CN111597031A - Scheduling method of scientific workflow in multi-cloud environment - Google Patents
Scheduling method of scientific workflow in multi-cloud environment Download PDFInfo
- Publication number
- CN111597031A CN111597031A CN202010438796.4A CN202010438796A CN111597031A CN 111597031 A CN111597031 A CN 111597031A CN 202010438796 A CN202010438796 A CN 202010438796A CN 111597031 A CN111597031 A CN 111597031A
- Authority
- CN
- China
- Prior art keywords
- task
- tasks
- time
- scientific workflow
- earliest
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000005520 cutting process Methods 0.000 claims description 10
- 230000000977 initiatory effect Effects 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 description 6
- 238000003860 storage Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000005457 optimization Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002243 precursor Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a scheduling method of a scientific workflow in a cloudy environment, and relates to the technical field of task scheduling; initializing scientific workflow in a cloud environment, defining the earliest ending time, the earliest starting time, the latest ending time and the deadline of tasks of the scientific workflow, preprocessing the scientific workflow, confirming the structure of the scientific workflow, merging adjacent tasks with directed cut edges through a directed acyclic graph of the scientific workflow, reducing the number of tasks, adding stacks aiming at local key paths, adding scheduled tasks to the stacks, adding parent tasks of the scheduled tasks added to the stacks, repeatedly adding the tasks in the way until no parent task exists, confirming the actual starting time and the actual ending time of all the tasks in the local critical path, and updating the earliest ending time, the earliest starting time, the latest ending time and the deadline, and scheduling the tasks of the local critical path to the cloud environment with the shortest execution time.
Description
Technical Field
The invention discloses a scheduling method, relates to the technical field of task scheduling, and particularly relates to a scheduling method of a scientific workflow in a cloudy environment.
Background
One type of cloud computing distributed computing refers to decomposing a huge data computing processing program into countless small programs through a network "cloud", and then processing and analyzing the small programs through a system composed of a plurality of servers to obtain results and returning the results to a user. Cloud computing has strong parallel computing capability, and therefore, the problem of scientific workflow optimization task allocation is generated in a multi-cloud environment.
However, different cloud service providers provide different services, and different services require different computing time, which results in differences in different cloud environments. Therefore, when the final task is completed and the scientific workflow task is distributed in a cloud environment, the task completion time is always increased, the cloud computing speed is not improved to the maximum extent, the user can respond the fastest time, the self consumption is increased by improving the cloud computing cost, and the profit of the cloud computing service of an enterprise is reduced.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a scheduling method of a scientific workflow in a cloud environment, which compresses a transmission path of data by adopting a series of optimization measures such as merging directed edge cutting tasks, distributing tasks of local critical paths to the most suitable cloud environment and the like according to the basic characteristics of the scientific workflow in the cloud environment, reduces the execution cost of the scientific workflow, and further improves the performance of an algorithm.
The specific scheme provided by the invention is as follows:
a scheduling method of scientific workflow in a cloud environment is provided, which is used for initializing scientific workflow in the cloud environment, defining the earliest ending time, the earliest starting time, the latest ending time and the deadline of the task of the scientific workflow,
the scientific workflow is preprocessed to confirm the structure of the scientific workflow, adjacent tasks with directed cut edges are combined through a directed acyclic graph of the scientific workflow, the number of the tasks is reduced,
adding a stack for the local critical path, adding the scheduled task to the stack, and adding the parent task of the scheduled task added to the stack, thereby repeatedly adding until no parent task exists,
and confirming the actual starting time and the actual ending time of all tasks in the local critical path, updating the earliest ending time, the earliest starting time, the latest ending time and the deadline, and scheduling the tasks of the local critical path to the cloud environment with the shortest task execution time.
The scheduling method defines a false-in task and a false-out task, wherein the running time of the false-in task and the false-out task is 0, and the false-in task and the false-out task are used for starting iteration of a true-in task and a true-out task.
The scheduling method combines adjacent tasks with directed cut edges:
acquiring DAG of scientific workflow, determining the in-degree and out-degree of each task,
and if the directed cutting edge exists, deleting the directed cutting edge, and combining the two tasks corresponding to the directed cutting edge into a new task.
