CN108536539A - A kind of method for scheduling task in industrial allocation formula data collecting system - Google Patents
A kind of method for scheduling task in industrial allocation formula data collecting system Download PDFInfo
- Publication number
- CN108536539A CN108536539A CN201810384049.XA CN201810384049A CN108536539A CN 108536539 A CN108536539 A CN 108536539A CN 201810384049 A CN201810384049 A CN 201810384049A CN 108536539 A CN108536539 A CN 108536539A
- Authority
- CN
- China
- Prior art keywords
- node
- acquisition
- task
- time
- tasks
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- 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
- G06F11/1474—Saving, restoring, recovering or retrying in transactions
-
- 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)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides the method for scheduling task in a kind of industrial allocation formula data collecting system, is related to data acquisition technology field.This method includes establishing the correspondence of each acquisition node acquisition resource and acquisition tasks, the correspondence of each acquisition node acquisition time and acquisition tasks and node resource utilization rate, the correspondence of each consumed resource of acquisition node task immigration and migration task number, the correspondence of the communication overhead of task immigration and migration task between each acquisition node, it determines the minimum acquisition node number of initialization and opens corresponding acquisition node, remaining acquisition node resource is spare as shared resource, the Redundant task of vital task in initial acquisition task and which part initiating task is distributed into each acquisition node.Present invention synthesis meets demand of the industrial allocation formula data acquisition to acquisition real-time, reliability, utilization of resources etc. under industrial big data environment, improves industrial data acquisition system resource utilization, collecting efficiency and acquisition reliability.
Description
Technical field
The present invention relates to the tasks in data acquisition technology field more particularly to a kind of industrial allocation formula data collecting system
Dispatching method.
Background technology
With the arrival of industrial big data environment, increasingly diversification, data scale increase data source in industrial process, face
The acquisition of industrial-scale high-frequency data and some new application demands, in order to ensure timing, the real-time of data acquisition
And reliability, more and more enterprises start to be considered as distributed system to carry out industrial data collection.In distributed data
In the design process of acquisition system, task scheduling strategy is the most key link, it will directly affect the property of distributed system
Energy.Good task scheduling approach can reduce the acquisition time of acquisition tasks, improve the collecting efficiency of system.
Under distributed environment, since each acquisition node can neatly be added or exit, what each acquisition node was acquired
Data no longer correspond to fixed production link, and have certain flexibility, due to network sudden and unstability and
The reason of acquisition node itself, in fact it could happen that certain acquisition node failures, overload, some gathered data group or number on acquisition node
It is more than the abnormal conditions such as collection period of setting according to item acquisition failure, acquisition time, how according to the resource of each acquisition node
Service condition carries out dynamic task scheduling and migration so that under abnormal conditions, the data of each production link are remained to realize and normally be adopted
Collection, and load balancing is still kept between each acquisition node, to ensure the reliability and high efficiency of acquisition, and ensure the task of migration
Amount is as few as possible, ensures the monotonicity of task immigration, is distributed number to save overhead when task immigration as far as possible
A problem urgently to be resolved hurrily is designed according to acquisition system, in addition, due to the data reliability requirement of industrial important production link
More stringent, the different data significance level of same production link is also not quite similar sometimes, for the acquisition tasks of significant data
It is generally necessary to be backed up, backup tasks are only acquired, and without storage, are only just deposited when main task acquires failure
How storage is ensureing main task rational management and migration, is meeting the premise of load balancing and monotonicity under distributed environment
Under, backup tasks are also realized with rational scheduling and migration, to ensure the monotonicity of task immigration as far as possible, while ensureing director
Business and its corresponding backup tasks will not be migrated to the same acquisition node, be needed to be considered in industrial data collection another ask
Topic.
The patent in terms of industrial allocation formula data collecting system task scheduling mainly has Patent No. at present
A kind of large-scale distributed data collecting system and method, Patent No. based on industrial process of CN105527948A
A kind of large-scale distributed data collecting system and method, Patent No. CN201610736266.1 of CN201610522950.X
A kind of large-scale distributed intelligent data acqusition system and method based on industrial cloud, Patent No. 201610622589.8
Task Scheduling Mechanism based on consistency hash algorithm and system.Patent CN105527948A and patent CN201610522950.X
When acquiring client failure, task scheduling is realized in such a way that acquisition tasks to be reassigned to remaining acquisition client,
Although the scheduling scheme meets the load balancing between each node, but redistributing for task leads to have on each node a large amount of appoint
Business is migrated, and the monotonicity of task immigration is cannot be satisfied, and task immigration expense is excessive, and resource utilization is caused to reduce,
And task scheduling approach described in above-mentioned two patent is not comprehensive enough, for not provided situations such as task acquisition failure on acquisition node
Corresponding solution;Although patent CN201610736266.1 more comprehensively gives acquisition node failure, overload, acquisition times
Business acquisition failure, task scheduling approach when acquisition time is unsatisfactory for collection period requirement, but for scheduling when specifically migrate it is more
Few task migrates which task is not described in detail, and scheduling scheme described in the patent is not in view of scheduling process saves respectively
The problems such as point load balancing, task immigration expense, to be easy to influence the efficiency and real-time of data acquisition;Patent
201610622589.8, by the way that the cryptographic Hash of task and corresponding execution unit to be mapped on the same Hash ring, hold according to task
Places number and the search direction of selection establish the mapping relations of task and several execution units on Hash annulus, when it is newly-increased,
When deleting execution unit or newly-increased task, realize that dynamic task scheduling, the dispatching method can be realized by correcting mapping relations
The monotonicity of task immigration, and can realize being flexibly matched with for task and execution unit.Although the technical solution in above-mentioned patent is all
It can realize the task scheduling of node failure, but its scheduling scheme is vital task only for all tasks, be required to backup
The occasion of backup is all not implemented in occasion or all tasks, and during actual industrial production, not all task all exists
Backup, in order to improve resource utilization, often only backs up the acquisition tasks corresponding to which part significant data, existing
There is the task scheduling mode in patent to be not suitable in this case.
Invention content
The technical problem to be solved by the present invention is in view of the above shortcomings of the prior art, provide a kind of industrial allocation formula data
Method for scheduling task in acquisition system improves industrial data acquisition system resource utilization, collecting efficiency and acquisition to reach
The purpose of reliability.
