CN105607956A - Task allocation method and system in computer - Google Patents
Task allocation method and system in computer Download PDFInfo
- Publication number
- CN105607956A CN105607956A CN201610009481.1A CN201610009481A CN105607956A CN 105607956 A CN105607956 A CN 105607956A CN 201610009481 A CN201610009481 A CN 201610009481A CN 105607956 A CN105607956 A CN 105607956A
- Authority
- CN
- China
- Prior art keywords
- task
- queue
- module
- pool
- computer
- 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]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
- Multi Processors (AREA)
Abstract
The invention provides a task allocation method and system in a computer. The task allocation method and system are used for allocating tasks in m task pools to p processes, and each task pool is provided with n storage units. The method includes the steps that firstly, the p processes are arranged in sequence, and a process queue is obtained; then, the tasks are allocated to the processes in the arrangement order of rows and columns in a two-dimensional matrix formed by m*n storage units in the m task pools. According to the task allocation method and system, the storage units of multiple task pools are regarded as the two-dimensional matrix, then elements in the matrix are allocated to the processes in sequence, and in this way, it can be guaranteed that the number of processes owned by each task pool and the task pools occupied by each process are balanced, so that the load of the processes is balanced, and it is guaranteed that the tasks are executed efficiently.
Description
Technical field
The invention belongs to computer realm, relate in particular to the method for allocating tasks in a kind of computer and beSystem.
Background technology
In software systems, there are many business scenarios all to need Timing Processing task, process these tasksMain body is born by " process " conventionally. In the situation that task amount is little, one " process " can completeThese tasks. But large at task amount, in the high situation of reliability requirement, have to adopt distributed manyThe pattern of thread is processed. When it come to arrive multithread mode, how that has just involved task at thisThe problem of distributing between " process " a bit.
Quartz is that OpenSymphony increases income and is organized in a Java in Jobscheduling fieldThe project of increasing income. The core of Quartz framework is scheduler, ring when it is in charge of Quartz application operationBorder. Scheduler is not to do on one's own account all work, but relies on some very important portions in frameworkPart. Quartz is not only process and management of process. For guaranteeing scalability, Quartz has adopted baseIn the framework of multi-process. When startup, framework initializes a set of worker process, and this cover process is scheduledDevice is used for carrying out predetermined task, and Quartz just can the multiple tasks of concurrent running like this. Quartz relies onThe process pool management component of a set of loose coupling carrys out management thread environment.
In brief, Quartz is a distributed timer framework that can the multiple tasks of parallel running,Mature and stable and use extensively. It by relevant database and JDBC job storage share data withRealize distributed object.
Quartz is powerful, but just because of its power, causes to the dependence of environment veryHarshness, resource consumption is also higher. Distributed deployment is done in usage data storehouse simultaneously, and efficiency is also lower.Its group scheme and configuration are also more loaded down with trivial details, have certain left-hand seat difficulty concerning new hand. In a word,Quartz is the framework that the support distributed task scheduling of a heavyweight is carried out, and needs cost but useCertain learning cost, although reliably not efficient.
Summary of the invention
(1) technical problem that will solve
The object of the invention is to, method for allocating tasks and system in a kind of computer are provided, realizeThe load balancing of process, and ensure that task is efficiently carried out.
(2) technical scheme
The invention provides the method for allocating tasks in a kind of computer, for appointing m task poolP process distributed in business, and m task pool is arranged in order, and it is single that each task pool has n storageUnit, method comprises:
S1, arranges in order to p process, obtains process queue;
S2, distributes to k process by the task in j memory cell of i task pool,IfRemainder equal 0, k equals p, otherwise k equalsRemainder,M, n, p, i, j, k are the integer that is more than or equal to 1, wherein, and 1≤i≤m, 1≤j≤n, 1≤k≤p。
The present invention also provides the distribution system of the task in a kind of computer, for by m task poolTask is distributed to p process, and m task pool is arranged in order, and it is single that each task pool has n storageUnit, system comprises:
Process order module, arranges in order to p process, obtains process queue;
Task distribution module, distributes to k by the task in j memory cell of i task poolIndividual process, ifRemainder equal 0, k equals p, otherwise k equals'sRemainder, m, n, p, i, j, k are the integer that is more than or equal to 1, wherein, 1≤i≤m, 1≤j≤ n, 1≤k≤p. , wherein, 1≤i≤m, 1≤j≤n, 1≤k≤p.