The method for scheduling the tasks of the scientific workflow comprises the steps of calculating the earliest end time, the earliest start time, the latest end time and the deadline of all the tasks of the scientific workflow, and analyzing the time of the tasks of the local critical path executed by the cloud environment.
The scheduling method comprises the following processes:
calculating the execution time of each task of the local key path by using the earliest end time, the earliest start time, the latest end time and the deadline under different cloud environments, adding the execution times to obtain the total operation time of the local key path under each cloud environment, and scheduling the cloud environment at the minimum value.
A scheduling system of scientific workflow under a multi-cloud environment comprises an initialization module, a preprocessing module, a path module and a selection scheduling module,
the initialization module initializes scientific workflow in cloud environment, defines the earliest ending time, the earliest starting time, the latest ending time and the deadline of the task of the scientific workflow,
the preprocessing module preprocesses the scientific workflow, confirms the structure of the scientific workflow, combines adjacent tasks with directed cut edges through a directed acyclic graph of the scientific workflow, reduces the number of the tasks,
the path module adds a stack for the local critical path, adds the scheduled task to the stack, and adds the parent task of the scheduled task added to the stack, repeatedly adding until there is no parent task,
and the selection scheduling module confirms the actual starting time and the actual ending time of all tasks in the local critical path, updates the earliest ending time, the earliest starting time, the latest ending time and the deadline, and schedules the tasks of the local critical path to the cloud environment with the shortest task execution time.
Scheduling device of scientific work flow under cloudy environment includes: at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is used for calling the machine readable program and executing the scheduling method of the scientific workflow in the cloud environment.
A computer readable medium: the computer readable medium has stored thereon computer instructions, which, when executed by a processor, cause the processor to execute the method for scheduling a scientific workflow in a cloudy environment.
The invention has the advantages that:
the invention provides a scheduling method of a scientific workflow in a multi-cloud environment, which is characterized in that the scientific workflow is preprocessed, adjacent tasks with directed cut edges are combined through a directed acyclic graph of the scientific workflow, the number of the tasks is reduced, the total time of task execution is shortened, the tasks are added in a stack through a local key path, the tasks are added according to the context of the tasks in the stack, all the tasks are sequentially executed, the actual starting time and the actual ending time of all the tasks in the local key path are determined, the earliest ending time, the earliest starting time, the latest ending time and the deadline are updated, the tasks of the local key path are scheduled to the cloud environment with the shortest task execution time, the execution cost of the scientific workflow is reduced, and the performance of an algorithm is further improved.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
Detailed Description
Scientific workflow (scientific workflow) represents a series of structured activities and computational steps required to solve a scientific computational problem. The computation involved in scientific computation has high intensity and complexity, and complex dependency relationship is processed. The earliest origins are from the office field or production automation, and the final result is that the whole command is divided into individual tasks and the tasks are executed in sequence according to a certain sequence.
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
The invention provides a scheduling method of scientific workflow in a multi-cloud environment, which is used for initializing the scientific workflow in the cloud environment, defining the earliest ending time, the earliest starting time, the latest ending time and the deadline of the task of the scientific workflow,
the scientific workflow is preprocessed to confirm the structure of the scientific workflow, adjacent tasks with directed cut edges are combined through a directed acyclic graph of the scientific workflow, the number of the tasks is reduced,
adding a stack for the local critical path, adding the scheduled task to the stack, and adding the parent task of the scheduled task added to the stack, thereby repeatedly adding until no parent task exists,
and confirming the actual starting time and the actual ending time of all tasks in the local critical path, updating the earliest ending time, the earliest starting time, the latest ending time and the deadline, and scheduling the tasks of the local critical path to the cloud environment with the shortest task execution time.
By using the method provided by the invention, the execution cost of the scientific workflow is reduced, the performance of the algorithm is further improved, and the purpose of scheduling the task in the most appropriate cloud environment is realized.