In order to solve the above technical problems, the technical solution used in the present invention is:
A kind of method for scheduling task in industrial allocation formula data collecting system, includes the following steps:
Step 1, the correspondence for establishing each acquisition node acquisition resource and acquisition tasks, detailed process are:
Step 1.1 individually distributes acquisition tasks to each acquisition node, changes acquisition tasks number, measures a large amount of different
Under the conditions of acquisition tasks, the acquisition tasks number T that is consumed on i-th of acquisition nodeiAnd its acquisition resource utilization data ui, with
And the required acquisition time data time of whole acquisition tasks distributed on the acquisition nodei;
Step 1.2, to the acquisition tasks data T on i-th of acquisition node obtained by step riveriWith corresponding resource consumption dataIt is fitted, acquires the functional relation of the two
Step 2, the correspondence for establishing each acquisition node acquisition time and acquisition tasks and node resource utilization rate, it is right
Step 1.1 gained acquisition tasks number TiWith corresponding acquisition resource utilization data ui, acquisition time data timeiIt is fitted,
Acquire acquisition time data time on i-th of acquisition nodeiWith acquisition tasks number TiWith resource utilization data uiFunction close
It is formula timei=g (Ti, ui);
Step 3, the correspondence for establishing task immigration is consumed on each acquisition node resource and migration task number;
Step 4, the correspondence for establishing the communication overhead of task immigration and migration task between each acquisition node;
Step 5, the minimum acquisition node number for determining initialization simultaneously open corresponding acquisition node, remaining acquisition node money
Source is spare as shared resource;
The Redundant task of vital task in initial acquisition task and which part initiating task is distributed to step 5 by step 6
Determining each acquisition node;
Acquisition node failure is judged whether, if so, thening follow the steps 7;Judge whether acquisition node overload or
Acquisition time is unsatisfactory for the case where collection period requirement on acquisition node, if so, thening follow the steps 8;It judges whether to acquire
On node the case where important acquisition tasks acquisition failure, if so, thening follow the steps 9;If the above situation is not present, task tune
Degree terminates;
Step 7, when there are acquisition node failure, be acquired task scheduling after node failure, specific dispatching method is:
Step 7.1, current collection period, by the corresponding Redundant task of vital task in initiating task on fault collection node
Collection result the database of data collecting system is stored in from its corresponding acquisition node;
Step 7.2, next collection period, find from the acquisition node of Redundant backup and fault collection node resource is remaining
The most similar node of situation, as the acquisition node being newly added, by the whole initiating tasks and redundancy times on fault collection node
Business, which is migrated to the acquisition node being newly added, to be acquired;
Step 8, when there are acquisition node overload or when being unsatisfactory for collection period requirement there are acquisition time on acquisition node,
It is acquired task scheduling after node overload, specific dispatching method is:
Step 8.1, the correspondence that step 4 is arrived according to step 1, establish the new Optimized model that acquisition node number is added,
Determine that the minimum value m for the acquisition node number being newly added, Optimized model are as follows:
min m
s.tTij×Tji=0 i=1,2...n;J=1,2...n, j ≠ i (1-1)
Wherein,
timei≤ Time i=1,2...n (1-7)
Wherein,
Wherein,
timej≤ Time j=n+1, n+2...n+m (1-12)
Wherein,
Wherein, formula (1-1) moves to node j if indicating that there are tasks from node i, and there is no tasks to migrate from node j
To node i, to ensure the monotonicity of task immigration, TijExpression moves to the task number of j-th of node from i-th of node,
TjiExpression moves to the task number of i-th of node from j-th of node;Formula (1-2) indicates that a node cannot both move out task
Task is moved into again, and n indicates the acquisition node number to work before newly added node;Formula (1-3) indicate task immigration during and
After the completion of task immigration in initialization acquisition node the resource utilization of i-th of acquisition node no more than set by user
The resource utilization upper limit, ui0Indicate the resource utilization before the unallocated task of i-th of node, R in initialization acquisition nodeiTable
Show the available resources total amount configured on i-th of acquisition node in initialization acquisition node, uhIndicate the node money of user setting
The source utilization rate upper limit, uwIndicate the node resource utilization threshold bandwidth of user setting, TiInitialization is adopted before indicating task immigration
Collect the acquisition tasks number on i-th of node in node,I-th of node in initialization acquisition node after expression task immigration
Upper acquisition tasks execute consumed resource,It is acquired on i-th of node in initialization acquisition node after expression task immigration
The resource that task immigration is consumed,WithValue determine that specific functional relation is by step by formula (1-4) and formula (1-5) respectively
Rapid 1 and step 3 determine;Formula (1-6) indicates the utilization of resources of i-th of node in initialization acquisition node after the completion of task immigration
Rate;Execute primary whole acquisition tasks after formula (1-7) expression task immigration in initialization acquisition node on i-th of acquisition node
Acquisition time timeiNo more than collection period Time, time set by useriInitialization acquires after indicating task immigration
The acquisition time of primary whole acquisition tasks, time are executed in node on i-th of nodeiValue by formula (1-8) determine, specifically
Functional relation is determined by step 2;Formula (1-9) indicates the resource utilization of j-th of acquisition node in newly added node no more than
The resource utilization upper limit set by user, ujThe resource of j-th of node is sharp in newly added node after the completion of expression task immigration
With rate, uj0Indicate the resource utilization before the unallocated task of j-th of node, R in newly added nodejIt indicates the in newly added node
The available resources total amount configured on j acquisition node,Acquisition is appointed on j-th of node in newly added node after expression task immigration
Business executes consumed resource,Acquisition tasks migration is consumed on j-th of node in newly added node after expression task immigration
Resource,WithValue determine that specific functional relation is true by step 1 and step 3 by formula (1-10) and formula (1-11) respectively
It is fixed;After formula (1-12) expression task immigration adopting for primary whole acquisition tasks is executed on j-th of acquisition node in newly added node
Collect the time no more than collection period Time, time set by userjJ-th of node in newly added node after expression task immigration
The upper acquisition time for executing primary whole acquisition tasks, timejValue by formula (1-13) determine, specific functional relation is by step
2 determine;
Optimized model in step 8.2, solution procedure 8.1, acquires the minimum acquisition node number being newly added, and from redundancy
The acquisition node that corresponding number is selected in hot standby acquisition node, is added collecting work;
Step 8.3 establishes acquisition node task immigration Optimized model, and model is as follows:
Wherein,
s.tTij×Tji=0 i=1,2...n;J=1,2...n, j ≠ i (1-18)
Wherein,
timei≤ Time i=1,2...n (1-23)
Wherein,
Wherein,
timej≤ Time j=n+1, n+2...n+m (1-27)
Wherein,
Wherein,
Wherein, the resource of each node and communication overhead and task are moved during formula (1-14) expression minimum task immigration
After the completion of shifting between each acquisition node load difference and Δ U, p (Tij) indicate that the communication of task immigration between node i and node j is opened
Pin, specific functional relation are determined that the sum of the load difference after the completion of Δ U expression task immigrations between each acquisition node takes by step 4
Value is determined by formula (1-17);Formula (1-29) indicates that i-th of node moves to the task number of j-th of node equal to i-th of node
Move to the sum of initiating task number and the Redundant task number of j-th of node, taskK, i, jExpression is moved to from i-th of node
The initiating task that j-th of node serial number is k,It indicates to move to the redundancy that j-th of node serial number is l from i-th of node
Task;Formula (1-30) indicates taskK, i, jValue, it is corresponding superfluous when the initiating task that number is k is located on i-th node
Remaining task is not located on j-th node, and when needing to be migrated to j-th of node, taskK, i, j=1, otherwise taskK, i, j=
0, TaskiAcquisition tasks serial number set before expression task immigration on i-th of acquisition node, TaskjJ-th before expression task immigration
Acquisition tasks serial number set on acquisition node, TaskijIndicate that i-th of acquisition node moves to the task of j-th of acquisition node
Serial number set;Formula (1-31) indicatesValue, it is corresponding when the Redundant task that number is l is located on i-th node
Redundant task is not located on j-th node, and when needing to be migrated to j-th of node,Otherwise
Optimized model in step 8.4, solution procedure 8.3 obtains i-th of acquisition node and moves to j-th of acquisition node
Task number set Taskij;
Step 8.5 is acquired the task immigration between node according to step 8.4 acquired results;
Step 9, on there are acquisition node when important acquisition tasks acquisition failure, by the acquisition of its corresponding Redundant task
As a result the database of data collecting system is stored in from corresponding acquisition node.
