet of things equipment carries out actual operation of the task, and when the cloud issues the task and the check point description file thereof, the mobile Internet of things equipment executes the task and generates a data backup file and a state collection file;
the cloud part is connected with all the mobile Internet of things devices;
the cloud end part is responsible for monitoring the actual running condition after the task is issued;
the cloud component can actively perform task migration scheduling.
2. The cloud-based converged task migration system according to claim 1, wherein the checkpointing the file component comprises:
quality of service description: the user's requirements for the task are mainly task execution time requirements;
task load description: the load type of the set segment task processing is described.
3. The cloud-based converged task migration system according to claim 2, wherein the checkpoint description file component further comprises:
description of backup strategy: describes at which moments the task should backup the data in what manner.
4. The cloud fusion task migration system based on checkpoint description files as claimed in claim 1, wherein the mobile internet of things device comprises:
a task execution module: according to the service quality description content, scheduling matched resources at the equipment end to execute the task;
a state collection module: and according to the task load description, collecting the load content processing state designated in the task execution, and generating a task state file.
5. The cloud fusion task migration system based on checkpoint description files as claimed in claim 4, wherein the mobile internet of things device further comprises:
the data backup module: and saving the task execution state according to the data backup strategy, generating a data backup file, and sending the data backup file to the cloud.
6. The checkpoint-description-file-based cloud fusion task migration system of claim 1, wherein the cloud component comprises:
a task migration module: a state collection module of a receiving Internet of things equipment end acquires current task state information, and migration scheduling of tasks is carried out according to the task state information;
a calculation module: when the task migration module requires to migrate part of task loads to the cloud, the computing module is responsible for processing the loads.
7. The checkpoint-description-file-based cloud fusion task migration system of claim 6, wherein the cloud component further comprises:
a task recovery module: and receiving state backup data uploaded by a data backup module of the equipment end of the Internet of things, and when the task needs to be restarted after being migrated, the task recovery module is responsible for recovering the execution state of the task according to the backup data.
8. The cloud-based converged task migration method based on checkpoint profiles as claimed in claim 6, wherein the cloud-based converged task migration system based on checkpoint profiles as claimed in any one of claims 1 to 7 is adopted, and comprises:
step S1: acquiring control information according to current and historical load processing rates, acquiring the current and historical load processing rates, and predicting the execution rate of the task in the period from the next moment to the time limit of task completion by using a long-short term memory neural network;
step S2: accumulating the predicted execution rates, if the ratio of the predicted execution rates to the residual load capacity is lower than a threshold value, marking the task as incapable of being completed in time, and acquiring information of a result of marking incapable of being completed in time;
step S3: according to the information that the marking result cannot be completed in time, if the load of the task can be processed in parallel, transferring part of the load to a cloud end for execution;
and if the load of the task cannot be processed in parallel, recovering the task execution state at the cloud end by using the state backup data of the task, and re-issuing the recovered task to the equipment end of the Internet of things.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110149647.0A CN112799818B (en) | 2021-02-03 | 2021-02-03 | Cloud fusion task migration method and system based on check point description file |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110149647.0A CN112799818B (en) | 2021-02-03 | 2021-02-03 | Cloud fusion task migration method and system based on check point description file |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112799818A true CN112799818A (en) | 2021-05-14 |
| CN112799818B CN112799818B (en) | 2022-09-23 |
Family
ID=75813983