In one embodiment of the present invention, the initialization process is described in detail:
defining a fake task and a fake task, wherein the running time of the two tasks is 0, the fake task is connected with all tasks without parent tasks, the fake task is connected with all tasks without child tasks, the data transmission time between the fake task and all tasks without child tasks is 0, which is equivalent to the fact that the two tasks do not exist actually, only plays the role of starting the next iteration, does not influence the time required by the total scientific workflow, only plays the role of starting the iteration of other tasks,
defining EFT, EST, LST, LFT, D (w) of all tasks and the cloud where the corresponding work is located, EFT (early finish time) is the earliest end time, EST (early start time) is the earliest start time, LST (last start time) is the latest start time, LFT (last finish time) is the latest end time, and D (w) refers to the latest end time of the whole scientific workflow, namely the expiration date. The data is to complete all tasks at the fastest speed before the end of the deadline of the whole scientific workflow, and also to prepare for the sequential execution and scheduling sequence of each task, the tasks can be reasonably arranged and distributed to the most suitable cloud to be executed at different times,
when the task does not have a predecessor task, EFT (i) ═ MET (i, P), i belongs to the real task P belongs to P, MET (i, P) means that the task has the minimum execution time in each cloud, and the cloud P where the task is located is determined at the same time, and when the task has the predecessor task, the EFT calculation method comprises two steps: the method comprises the following steps that firstly, when the calculation is carried out in different clouds, for different father tasks, EFT [ father task ] + MET (i, p) + TE can be calculated, wherein EFT [ predecessor task ] is the EFT size of the predecessor task, EFT (i) is min { max (EFT (j) + MET (i, p) + TE) }, i belongs to a direct subtask which really enters a task p belongs to Pj belongs to i, TE is the data transfer time between the task and the predecessor task, if the task and the predecessor task are in the same cloud, TE is 0, otherwise, TE is the data size to be transmitted when the father task ends and is divided by the data transmission rate between clouds, the ending time with different sizes can be obtained, and the largest ending time can be taken out from the different father tasks; secondly, respectively obtaining a data time in different cloud environments in the first step, and selecting the minimum data time as the final EFT of the task;
EST (i) ═ EFT (i) — MET (i, p) as shown in the formula, and EST (i) ═ EFT (i) — MET (i, p) after EST of the same task is obtained, the result is the obtained result;
when the task does not have subsequent tasks, LST (i) ═ D (w) — MET (i, P), MET (i, P), i epsilon really enters the task P epsilon P, which means that the task has the minimum execution time in each cloud, and the cloud where the task is located at the moment is determined; when the task has a subsequent task, LST (i) ═ max (min (LST (j) — MET (i, p) -TE)), i ∈ real to enter a direct subtask of the task p ∈ Pj ∈ i, and the calculation method of the LST is two steps: calculating LST [ successor task ] + MET (i, p) + TE in different clouds for different subtasks, wherein the LST [ successor task ] is the LST size for calculating the successor task, the TE is the data transfer time between the task and the successor task, if the task and the successor task are in the same cloud, the TE is 0, otherwise, the TE is the data size to be transmitted when the subtask is finished and is divided by the data transmission rate between the clouds, the finishing time with different sizes of different subtasks can be obtained, and the minimum finishing time is taken out; secondly, respectively obtaining a data time in different cloud environments in the first step, and selecting the maximum data time as the LST of the task;
LFT (i) ═ LST (i) + MET (i, p), after finding the LFT of the same task, LFT (i) ═ LST (i) + MET (i, p), the result is the result;
since the leave-in task is the task that is executed first and the execution time is 0, the corresponding EFT and EST are both 0, and the leave-in task is the task that is executed last, the corresponding LST and LFT are both D (w), the deadline,
and mark the vacated tasks and the vacated tasks as scheduled tasks.
According to the invention, the corresponding task execution time and the specific algorithm are defined through the initialization process, so that sufficient preparation is made for reasonably arranging the task to be dispatched to the most appropriate cloud environment according to the task execution time.
In another embodiment of the present invention, the pre-treatment process is specifically described:
inputting a DAG of scientific workflow, creating a graph matrix of father-son nodes, determining the in-degree and the out-degree of each task, wherein the graph matrix is an n multiplied by n matrix, if an edge exists between a certain task and another task, the value at the corresponding position is set to be 1, otherwise, the value is initially set to be 0, after the steps are completed, the in-degree and the out-degree of each task can be confirmed,
if the directed cutting edge exists, deleting the edge, and merging the two tasks corresponding to the edge into a new task, namely, the out degree of the parent task is 1, and the in degree of the child task is 1. Changing the value of the corresponding position of the directed cutting edge on the matrix into 0, namely deleting the edge, simultaneously changing the outgoing degree and the later-linked edge of the sub task into the outgoing degree and the later-linked edge of the parent task, namely combining the parent task and the sub task, and changing the combined tasks into a new task, wherein the execution time of the task is the sum of the execution times of the two tasks, and at the moment, because the two tasks are operated in the same cloud environment later, the data transmission time between the two tasks is 0;
the above process is repeatedly executed until there is no longer a directional cut edge.