It is using advantageous effect caused by above-mentioned technical proposal:Industrial allocation formula data acquisition provided by the invention system
Method for scheduling task in system considers acquisition node resource utilization, collecting efficiency, load balancing, acquires reliability, is different
Multiple factors such as the resource of dynamic task migrating and communication overhead between structure acquisition node, node are established a series of based on emulation
Optimized model realizes acquisition node failure in industrial allocation formula data collecting system, overload, acquisition time and is more than acquisition week
After phase, the acquisition failure of part acquisition tasks, not only asked comprising 1: 1 Redundant task but also comprising the Real-time Task Dispatch of nonredundancy task
Topic meets industrial allocation formula data under industrial big data environment and acquires to acquisition real-time, reliability, utilization of resources etc.
Demand.
Description of the drawings
Fig. 1 is the method for scheduling task flow chart in industrial allocation formula data collecting system provided in an embodiment of the present invention.
Specific implementation mode
With reference to the accompanying drawings and examples, the specific implementation mode of the present invention is described in further detail.Implement below
Example is not limited to the scope of the present invention for illustrating the present invention.
In the present embodiment, 4 acquisition nodes are set altogether, number is 1,2,3,4 respectively.The memory configurations of 4 acquisition nodes
Situation is respectively 4G, 8G, 4G, 16G, the original allocation moment, the resource utilizations of 4 acquisition nodes is respectively 10%, 15%,
13%, 27%.
In the present embodiment, F1-F35 this 35 initial acquisition tasks are set, correspond to respectively several on F1-F35 this 35 Fans
According to acquisition, each acquisition tasks include 404 gathered data items, wherein Single Float types data 31, Double
Float types data 52, Boolean types data 319, Unsigned Integer types data 2.Wherein F11-F22 attaches most importance to
Task is wanted, needs to carry out redundancy.It is 1-35 to be numbered respectively to F1-F35 this 35 tasks, the Redundant task number point of F11-F22
It Wei 11*, 12*, 13*, 14*, 15*, 16*, 17*, 18*, 19*, 20*, 21*, 22*。
Method for scheduling task in a kind of industrial allocation formula data collecting system of the present embodiment, as shown in Figure 1, specific side
Method is as described below.
Step 1, the correspondence for establishing each acquisition node acquisition resource and acquisition tasks.
Step 1.1 individually distributes acquisition tasks to each acquisition node, changes acquisition tasks number, it is a large amount of to measure acquisition
When the acquisition tasks of different numbers, the acquisition tasks number T that is consumed on i-th of acquisition nodeiAnd its acquisition resource utilization number
According to uiAnd the required acquisition time data time of whole acquisition tasks distributed on the acquisition nodei;
In the present embodiment, the acquisition resource data refers to memory source.
In the present embodiment, acquisition tasks are distributed to 1,2,3, No. 4 acquisition node respectively, measure the acquisition for acquiring different numbers
The memory source and its memory usage and acquisition time consumed when task, the data measured are respectively such as table 1, table 2, table
3, shown in table 4.