(3) beneficial effect
The present invention regards the memory cell of multiple task pools as a two-dimensional matrix, then by matrixElement distribute in order process, make like this relations of distribution cheer and bright, be easy to safeguard and monitoring;In addition, by the method for salary distribution above, make the process number that each task pool has, each processThe task pool occupying can ensure equilibrium, thereby the load balancing of implementation process and guarantee task are by efficientCarry out.
Brief description of the drawings
Fig. 1 is the schematic diagram of the relations of distribution between task pool of the present invention and process.
Fig. 2 is the schematic diagram that carries out task distribution according to the relations of distribution of Fig. 1.
Fig. 3 is the flow chart of the method for allocating tasks in the computer that provides of the embodiment of the present invention.
Fig. 4 is the structural representation of the task distribution system in the computer that provides of the embodiment of the present invention.
Detailed description of the invention
The invention provides method for allocating tasks and system in a kind of computer, for by m task poolIn task distribute to p process, each task pool has n memory cell, method is first to pIndividual process is arranged in order, obtains process queue, then individual at the common m × n of m task poolIn the two-dimensional matrix of memory cell composition, assign the task to process according to putting in order of ranks. ThisInvention is to regard the memory cell of multiple task pools as a two-dimensional matrix, then by the unit in matrixElement is distributed to process in order, makes like this accounting for of the process number that each task pool has, each processSome task pools can ensure equilibrium, thereby the load balancing of implementation process and guarantee task are efficiently heldOK.
According to one embodiment of the present invention, task is carried out to centralized buffer memory, namely instituteSome tasks in unified buffer memory, but and are not meant that these tasks are all placed on one by centralized storesIn buffer queue, but be dispersed in multiple " task pools " according to hash algorithm, operation system canUse " task management interface " to increase newly or deletion task in task pool. A task pool just represents oneIndividual buffer queue, and the number of task pool configures according to portfolio. Like this by disperseing task poolCan improve system reliability, and adopt the mode of buffer memory to store the storage of greatly lifting task, look intoInquiry, deletion efficiency. The present invention regards m × n memory cell in m task pool as a two dimensionMatrix, method for allocating tasks comprises:
S1, arranges in order to p process, obtains process queue;
S2, in the two-dimensional matrix forming in common m × n memory cell of m task pool, according to rowPutting in order of row assigns the task to process, is specially: by j storage of i task poolTask in unit is distributed to k process, wherein, and wherein, ifRemainder equal0, k equals p, otherwise k equalsRemainder, m, n, p, i, j, k are greatlyIn the integer that equals 1, wherein, 1≤i≤m, 1≤j≤n, 1≤k≤p.
Principle of the present invention is, the process number of establishing is P, and task pool adds up to B, the entering of each task poolNumber of passes is A=P/B, if A was less than for 1 (task pool sum is greater than survival process number), makes A=1,Make like this survival process can be evenly distributed in task pool as much as possible, (appoint if A is greater than 1Business pond sum is less than in survival process number), A is carried out to round, this is in order not allowProcess is carried out too much leaving unused, thereby makes process load balancing.
As shown in Figure 1-2, b0-b4 represents 5 task pools, and each task pool has N+1 storageUnit, i.e. the memory location of s0-sN, has formed two-dimensional matrix like this, for convenience of description, by squareEach element of battle array represents by numbering, in the time that task is distributed, first distributes the element of the first row of the first row,The element of the secondary series of reallocation the first row ... by that analogy, press the numerical order pair of 0-14 in figureEach element distributes. As shown in Figure 2, suppose to have 5 processes and 5 task pools, need to divideJoin the element that is numbered 0-14, in distribution, for convenience of description, can be decomposed into three phases, headFirst, by element in figure 0,1,2,3,4 distribute to respectively the 1st, 2,3,4,5 processes, then, by unit in figureElement 5,6,7,8,9 distributes to respectively the 1st, and 2,3,4,5 processes are last, and by 10,11,12,13,14 respectivelyDistribute to the 1st, 2,3,4,5 processes, like this, 5 processes have all obtained that 5 task pools distributeTask, and the task quantity homogeneous phases that each process obtains etc., have realized load balancing.