The embodiment enables the directed cutting edge to no longer exist in the whole scientific workflow, namely the whole scientific workflow is simplified to the greatest extent, the workload of the next task distribution can be greatly reduced, the total task number is effectively reduced, and the calculation time of the algorithm is shortened.
In another embodiment of the present invention, the sequence of scheduling tasks is specifically described, wherein a scheduled task is input, a dummy task is input at the beginning, the dummy task is used as the start, the search is continued forward, and finally a local critical path (pep) in a single direction is formed,
the specific process is as follows:
the specific path selection method is that each selected scheduled task is used as input, a key parent task is selected as a direct precursor task of the local key path, then the parent task is used as a current task to continue searching the key parent task until the parent task cannot be searched, and the key parent task means that data is transmitted to the parent task corresponding to the task at the latest in all the parent tasks;
the stack is added into the pcp in sequence to form a complete local critical path, the head task of the local critical path does not have an unscheduled parent task, the tail task does not have an unscheduled child task,
scheduling the tasks in each pep to the most appropriate cloud, and determining the actual starting time and the actual ending time of each task in the pep after scheduling, namely, the tasks become scheduled tasks;
the actual start time and the actual end time of scheduled tasks are confirmed so the EFT, EST, LST, LFT of all other unscheduled parent and child tasks of these tasks need to be updated.
In another embodiment of the present invention, a process of scheduling the entire local critical path to the most suitable cloud environment is specifically described:
confirming the actual start time and the actual end time of all tasks in the pep, searching a cloud environment with the shortest total operation time for the pep, adding the execution time of each task in the pep under different cloud environments, finally obtaining the total operation time of the pep under each cloud environment, obtaining the cloud with the minimum value as the most appropriate cloud environment, if the task of the pep and the executed subtasks derived from the task of the pep are in the same virtual machine under the same cloud environment, distributing the task of the pep to other cloud environments for execution, if the virtual machines under other cloud environments are executed in an overflowing manner, namely the task cannot be executed before the subtasks are executed, under the premise of meeting the constraint of cost conditions, newly building a virtual machine under the original most appropriate cloud environment, and putting all tasks of the pep on the newly built virtual machine for execution,
and confirming the actual ending time of the parent task of the head task in the pep, if the cloud environment of the pep is the same as the cloud environment of the parent task, the actual starting time of the pep head task is the actual ending time of the parent task, otherwise, the time of data transmission between different cloud environments needs to be added, the actual starting execution time of each other task is the actual completion time of the previous task, and the actual completion time is the actual starting execution time of the task plus the operation time of the task in the cloud environment. And finally, setting each task in the pcp as a scheduled task to finish the scheduling of the tasks.
The invention also provides a scheduling device of the scientific workflow in the cloud environment, which comprises: at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is used for calling the machine readable program and executing the scheduling method of the scientific workflow in the cloud environment.
A computer readable medium: the computer readable medium has stored thereon computer instructions, which, when executed by a processor, cause the processor to execute the method for scheduling a scientific workflow in a cloudy environment.
Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion unit to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
Meanwhile, the invention also provides a dispatching system of scientific work flow under the multi-cloud environment, which comprises an initialization module, a preprocessing module, a path module and a selection dispatching module,
the initialization module initializes scientific workflow in cloud environment, defines the earliest ending time, the earliest starting time, the latest ending time and the deadline of the task of the scientific workflow,
the preprocessing module preprocesses the scientific workflow, confirms the structure of the scientific workflow, combines adjacent tasks with directed cut edges through a directed acyclic graph of the scientific workflow, reduces the number of the tasks,
the path module adds a stack for the local critical path, adds the scheduled task to the stack, and adds the parent task of the scheduled task added to the stack, repeatedly adding until there is no parent task,
and the selection scheduling module confirms the actual starting time and the actual ending time of all tasks in the local critical path, updates the earliest ending time, the earliest starting time, the latest ending time and the deadline, and schedules the tasks of the local critical path to the cloud environment with the shortest task execution time.