No. 11 acquisition node acquisition tasks numbers of table, memory usage and acquisition time data
Acquisition tasks number | 1 | 2 | 3 | 4 | 5 | 6 |
Consume memory source (G) | 0.19 | 0.37 | 0.55 | 0.76 | 0.97 | 1.15 |
Memory usage (%) | 14.75 | 19.25 | 23.75 | 29 | 34.25 | 38.75 |
Acquisition time (ms) | 305 | 330 | 357 | 390 | 427 | 469 |
Acquisition tasks number | 7 | 8 | 9 | 10 | 11 | 12 |
Consume memory source (G) | 1.33 | 1.53 | 1.72 | 1.91 | 2.11 | 2.29 |
Memory usage (%) | 43.25 | 48.25 | 53 | 57.25 | 62.75 | 67.25 |
Acquisition time (ms) | 508 | 559 | 607 | 661 | 728 | 788 |
No. 22 acquisition node acquisition tasks numbers of table, memory usage and acquisition time data
No. 33 acquisition node acquisition tasks numbers of table, memory usage and acquisition time data
Acquisition tasks number | 1 | 2 | 3 | 4 | 5 | 6 |
Consume memory source (G) | 0.18 | 0.38 | 0.59 | 0.77 | 0.98 | 1.17 |
Memory usage (%) | 17.5 | 22.5 | 27.75 | 32.25 | 37.5 | 42.25 |
Acquisition time (ms) | 319 | 346 | 380 | 415 | 454 | 498 |
Acquisition tasks number | 7 | 8 | 9 | 10 | 11 | 12 |
Consume memory source (G) | 1.34 | 1.52 | 1.72 | 1.88 | 2.09 | 2.27 |
Memory usage (%) | 46.5 | 51 | 56 | 60 | 65.25 | 69.75 |
Acquisition time (ms) | 539 | 589 | 647 | 694 | 763 | 824 |
No. 44 acquisition node acquisition tasks numbers of table, memory usage and acquisition time data
Step 1.2, to the acquisition tasks data T on i-th of acquisition node obtained by step 1.1iNumber is consumed with corresponding resource
According toIt is fitted, acquiresAnd TiFunctional relation
In the present embodiment, No. 1 acquisition node being fittedAnd T1Functional relation be
No. 2 acquisition nodesAnd T2Functional relation beNo. 3 acquisition nodesAnd T3Functional relation
Formula isNo. 3 acquisition nodesAnd T4Functional relation be
Step 2 establishes resource utilization corresponding to each acquisition node acquisition time number acquisition tasks different from acquisition
Correspondence, to acquiring T on each node obtained by step 1.1iAcquisition resource utilization u corresponding to a acquisition tasksi, acquisition
Time timeiIt is fitted, acquires acquisition time time on i-th of acquisition nodeiT is acquired on the nodeiA acquisition tasks institute
Corresponding acquisition resource utilization uiFunctional relation timei=g (ui)。
In the present embodiment, step 1.1 the data obtained is fitted using least square method, be fitted No. 1 acquisition
Node time1And u1Functional relation be time1=0.0869u1 2+2.0390u1+ 257.6073, No. 2 acquisition node time2
And u2Functional relation be time2=0.1370u2 2-0.2669u2+ 161.6201, No. 3 acquisition node time3And u3Function
Relational expression is time3=0.0880u3 2+2.0146u3+ 256.5218, No. 4 acquisition node time4And u4Functional relation be
time4=0.3159u4 2-10.3928u4+190.3659。
Step 3, the correspondence for establishing task immigration is consumed on each acquisition node resource and migration task number.
In the present embodiment, the resource overhead due to migrating task between each node comes compared to the resource overhead of task execution
Say it is very small, can be neglected.
Step 4, the correspondence for establishing the communication overhead of task immigration and migration task number between each acquisition node.
In the present embodiment, due to using fiber optic communication between each node, the communication overhead of task immigration is very small, can be ignored not
Meter.
Step 5, the minimum acquisition node number for determining initialization simultaneously open corresponding acquisition node, remaining acquisition node money
Source is spare as shared resource.
In the present embodiment, the acquisition node number of initialization is 2, respectively No. 3 and No. 4 acquisition nodes, No. 1 and 2
Number acquisition node is spare as shared resource.The spare purpose of shared resource is to save the server resource of data acquisition, is improved
Entire resource utilization proposes resource bid when acquisition tasks change or need to increase resource, then to system.
The Redundant task of vital task in initial acquisition task and which part initiating task is distributed to step 5 by step 6
Determining each acquisition node.
In the present embodiment, the set of tasks for distributing to No. 3 acquisition nodes is { 11-16,17*-22* }, distributes to No. 4 acquisitions
The set of tasks of node is { 1-10,11*-16*, 17-35 }.
Step 7, when there are acquisition node failure, be acquired task scheduling after node failure.
In the present embodiment, it is assumed that No. 3 acquisition node failures, then the method for scheduling task after acquisition node failure is as follows.
Step 7.1, current collection period, by the corresponding Redundant task of vital task in initiating task on fault collection node
Collection result the database of data collecting system is stored in from its corresponding acquisition node;The data carried depending on data collecting system
Can be SQL Server databases or HBase databases depending on the type of library;
It is in the current collection period of failure, the 11-16 acquisition tasks on No. 3 acquisition nodes are corresponding in the present embodiment
Redundant task, i.e., 11*-16*Number acquisition tasks are stored in databases from No. 4 acquisition nodes;
Step 7.2, next collection period, find from the acquisition node of Redundant backup and fault collection node resource is remaining
The most similar node of situation, as the acquisition node being newly added, by the whole initiating tasks and redundancy times on fault collection node
Business, which is migrated to the acquisition node being newly added, to be acquired;
In the present embodiment, next collection period, selects No. 3 with failure to adopt from No. 1 and No. 2 acquisition nodes after a failure
No. 1 acquisition node is as the acquisition node being newly added similar in collection node resource residue situation, by the whole on No. 3 acquisition nodes
Initiating task { 11-16 } and Redundant task { 17*-22*Migrate to No. 1 acquisition node and be acquired.
Step 8, when there are acquisition node overload or when being unsatisfactory for collection period requirement there are acquisition time on acquisition node,
It is acquired task scheduling after node overload;
In the present embodiment, it is assumed that No. 3 acquisition node overloads, then method for scheduling task is as follows after acquisition node overload.