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110149647.0A Active CN112799818B (en) | 2021-02-03 | 2021-02-03 | Cloud fusion task migration method and system based on check point description file |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112799818B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103248659A (en) * | 2012-02-13 | 2013-08-14 | 北京华胜天成科技股份有限公司 | Method and system for dispatching cloud computed resources |
| CN105912390A (en) * | 2016-04-07 | 2016-08-31 | 四川大学 | Template based task scheduling strategy in cloud computation environment |
| CN107196865A (en) * | 2017-06-08 | 2017-09-22 | 中国民航大学 | A kind of adaptive threshold overload moving method of Load-aware |
| CN109194746A (en) * | 2018-09-06 | 2019-01-11 | 广州知弘科技有限公司 | Heterogeneous Information processing method based on Internet of Things |
| CN109271142A (en) * | 2018-07-24 | 2019-01-25 | 武汉空心科技有限公司 | Exploitation cloud platform based on template component |
| US20190141593A1 (en) * | 2017-11-03 | 2019-05-09 | Hon Hai Precision Industry Co., Ltd. | Methods and related devices for optimizing a mobile edge computing (mec) system |
| CN109831427A (en) * | 2019-01-29 | 2019-05-31 | 浪潮金融信息技术有限公司 | A kind of internet of things equipment registers cloud platform method automatically |
| CN111682973A (en) * | 2020-08-17 | 2020-09-18 | 烽火通信科技股份有限公司 | Method and system for arranging edge cloud |
-
2021
- 2021-02-03 CN CN202110149647.0A patent/CN112799818B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103248659A (en) * | 2012-02-13 | 2013-08-14 | 北京华胜天成科技股份有限公司 | Method and system for dispatching cloud computed resources |
| CN105912390A (en) * | 2016-04-07 | 2016-08-31 | 四川大学 | Template based task scheduling strategy in cloud computation environment |
| CN107196865A (en) * | 2017-06-08 | 2017-09-22 | 中国民航大学 | A kind of adaptive threshold overload moving method of Load-aware |
| US20190141593A1 (en) * | 2017-11-03 | 2019-05-09 | Hon Hai Precision Industry Co., Ltd. | Methods and related devices for optimizing a mobile edge computing (mec) system |
| CN109271142A (en) * | 2018-07-24 | 2019-01-25 | 武汉空心科技有限公司 | Exploitation cloud platform based on template component |
| CN109194746A (en) * | 2018-09-06 | 2019-01-11 | 广州知弘科技有限公司 | Heterogeneous Information processing method based on Internet of Things |
| CN109831427A (en) * | 2019-01-29 | 2019-05-31 | 浪潮金融信息技术有限公司 | A kind of internet of things equipment registers cloud platform method automatically |
| CN111682973A (en) * | 2020-08-17 | 2020-09-18 | 烽火通信科技股份有限公司 | Method and system for arranging edge cloud |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112799818B (en) | 2022-09-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10761765B2 (en) | Distributed object replication architecture | |
| US10585746B2 (en) | Tracking status and restarting distributed replication | |
| CN110119377B (en) | Docker container-oriented online migration system implementation and optimization method | |
| US10509675B2 (en) | Dynamic allocation of worker nodes for distributed replication | |
| US10853111B1 (en) | Virtual machine instance migration feedback | |
| US20190245918A1 (en) | Distributed replication of an object | |
| US11531909B2 (en) | Computer system and method for machine learning or inference | |
| US11113158B2 (en) | Rolling back kubernetes applications | |
| US7610511B2 (en) | Journal migration method and data recovery management method | |
| US11347684B2 (en) | Rolling back KUBERNETES applications including custom resources | |
| US9244719B2 (en) | Batch processing system | |
| CN111506253B (en) | Distributed storage system and storage method thereof | |
| CN111752545B (en) | Stream computing method supporting data replay | |
| US11934665B2 (en) | Systems and methods for ephemeral storage snapshotting | |
| CN103678051A (en) | On-line fault tolerance method in cluster data processing system | |
| CN112486639A (en) | Data saving and restoring method and device for task, server and storage medium | |
| WO2023082681A1 (en) | Data processing method and apparatus based on batch-stream integration, computer device, and medium | |
| CN112799818B (en) | Cloud fusion task migration method and system based on check point description file | |
| WO2024041363A1 (en) | Serverless-architecture-based distributed fault-tolerant system, method and apparatus, and device and medium | |
| CN112231303A (en) | Data migration system and method | |
| WO2023165512A1 (en) | Fault file storage method and related apparatus | |
| CN112612635A (en) | Multi-level protection method for application program | |
| CN111628924A (en) | Method, system, storage medium and electronic device for sending e-mail | |
| CN115718674A (en) | Data disaster tolerance recovery method and device | |
| CN106250266B (en) | System repairing method and device |
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 |