Such method of salary distribution, ensures that process number and the each process in each task pool is sharedTask pool can ensure equilibrium, thereby the load balancing of implementation process and guarantee task are efficiently carried out.
According to one embodiment of the present invention, after step S2, if in p process oneProcess finishes, and in process queue, delete this process, and make p=p-1, then, execution step S1～S2; Or, if there is new process initiation, this is added new process to described process queue,And make p=p+1, then, carry out described step S1～S2. The p=p-1 is here the numerical value of instigating pSubtract the numerical value that 1, p=p+1 refers to p and add 1, in the time of execution step S1～S2, can make the number of p soBe worth consistent with actual process number.
According to one embodiment of the present invention, before step S1, the each process in p processIn the time starting, it is identifier of this course allocation. In the time sorting, obtain in p process everyThe identifier of individual process, according to the size of identifier, arranges p process, obtains process teamRow, using first process in process queue as main thread, for performing step S2. Identifier canTo be a bill, bill is the name of process, is the numbering of a numeric type, dull growth, noCan repeat, as shown in table 1.
| Bill | 1 | 2 | … | N |
| Process 0 | Process 1 | Process 2 | … |
Table 1
Each process can be obtained a bill in the time starting, and obtains bill rule as follows: from numbering 0Start whether search successively corresponding bill occupied, until find " N " not have when occupied, thatThe success of neck ticket, " N " be the bill of this process namely. After " success of neck ticket ", also need carrying outCheck whether " master note " of this bill, number minimum bill, like this in process queue,It also can be arranged on first process, thereby as host process. And host process is at execution step S2Time, being actually the preservation relations of distribution, the relations of distribution refer to the unit in two-dimensional array as shown in Figure 1The corresponding relation of element and process, main thread is according to this corresponding relation, and it is right that the task in element is distributed toThe process of answering.
According to one embodiment of the present invention, when each process initiation, can open a timer fixedTime go to attempt safeguarding all relations of distribution, the object designing be like this ensure in original host process differentIn normal situation, the process of new bill minimum will become host process and continue to safeguard the relations of distribution.
According to one embodiment of the present invention, host process is not to go repeatedly to carry out task distribution,But be only just to carry out task distribution in load inequality. In the time of original allocation task, such as survive intoNumber of passes is P, and task pool adds up to B, follow-up because of newly-increased process, the process exception process number that causes survivingAmount changes (the survival information of process is placed in unique caching), bearing while causing original allocation taskCarry no longer equilibrium, host process just can perform step S1～S2.
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with concrete realityExecute example, and with reference to accompanying drawing, the present invention is described in more detail.
Fig. 3 is the method for allocating tasks in the computer that provides of the embodiment of the present invention, and method comprises:
S1, the each process in p process, in the time starting, for identifier of this course allocation, obtainsGet the identifier of each process in p process, according to the size of identifier, p process arrangedRow, obtain process queue, and using first process in process queue as main thread, for carrying outStep S2;
S2, distributes to k process by the task in j memory cell of i task pool,IfRemainder equal 0, k equals p, otherwise k equalsRemainder,M, n, p, i, j, k are the integer that is more than or equal to 1, wherein, and 1≤i≤m, 1≤j≤n, 1≤ k≤p. If a process in p process finishes, in process queue, delete this process, andMake p=p-1, then, execution step S1～S2; Or, if there is new process initiation, this is addedEnter new process to process queue, and make p=p+1, then, execution step S1～S2.
Fig. 4 is the task distribution system in the computer that provides of the embodiment of the present invention, comprising:
Identifier allocation module in the time starting, is this process for the each process p processDistribute an identifier.