Step 8.1, the simulation result that step 4 is arrived according to step 1, establish the new Optimized model that acquisition node number is added,
Determine that the minimum value m for the acquisition node number being newly added, Optimized model are as follows:
min m
s.tTij×Tji=0 i=1,2...n;J=1,2...n, j ≠ i (1-1)
Wherein,
timei≤ Time i=1,2...n (1-7)
Wherein,
Wherein,
timej≤ Time j=n+1, n+2...n+m (1-12)
Wherein,
Wherein, formula (1-1) moves to node j if indicating that there are tasks from node i, and there is no tasks to migrate from node j
To node i, to ensure the monotonicity of task immigration, TijExpression moves to the task number of j-th of node from i-th of node,
TjiExpression moves to the task number of i-th of node from j-th of node;Formula (1-2) indicates that a node cannot both move out task
Task is moved into again, and n indicates the acquisition node number to work before newly added node;Formula (1-3) indicate task immigration during and
After the completion of task immigration in initialization acquisition node the resource utilization of i-th of acquisition node no more than set by user
The resource utilization upper limit, ui0Indicate the resource utilization before the unallocated task of i-th of node, R in initialization acquisition nodeiTable
Show the available resources total amount configured on i-th of acquisition node in initialization acquisition node, uhIndicate the node money of user setting
The source utilization rate upper limit, uwIndicate the node resource utilization threshold bandwidth of user setting, TiInitialization is adopted before indicating task immigration
Collect the acquisition tasks number on i-th of node in node,I-th of node in initialization acquisition node after expression task immigration
Upper acquisition tasks execute consumed resource,Indicate that acquisition tasks migrate institute on i-th of node in initialization acquisition node
The resource of consumption,WithValue determine that specific functional relation is by step 1 and step 3 by formula (1-4) and formula (1-5) respectively
It determines;Formula (1-6) indicates the resource utilization of i-th of node in initialization acquisition node after the completion of task immigration;Formula (1-7)
Execute the acquisition time of primary whole acquisition tasks after expression task immigration in initialization acquisition node on i-th of acquisition node
No more than collection period Time, time set by useriI-th of node in initialization acquisition node after expression task immigration
The upper acquisition time for executing primary whole acquisition tasks, timeiValue by formula (1-8) determine, specific functional relation is by step 2
It determines;Formula (1-9) indicates the resource utilization of j-th of acquisition node in newly added node no more than the resource set by user
The utilization rate upper limit, ujIndicate the resource utilization of j-th of node in newly added node after the completion of task immigration, uj0It indicates new to be added
Resource utilization in node before the unallocated task of j-th of node, RjIt indicates to configure on j-th of acquisition node in newly added node
Available resources total amount,Acquisition tasks execute consumed money on j-th of node in newly added node after expression task immigration
Source,Indicate that acquisition tasks migrate consumed resource on j-th of node in newly added node,WithValue respectively by formula
(1-10) and formula (1-11) determine that specific functional relation is determined by step 1 and step 3;It is new after formula (1-12) expression task immigration
The acquisition time that primary whole acquisition tasks are executed on j-th of acquisition node is added in node no more than acquisition set by user
Cycle T ime, timejExecute the acquisition of primary whole acquisition tasks after expression task immigration in newly added node on j-th of node
Time, timejValue determine that specific functional relation determines by step 2 by formula (1-13);
In the present embodiment, R1=4G, R2=8G, R3=4G, R4=16G, uh=80%, uw=5%, u10=10%, u20=
15%, u30=13%, u40=27%, Time=1000ms, M=47, then Optimized model be specially:
min m
s.tT43×T34=0 (2-1)
Wherein,
Wherein,
time3≤1000 (2-10)
Wherein, time3=0.0880u3 2+2.0146u3+256.5218 (2-11)
time4≤1000 (2-12)
Wherein, time4=0.3159u4 2-10.3928u4+190.3659 (2-13)
Wherein,
Wherein,
time1≤1000 (2-18)
Wherein, time1=0.0869u1 2+2.0390u1+257.6073 (2-19)
time2≤1000 (2-20)
Wherein, time2=0.1370u2 2-0.2669u2+161.6201 (2-21)
Optimized model in step 8.2, solution procedure 8.1, acquires the minimum acquisition node number being newly added, and from redundancy
Corresponding acquisition node is selected in hot standby acquisition node, and collecting work is added;
In the present embodiment, the minimum acquisition node number for solving gained is 1, and the acquisition node being newly added selected is No. 1
Acquisition node;
Step 8.3 establishes acquisition node task immigration Optimized model, and model is as follows:
Wherein,
s.tTij×Tji=0 i=1,2...n;J=1,2...n, j ≠ i (1-18)
Wherein,
timej≤ Time i=1,2...n (1-23)
Wherein,
Wherein,
timej≤ Time j=n+1, n+2...n+m (1-27)
Wherein,
Wherein,
Wherein, the resource of each node and communication overhead and task are moved during formula (1-14) expression minimum task immigration
After the completion of shifting between each acquisition node load difference and Δ U,I-th of section in initialization acquisition node after expression task immigration
Acquisition tasks migrate consumed resource on point,Indicate after task immigration in newly added node acquisition tasks on j-th of node
The consumed resource of migration,WithValue determine that specific functional relation is by step 3 by formula (1-15) and formula (1-16) respectively
It determines, p (Tii) communication overhead of task immigration, specific functional relation are determined by step 4 between expression node i and node j, Δ U
The sum of load difference after the completion of expression task immigration between each acquisition node, value are determined by formula (1-17);Formula (1-18) indicates
Node j is moved to from node i if there are tasks, there is no tasks to move to node i from node j, to ensure task immigration
Monotonicity, TijExpression moves to the task number of j-th of node, T from i-th of nodejiExpression is moved to from j-th of node
The task number of i-th of node;Formula (1-19) indicates that a node cannot not only move out task and moves into task, and n indicates new addition
The acquisition node number to work before node;Formula (1-20) indicate task immigration during and task immigration after the completion of initialization
The resource utilization of i-th of acquisition node is no more than the resource utilization upper limit set by user, u in acquisition nodei0Table
Show the resource utilization before the unallocated task of i-th of node, R in initialization acquisition nodeiIndicate initialization acquisition node
In the available resources total amount that configures on i-th of acquisition node, uhIndicate the node resource utilization rate upper limit of user setting, uwIt indicates
The node resource utilization threshold bandwidth of user setting, TiI-th of node in initialization acquisition node before expression task immigration
On acquisition tasks number,Acquisition tasks execute institute on i-th of node in initialization acquisition node after expression task immigration
The resource of consumption,Value determine that specific functional relation determines by step 1 by formula (1-21);Formula (1-22) expression task is moved
After shifting in initialization acquisition node i-th of node resource utilization;Initialization is adopted after formula (1-23) indicates task immigration
Collect the acquisition time time for executing primary whole acquisition tasks in node on i-th of acquisition nodeiNo more than set by user
The value of collection period Time, timei are determined that specific functional relation is determined by step 2 by formula (1-24);Formula (1-25) indicates new
The resource utilization of j-th of acquisition node in node is added no more than the resource utilization upper limit set by user, ujIt indicates
After the completion of task immigration in newly added node j-th of node resource utilization, uj0Indicate that j-th of node is not in newly added node
Resource utilization before distribution task, RjIndicate the available resources total amount configured on j-th of acquisition node in newly added node,Table
Show that acquisition tasks execute consumed resource on j-th of node in newly added node after task immigration,Value by formula (1-
26) it determines, specific functional relation is determined by step 1;J-th of acquisition section in newly added node after formula (1-27) expression task immigration
The acquisition time time of primary whole acquisition tasks is executed on pointjNo more than collection period Time, time set by userj's
Value is determined that specific functional relation is determined by step 2 by formula (1-28);Formula (1-29) indicates that i-th of node moves to j-th of section
The task number of point moves to the sum of initiating task number and Redundant task number of j-th of node equal to i-th of node,
taskK, i, jIt indicates to move to the initiating task that j-th of node serial number is k from i-th of node,It indicates from i-th of node
Move to the Redundant task that j-th of node serial number is l;Formula (1-30) indicates taskK, i, jValue, when number is that the initial of k is appointed
Business is located on i-th of node, and corresponding Redundant task is not located on j-th of node, and needs to be migrated to j-th of node
When, taskK, i, j=1, otherwise taskK, i, j=0, TaskiAcquisition tasks serial number before expression task immigration on i-th of acquisition node
Set, TaskjAcquisition tasks serial number set before expression task immigration on j-th of acquisition node, TaskijIndicate i-th of acquisition
Node moves to the task number set of j-th of acquisition node;Formula (1-31) indicatesValue, when number be the superfluous of l
Remaining task is located on i-th of node, and corresponding Redundant task is not located on j-th of node, and needs to be migrated to j-th
When node,Otherwise
In the present embodiment, Optimized model is as follows:
Min Cost=Δs U (2-22)
Wherein,
s.tT43×T34=0 (2-24)
(T31+T34)×T43=0 (2-25)
(T41+T43)×T34=0 (2-26)
Wherein,
Wherein,
time3≤1000 (2-33)
Wherein, time3=0.0880u3 2+2.0146u3+256.5218 (2-34)
time4≤1000 (2-35)
Wherein, time4=0.3159u4 2-10.3928u4+190.3659 (2-36)
Wherein,
time1≤1000 (2-39)
Wherein, time1=0.0869u1 2+2.0390u1+257.6073 (2-40)
Wherein,
Optimized model in step 8.4, solution procedure 8.3 obtains i-th of acquisition node and moves to j-th of acquisition node
Task number set Taskij;
In the present embodiment, a kind of feasible migration scheme for solving gained is:No. 3 acquisition node moves to No. 1 acquisition
The task number set Task of node31={ 11,12 }, No. 4 acquisition node move to the task number collection of No. 1 acquisition node
Close Task41={ 1,2,3,4,5,6,7,8 };
Step 8.5 is acquired the task immigration between node according to step 8.4 acquired results.