Through the program module, the execution cost of the scientific workflow can be reduced by utilizing the system, the performance of the algorithm is further improved, and the purpose of scheduling the task in the most appropriate cloud environment is achieved.
It should be noted that not all steps and modules in the above flows and system structures are necessary, and some steps or modules may be omitted according to actual needs. The execution order of the steps is not fixed and can be adjusted as required. The system structure described in the above embodiments may be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by a plurality of physical entities, or some components in a plurality of independent devices may be implemented together.
In the above embodiments, the hardware unit may be implemented mechanically or electrically. For example, a hardware element may comprise permanently dedicated circuitry or logic (such as a dedicated processor, FPGA or ASIC) to perform the corresponding operations. The hardware elements may also comprise programmable logic or circuitry, such as a general purpose processor or other programmable processor, that may be temporarily configured by software to perform the corresponding operations. The specific implementation (mechanical, or dedicated permanent, or temporarily set) may be determined based on cost and time considerations.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (8)
1. A method for scheduling scientific workflow in a cloud environment is characterized in that the scientific workflow in the cloud environment is initialized, the earliest ending time, the earliest starting time, the latest ending time and the deadline of the task of the scientific workflow are defined,
the scientific workflow is preprocessed to confirm the structure of the scientific workflow, adjacent tasks with directed cut edges are combined through a directed acyclic graph of the scientific workflow, the number of the tasks is reduced,
adding a stack for the local critical path, adding the scheduled task to the stack, and adding the parent task of the scheduled task added to the stack, thereby repeatedly adding until no parent task exists,
and confirming the actual starting time and the actual ending time of all tasks in the local critical path, updating the earliest ending time, the earliest starting time, the latest ending time and the deadline, and scheduling the tasks of the local critical path to the cloud environment with the shortest task execution time.
2. A scheduling method as claimed in claim 1 wherein a dummy task and a dummy task are defined, the run time of the dummy task and the dummy task being 0 for initiating an iteration of the true task and the true task.
3. A scheduling method according to claim 1 or 2 wherein adjacent tasks having directed cut edges are merged:
acquiring DAG of scientific workflow, determining the in-degree and out-degree of each task,
and if the directed cutting edge exists, deleting the directed cutting edge, and combining the two tasks corresponding to the directed cutting edge into a new task.
4. The scheduling method of claim 3 wherein the earliest end time, the earliest start time, the latest end time and the deadline of all tasks of the scientific workflow are calculated for analyzing the time of the cloud environment to execute the tasks of the local critical path.
5. The scheduling method of claim 4, wherein the process is:
calculating the execution time of each task of the local key path by using the earliest end time, the earliest start time, the latest end time and the deadline under different cloud environments, adding the execution times to obtain the total operation time of the local key path under each cloud environment, and scheduling the cloud environment at the minimum value.
6. A scheduling system of scientific work flow under a multi-cloud environment is characterized by comprising an initialization module, a preprocessing module, a path module and a selection scheduling module,
the initialization module initializes scientific workflow in cloud environment, defines the earliest ending time, the earliest starting time, the latest ending time and the deadline of the task of the scientific workflow,
the preprocessing module preprocesses the scientific workflow, confirms the structure of the scientific workflow, combines adjacent tasks with directed cut edges through a directed acyclic graph of the scientific workflow, reduces the number of the tasks,
the path module adds a stack for the local critical path, adds the scheduled task to the stack, and adds the parent task of the scheduled task added to the stack, repeatedly adding until there is no parent task,
and the selection scheduling module confirms the actual starting time and the actual ending time of all tasks in the local critical path, updates the earliest ending time, the earliest starting time, the latest ending time and the deadline, and schedules the tasks of the local critical path to the cloud environment with the shortest task execution time.
7. Scheduling device of scientific work flow under cloudy environment, characterized by includes: at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is used for calling the machine readable program to execute the scheduling method of the scientific workflow in the cloudy environment according to any one of claims 1 to 5.