Process order module, for obtaining the identifier of described p the each process of process, according to instituteState the size of identifier, a described p process is arranged, obtain process queue, and by process teamFirst process in row is as main thread.
Task distribution module, carries out task distribution by carrying out this main thread, specifically comprises: byTask in j memory cell of i task pool is distributed to k process, wherein, ifRemainder equal 0, k equals p, otherwise k equalsRemainder, m,N, p, i, j, k are the integer that is more than or equal to 1, wherein, and 1≤i≤m, 1≤j≤n, 1≤k≤p。
Management of process module, for after carrying out task distribution, if in a described p process oneProcess finishes, and in this process queue, deletes this process, and makes p=p-1; Or, for enteringAfter row task is distributed, if there is new process initiation, this new process is added to described process teamIn row, and make p=p+1.
In sum, the present invention can be applicable in large-scale timing task processing system, have efficiently,Reliably characteristic, the many abnormal conditions such as machine of can answering system extremely delaying, accomplish runtime self adaptationAdjust, can ensure promptness and the harmony of task processing simultaneously by core allocation algorithm, also canEnough each distributed system resources of executing the task that keeps are rationally utilized.
Above-described specific embodiment, carries out object of the present invention, technical scheme and beneficial effectFurther description, institute it should be understood that the foregoing is only specific embodiments of the invention and, be not limited to the present invention, within the spirit and principles in the present invention all, any repairing of doingProtection scope of the present invention changes, be equal to replacement, improvement etc., within all should be included in.
Claims (12)
1. the method for allocating tasks in computer, for distributing the task of m task poolGive p process, it is characterized in that, a described m task pool is arranged in order, and each task pool has nIndividual memory cell, method comprises:
S1, arranges in order to a described p process, obtains process queue;
S2, distributes to k process by the task in j memory cell of i task pool,Wherein, ifRemainder equal 0, k equals p, otherwise k equals'sRemainder, m, n, p, i, j, k are the integer that is more than or equal to 1, wherein, 1≤i≤m, 1≤j≤n，1≤k≤p。
2. the method for allocating tasks in computer according to claim 1, is characterized in that,After described step S2, if a process in a described p process finishes, in process queueThis process of middle deletion, and make p=p-1, then, carry out described step S1～S2.
3. the method for allocating tasks in computer according to claim 1, is characterized in that,After described step S2, if there is new process initiation, this is added new process to described in enterIn journey queue, and make p=p+1, then, carry out described step S1～S2.
4. the method for allocating tasks in computer according to claim 1, is characterized in that,Before described step S1, comprising:
Each process in a described p process, in the time starting, is identifier of this course allocation.
5. the method for allocating tasks in computer according to claim 4, is characterized in that,Described step S1 comprises:
Obtain the identifier of each process in a described p process, according to the size of described identifier, rightA described p process is arranged, and obtains process queue.
6. the method for allocating tasks in computer according to claim 1, is characterized in that,
In described step S1, using first process in described process queue as main thread, useIn carrying out described step S2.
7. the task distribution system in computer, for distributing the task of m task poolGive p process, it is characterized in that, a described m task pool is arranged in order, and each task pool has nIndividual memory cell, system comprises:
Process order module, arranges in order to a described p process, obtains process queue;
Task distribution module, distributes to k by the task in j memory cell of i task poolIndividual process, wherein, ifRemainder equal 0, k equals p, otherwise k equalsRemainder, m, n, p, i, j, k are the integer that is more than or equal to 1, wherein, 1≤i≤m，1≤j≤n，1≤k≤p。
8. the task distribution system in computer according to claim 7, is characterized in that,Also comprise:
Management of process module, for after carrying out task distribution, if in a described p process oneProcess finishes, and in this process queue, deletes this process, and makes p=p-1.
9. the task distribution system in computer according to claim 7, is characterized in that,Also comprise:
Management of process module, for after carrying out task distribution, if there is new process initiation, willThis new process is added in described process queue, and makes p=p+1.
10. the task distribution system in computer according to claim 7, is characterized in that,Also comprise:
Identifier allocation module in the time starting, is this for the each process in a described p processIdentifier of course allocation.