In the present embodiment, according to step 8.4 acquired results, by 11, No. 12 acquisition tasks on No. 3 acquisition nodes migrate to
No. 1 acquisition node being newly added migrates the 1-8 acquisition tasks on No. 4 acquisition nodes to No. 1 acquisition node being newly added.
Step 9, on there are acquisition node when important acquisition tasks acquisition failure, by the acquisition of its corresponding Redundant task
As a result it is stored in database from corresponding acquisition node;
In the present embodiment, it is assumed that No. 11 acquisition tasks acquisition failure on No. 3 acquisition nodes, by its corresponding Redundant task
11*It is stored in database from No. 4 acquisition nodes.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, it will be understood by those of ordinary skill in the art that:It still may be used
To modify to the technical solution recorded in previous embodiment, either which part or all technical features are equal
It replaces;And these modifications or replacements, model defined by the claims in the present invention that it does not separate the essence of the corresponding technical solution
It encloses.
Claims (3)
1. the method for scheduling task in a kind of industrial allocation formula data collecting system, it is characterised in that:This method includes following step
Suddenly:
Step 1, the correspondence for establishing each acquisition node acquisition resource and acquisition tasks;
Step 2, the correspondence for establishing each acquisition node acquisition time and acquisition tasks and node resource utilization rate;
Step 3, the correspondence for establishing task immigration is consumed on each acquisition node resource and migration task number;
Step 4, the correspondence for establishing the communication overhead of task immigration and migration task between each acquisition node;
Step 5, the minimum acquisition node number for determining initialization simultaneously open corresponding acquisition node, remaining acquisition node resource is made
It is spare for shared resource;
The Redundant task of vital task in initial acquisition task and which part initiating task is distributed to step 5 and is determined by step 6
Each acquisition node;
Acquisition node failure is judged whether, if so, thening follow the steps 7;Judge whether acquisition node overload or acquisition
Acquisition time is unsatisfactory for the case where collection period requirement on node, if so, thening follow the steps 8;Judge whether acquisition node
The case where upper important acquisition tasks acquisition failure, if so, thening follow the steps 9;If the above situation is not present, task scheduling knot
Beam;
Step 7, when there are acquisition node failure, be acquired task scheduling after node failure, specific dispatching method is:
Step 7.1, current collection period adopt the corresponding Redundant task of vital task in initiating task on fault collection node
Collect the database that result is stored in data collecting system from its corresponding acquisition node;
Step 7.2, next collection period are found and fault collection node resource residue situation from the acquisition node of Redundant backup
Most similar node, as the acquisition node being newly added, by fault collection node whole initiating tasks and Redundant task move
The acquisition node being newly added is moved to be acquired;
Step 8, when there are acquisition node overload or when being unsatisfactory for collection period requirement there are acquisition time on acquisition node, carrying out
Task scheduling after acquisition node overload, specific dispatching method are:
Step 8.1, the correspondence that step 4 is arrived according to step 1, establish the new Optimized model that acquisition node number is added, determine
The minimum value m for the acquisition node number being newly added;
Optimized model in step 8.2, solution procedure 8.1, acquires the minimum acquisition node number being newly added, and from Redundant backup
Acquisition node in select corresponding number acquisition node, be added collecting work;
Step 8.3 establishes acquisition node task immigration Optimized model;
Optimized model in step 8.4, solution procedure 8.3 obtains times that i-th of acquisition node moves to j-th of acquisition node
Serial number set of being engaged in Taskij;
Step 8.5 is acquired the task immigration between node according to step 8.4 acquired results;
Step 9, on there are acquisition node when important acquisition tasks acquisition failure, by the collection result of its corresponding Redundant task
The database of data collecting system is stored in from corresponding acquisition node.