8. Computer readable medium having stored thereon computer instructions, which when executed by a processor, cause the processor to execute the method of scheduling scientific workflows in a cloudy environment according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010438796.4A CN111597031A (en) | 2020-05-22 | 2020-05-22 | Scheduling method of scientific workflow in multi-cloud environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010438796.4A CN111597031A (en) | 2020-05-22 | 2020-05-22 | Scheduling method of scientific workflow in multi-cloud environment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111597031A true CN111597031A (en) | 2020-08-28 |
Family
ID=72182862
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010438796.4A Pending CN111597031A (en) | 2020-05-22 | 2020-05-22 | Scheduling method of scientific workflow in multi-cloud environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111597031A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112379998A (en) * | 2020-10-20 | 2021-02-19 | 广东工业大学 | Switching time delay optimization method based on task fusion and task scheduling |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015082253A1 (en) * | 2013-12-04 | 2015-06-11 | Koninklijke Philips N.V. | Prediction of critical work load in radiation therapy workflow |
CN105068863A (en) * | 2015-07-16 | 2015-11-18 | 福州大学 | Cost-driven scheduling method for workflow with deadline constraints in cloudy environment |
CN108154317A (en) * | 2018-01-25 | 2018-06-12 | 福建师范大学 | The workflow group scheduling method that Case-based Reasoning self-adjusted block is integrated under cloudy environment |
CN109784663A (en) * | 2018-12-20 | 2019-05-21 | 西北大学 | A kind of workflow schedule method and apparatus |
-
2020
- 2020-05-22 CN CN202010438796.4A patent/CN111597031A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015082253A1 (en) * | 2013-12-04 | 2015-06-11 | Koninklijke Philips N.V. | Prediction of critical work load in radiation therapy workflow |
CN105068863A (en) * | 2015-07-16 | 2015-11-18 | 福州大学 | Cost-driven scheduling method for workflow with deadline constraints in cloudy environment |
CN108154317A (en) * | 2018-01-25 | 2018-06-12 | 福建师范大学 | The workflow group scheduling method that Case-based Reasoning self-adjusted block is integrated under cloudy environment |
CN109784663A (en) * | 2018-12-20 | 2019-05-21 | 西北大学 | A kind of workflow schedule method and apparatus |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112379998A (en) * | 2020-10-20 | 2021-02-19 | 广东工业大学 | Switching time delay optimization method based on task fusion and task scheduling |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | Online multi-workflow scheduling under uncertain task execution time in IaaS clouds | |
JP4781089B2 (en) | Task assignment method and task assignment device | |
US8752059B2 (en) | Computer data processing capacity planning using dependency relationships from a configuration management database | |
US20220164222A1 (en) | Execution of Services Concurrently | |
CN107678848A (en) | Repeater satellite task scheduling constrains the construction method and device of plan model | |
US20060195336A1 (en) | Methods and systems for dynamic parallel processing | |
CN112416585A (en) | GPU resource management and intelligent scheduling method for deep learning | |
CN112835714A (en) | Container arrangement method, system and medium for CPU heterogeneous cluster in cloud edge environment | |
Kumar et al. | Time and cost optimization algorithm for scheduling multiple workflows in hybrid clouds | |
Qian et al. | A workflow-aided Internet of things paradigm with intelligent edge computing | |
Yi et al. | Fast training of deep learning models over multiple gpus | |
CN111597031A (en) | Scheduling method of scientific workflow in multi-cloud environment | |
Kuai et al. | Fair virtual network function mapping and scheduling using proximal policy optimization | |
US20120059938A1 (en) | Dimension-ordered application placement in a multiprocessor computer | |
Li et al. | Time‐cost efficient memory configuration for serverless workflow applications | |
US8073722B2 (en) | Methods and systems for a multi-level scheduling framework | |
Lee et al. | Reliable workflow execution in distributed systems for cost efficiency | |
Vinothina | Scheduling scientific workflow tasks in cloud using swarm intelligence | |
CN115658263A (en) | Task scheduling method and system for cloud computing platform | |
CN112839109B (en) | Cloud resource arranging method based on cloud function and BPMN (Business Process management) specification | |
US7489676B2 (en) | Communication method of exchanging real-time data in a collision recognition-based communication network, corresponding memory medium and communication network | |
Gupta et al. | Real-time scheduling using compact task graphs | |
Wang et al. | Monte carlo tree search for task mapping onto heterogeneous platforms | |
US20130144590A1 (en) | Configuration management | |
Al-Masri et al. | Enhancing Resource Provisioning Across Edge-based Environments |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200828 |