2. the method for scheduling task in industrial allocation formula data collecting system according to claim 1, it is characterised in that:Institute
The detailed process for stating step 1 is:
Step 1.1 individually distributes acquisition tasks to each acquisition node, changes acquisition tasks number, measures a large amount of different acquisitions
Under task condition, the acquisition tasks number T that is consumed on i-th of acquisition nodeiAnd its acquisition resource utilization data ui, and should
The required acquisition time data time of whole acquisition tasks distributed on acquisition nodei;
Step 1.2, to the acquisition tasks data T on i-th of acquisition node obtained by step 1.1iWith corresponding resource consumption dataInto
Row fitting, acquires the functional relation of the two
In the step 2, to step 1.1 gained acquisition tasks number TiWith corresponding acquisition resource utilization data ui, acquisition time
Data timeiIt is fitted, acquires acquisition time data time on i-th of acquisition nodeiWith acquisition tasks number TiWith resource profit
With rate data uiFunctional relation timei=g (Ti, ui)。
3. the method for scheduling task in industrial allocation formula data collecting system according to claim 2, it is characterised in that:Institute
The Optimized model stated in step 8.1 is as follows:
min m
s.t Tij×Tji=0 i=1,2 ... n;J=1,2 ... n, j ≠ i (1-1)
Wherein,
timei≤ Time i=1,2 ... n (1-7)
Wherein,
Wherein,
timej≤ Time j=n+1, n+2 ... n+m (1-12)
Wherein,
Wherein, formula (1-1) moves to node j if indicating that there are tasks from node i, and there is no tasks to move to section from node j
Point i, to ensure the monotonicity of task immigration, TijExpression moves to the task number of j-th of node, T from i-th of nodejiTable
Show the task number that i-th of node is moved to from j-th of node;Formula (1-2) indicates that a node cannot not only move out task but also move
Enter task, n indicates the acquisition node number to work before newly added node;Formula (1-3) indicate task immigration during and task
After the completion of migration in initialization acquisition node the resource utilization of i-th of acquisition node no more than the resource set by user
The utilization rate upper limit, ui0Indicate the resource utilization before the unallocated task of i-th of node, R in initialization acquisition nodeiIt indicates just
The available resources total amount configured on i-th of acquisition node in beginning work acquisition node, uhIndicate the node resource profit of user setting
With the rate upper limit, uwIndicate the node resource utilization threshold bandwidth of user setting, TiInitialization acquisition saves before indicating task immigration
Acquisition tasks number in point on i-th of node,It is adopted on i-th of node in initialization acquisition node after expression task immigration
Set task executes consumed resource,Indicate after task immigration in initialization acquisition node acquisition tasks on i-th of node
The consumed resource of migration,WithValue determined respectively by formula (1-4) and formula (1-5), specific functional relation by step 1 and
Step 3 determines;Formula (1-6) indicates the resource utilization of i-th of node in initialization acquisition node after the completion of task immigration;Formula
After (1-7) expression task immigration adopting for primary whole acquisition tasks is executed on i-th of acquisition node in initialization acquisition node
Collect time timeiNo more than collection period Time, time set by useriInitialization acquisition node after expression task immigration
In the acquisition times of primary whole acquisition tasks, time are executed on i-th of nodeiValue by formula (1-8) determine, specific function
Relationship is determined by step 2;Formula (1-9) indicates the resource utilization of j-th of acquisition node in newly added node no more than user
The set resource utilization upper limit, ujIndicate the resource utilization of j-th of node in newly added node after the completion of task immigration,
uj0Indicate the resource utilization before the unallocated task of j-th of node, R in newly added nodejIt indicates to adopt for j-th in newly added node
The available resources total amount configured on collection node,Acquisition tasks are held on j-th of node in newly added node after expression task immigration
The consumed resource of row,Acquisition tasks migrate consumed money on j-th of node in newly added node after expression task immigration
Source,WithValue determine that specific functional relation is determined by step 1 and step 3 by formula (1-10) and formula (1-11) respectively;Formula
Execute the acquisition time of primary whole acquisition tasks after (1-12) expression task immigration in newly added node on j-th of acquisition node
No more than collection period Time, time set by userjIt is executed on j-th of node in newly added node after expression task immigration
The acquisition time of primary whole acquisition tasks, timejValue determine that specific functional relation determines by step 2 by formula (1-13);
Acquisition node task immigration Optimized model in the step 8.3 is as follows:
Wherein,
s.t Tij×Tji=0 i=1,2 ... n;J=1,2 ... n, j ≠ i (1-18)
Wherein,
timei≤ Time i=1,2 ... n (1-23)
Wherein,
Wherein,
timej≤ Time j=n+1, n+2 ... n+m (1-27)
Wherein,
Wherein,
Wherein, the resource of each node and communication overhead and task immigration are complete during formula (1-14) expression minimum task immigration
At the load difference between rear each acquisition node and Δ U, p (Tij) communication overhead of task immigration between expression node i and node j,
Specific functional relation is determined by step 4, and poor sum, value are loaded between each acquisition node after the completion of Δ U expression task immigrations
It is determined by formula (1-17);The task number that formula (1-29) i-th of node of expression moves to j-th of node is moved equal to i-th of node
Move on to the sum of initiating task number and the Redundant task number of j-th of node, taskK, i, jIt indicates to move to jth from i-th of node
A node serial number is the initiating task of k,Indicate that moving to the redundancy that j-th of node serial number is l from i-th of node appoints
Business;Formula (1-30) indicates taskK, i, jValue, when the initiating task that number is k is located on i-th node, corresponding redundancy
Task is not located on j-th node, and when needing to be migrated to j-th of node, taskK, i, j=1, otherwise taskK, i, j=0,
TaskiAcquisition tasks serial number set before expression task immigration on i-th of acquisition node, TaskjJ-th before expression task immigration
Acquisition tasks serial number set on acquisition node, TaskijIndicate that i-th of acquisition node moves to the task of j-th of acquisition node
Serial number set;Formula (1-31) indicatesValue, it is corresponding when the Redundant task that number is l is located on i-th node
Redundant task is not located on j-th node, and when needing to be migrated to j-th of node,Otherwise
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810384049.XA CN108536539B (en) | 2018-04-26 | 2018-04-26 | Task scheduling method in industrial distributed data acquisition system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810384049.XA CN108536539B (en) | 2018-04-26 | 2018-04-26 | Task scheduling method in industrial distributed data acquisition system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108536539A true CN108536539A (en) | 2018-09-14 |
CN108536539B CN108536539B (en) | 2021-12-17 |
Family
ID=63478226
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810384049.XA Active CN108536539B (en) | 2018-04-26 | 2018-04-26 | Task scheduling method in industrial distributed data acquisition system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108536539B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109634915A (en) * | 2018-11-28 | 2019-04-16 | 深圳市网心科技有限公司 | File dispositions method, Cloud Server, system and storage medium |
CN110377430A (en) * | 2019-07-24 | 2019-10-25 | 中南民族大学 | Data migration method, equipment, storage medium and device |
CN110389555A (en) * | 2019-08-16 | 2019-10-29 | 山东省金曼克电气集团股份有限公司 | A kind of monitoring system and monitoring method of marine transformer |
CN111142788A (en) * | 2019-11-29 | 2020-05-12 | 浪潮电子信息产业股份有限公司 | Data migration method and device and computer readable storage medium |
CN112753018A (en) * | 2020-03-04 | 2021-05-04 | 深圳市大疆创新科技有限公司 | Method, system and storage medium for adaptive load balancing |
CN114374695A (en) * | 2021-11-26 | 2022-04-19 | 智昌科技集团股份有限公司 | Operation method and system of distributed data acquisition system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1756190A (en) * | 2004-09-30 | 2006-04-05 | 北京航空航天大学 | Distributed performance data acquisition method |
US20140207920A1 (en) * | 2013-01-22 | 2014-07-24 | Hitachi, Ltd. | Virtual server migration plan making method and system |
CN104702598A (en) * | 2015-02-16 | 2015-06-10 | 南京邮电大学 | Distributed network protocol security detection method for smart power grid |
CN105527948A (en) * | 2015-12-11 | 2016-04-27 | 东北大学 | Large scale distributed data acquisition system and method based on industrial process |
CN106126346A (en) * | 2016-07-05 | 2016-11-16 | 东北大学 | A kind of large-scale distributed data collecting system and method |
CN106357426A (en) * | 2016-08-26 | 2017-01-25 | 东北大学 | Large-scale distribution intelligent data collection system and method based on industrial cloud |
CN107918561A (en) * | 2017-11-17 | 2018-04-17 | 东北大学 | A kind of method for allocating tasks in industrial allocation formula data collecting system |
-
2018
- 2018-04-26 CN CN201810384049.XA patent/CN108536539B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1756190A (en) * | 2004-09-30 | 2006-04-05 | 北京航空航天大学 | Distributed performance data acquisition method |
US20140207920A1 (en) * | 2013-01-22 | 2014-07-24 | Hitachi, Ltd. | Virtual server migration plan making method and system |
CN104702598A (en) * | 2015-02-16 | 2015-06-10 | 南京邮电大学 | Distributed network protocol security detection method for smart power grid |
CN105527948A (en) * | 2015-12-11 | 2016-04-27 | 东北大学 | Large scale distributed data acquisition system and method based on industrial process |
CN106126346A (en) * | 2016-07-05 | 2016-11-16 | 东北大学 | A kind of large-scale distributed data collecting system and method |
CN106357426A (en) * | 2016-08-26 | 2017-01-25 | 东北大学 | Large-scale distribution intelligent data collection system and method based on industrial cloud |
CN107918561A (en) * | 2017-11-17 | 2018-04-17 | 东北大学 | A kind of method for allocating tasks in industrial allocation formula data collecting system |
Non-Patent Citations (1)
Title |
---|
吴永建 等: "流程工业中设备逻辑控制程序测试平台的建立", 《第25届中国过程控制会议》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109634915A (en) * | 2018-11-28 | 2019-04-16 | 深圳市网心科技有限公司 | File dispositions method, Cloud Server, system and storage medium |
CN110377430A (en) * | 2019-07-24 | 2019-10-25 | 中南民族大学 | Data migration method, equipment, storage medium and device |
CN110389555A (en) * | 2019-08-16 | 2019-10-29 | 山东省金曼克电气集团股份有限公司 | A kind of monitoring system and monitoring method of marine transformer |
CN111142788A (en) * | 2019-11-29 | 2020-05-12 | 浪潮电子信息产业股份有限公司 | Data migration method and device and computer readable storage medium |
CN111142788B (en) * | 2019-11-29 | 2021-10-15 | 浪潮电子信息产业股份有限公司 | Data migration method and device and computer readable storage medium |
CN112753018A (en) * | 2020-03-04 | 2021-05-04 | 深圳市大疆创新科技有限公司 | Method, system and storage medium for adaptive load balancing |
CN112753018B (en) * | 2020-03-04 | 2022-06-10 | 深圳市大疆创新科技有限公司 | Method, system and storage medium for adaptive load balancing |
CN114374695A (en) * | 2021-11-26 | 2022-04-19 | 智昌科技集团股份有限公司 | Operation method and system of distributed data acquisition system |
CN114374695B (en) * | 2021-11-26 | 2022-06-28 | 智昌科技集团股份有限公司 | Operation method and system of distributed data acquisition system |
Also Published As
Publication number | Publication date |
---|---|
CN108536539B (en) | 2021-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108536539A (en) | A kind of method for scheduling task in industrial allocation formula data collecting system | |
Shi et al. | Empirical evaluation of vector bin packing algorithms for energy efficient data centers | |
CN102521044B (en) | Distributed task scheduling method and system based on messaging middleware | |
CN104050042B (en) | The resource allocation methods and device of ETL operations | |
CN105468435A (en) | NFV dynamic resource distribution method | |
AU2015301411A1 (en) | Systems and methods for auto-scaling a big data system | |
CN107291550B (en) | A kind of Spark platform resource dynamic allocation method and system for iterated application | |
CN103634394A (en) | Data flow processing-oriented elastic expandable resource managing method and system | |
Zhao et al. | A data placement strategy for data-intensive scientific workflows in cloud | |
CN102983996A (en) | Dynamic allocation method and system for high-availability cluster resource management | |
Wang et al. | Hybrid pulling/pushing for i/o-efficient distributed and iterative graph computing | |
CN108234356A (en) | Optimization application resource Distribution Strategy based on application relational network | |
TW201702908A (en) | Method and apparatus for flexible scheduling in a database system for reducing computation complexity during a database flexible scheduling process, and meanwhile achieving objects of saving resources as well as reducing scheduling cost | |
CN116777182B (en) | Task dispatch method for semiconductor wafer manufacturing | |
CN106681823A (en) | Load balancing method for processing MapReduce data skew | |
CN106033442A (en) | Parallel breadth-first search method based on shared memory system structure | |
CN109739634A (en) | A kind of atomic task execution method and device | |
Zhang et al. | Design and implementation of cloud-based performance testing system for web services | |
Zhou et al. | Simulation model of dynamic service scheduling in cloud manufacturing | |
CN103973811A (en) | High-availability cluster management method capable of conducting dynamic migration | |
CN110399206B (en) | IDC virtualization scheduling energy-saving system based on cloud computing environment | |
CN105468451A (en) | Job scheduling system of computer cluster on the basis of high-throughput sequencing data | |
CN110362383A (en) | The VM moving method of P-E tradeoff towards seasonal form non-stationary concurrency | |
Sun et al. | Probability-Based Online Algorithm for Switch Operation of Energy Efficient Data Center | |
Sun et al. | T2FA: A heuristic algorithm for deadline-constrained workflow scheduling in cloud with multicore resource |
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 |