Task distribution system in 11. computers according to claim 10, is characterized in that,Described process order module is obtained the identifier of each process in a described p process, according to described markThe size of symbol, arranges a described p process, obtains process queue.
Task distribution system in 12. computers according to claim 7, is characterized in that,
Described process order module is using first process in described process queue as main thread, described inTask distribution module is carried out task distribution by carrying out this main thread.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610009481.1A CN105607956B (en) | 2016-01-06 | 2016-01-06 | Method for allocating tasks and system in a kind of computer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610009481.1A CN105607956B (en) | 2016-01-06 | 2016-01-06 | Method for allocating tasks and system in a kind of computer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105607956A true CN105607956A (en) | 2016-05-25 |
| CN105607956B CN105607956B (en) | 2019-06-04 |
Family
ID=55987910
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610009481.1A Active CN105607956B (en) | 2016-01-06 | 2016-01-06 | Method for allocating tasks and system in a kind of computer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105607956B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106201696A (en) * | 2016-07-15 | 2016-12-07 | 东软集团股份有限公司 | Method and apparatus for thread |
| CN106375372A (en) * | 2016-08-23 | 2017-02-01 | 东方网力科技股份有限公司 | Big data resource allocation method and device |
| CN106803841A (en) * | 2017-02-14 | 2017-06-06 | 北京奇虎科技有限公司 | The read method of message queue data, device and distributed data-storage system |
| CN108848150A (en) * | 2018-06-06 | 2018-11-20 | 郑州云海信息技术有限公司 | A kind of plan target method and system in cloud management platform |
| CN110795222A (en) * | 2019-10-25 | 2020-02-14 | 北京浪潮数据技术有限公司 | Multithreading task scheduling method, device, equipment and readable medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050034130A1 (en) * | 2003-08-05 | 2005-02-10 | International Business Machines Corporation | Balancing workload of a grid computing environment |
| US20120291041A1 (en) * | 2011-05-11 | 2012-11-15 | James Adam Cipar | Assigning resources for tasks |
| CN102929707A (en) * | 2012-11-06 | 2013-02-13 | 无锡江南计算技术研究所 | Parallel task dynamical allocation method |
| CN103279390A (en) * | 2012-08-21 | 2013-09-04 | 中国科学院信息工程研究所 | Parallel processing system for small operation optimizing |
| CN104700255A (en) * | 2013-12-06 | 2015-06-10 | 腾讯科技（北京）有限公司 | Multi-process processing method, device and system |
-
2016
- 2016-01-06 CN CN201610009481.1A patent/CN105607956B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050034130A1 (en) * | 2003-08-05 | 2005-02-10 | International Business Machines Corporation | Balancing workload of a grid computing environment |
| US20120291041A1 (en) * | 2011-05-11 | 2012-11-15 | James Adam Cipar | Assigning resources for tasks |
| CN103279390A (en) * | 2012-08-21 | 2013-09-04 | 中国科学院信息工程研究所 | Parallel processing system for small operation optimizing |
| CN102929707A (en) * | 2012-11-06 | 2013-02-13 | 无锡江南计算技术研究所 | Parallel task dynamical allocation method |
| CN104700255A (en) * | 2013-12-06 | 2015-06-10 | 腾讯科技（北京）有限公司 | Multi-process processing method, device and system |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106201696A (en) * | 2016-07-15 | 2016-12-07 | 东软集团股份有限公司 | Method and apparatus for thread |
| CN106375372A (en) * | 2016-08-23 | 2017-02-01 | 东方网力科技股份有限公司 | Big data resource allocation method and device |
| CN106375372B (en) * | 2016-08-23 | 2019-12-06 | 东方网力科技股份有限公司 | big data resource allocation method and device |
| CN106803841A (en) * | 2017-02-14 | 2017-06-06 | 北京奇虎科技有限公司 | The read method of message queue data, device and distributed data-storage system |
| CN106803841B (en) * | 2017-02-14 | 2020-04-10 | 北京奇虎科技有限公司 | Method and device for reading message queue data and distributed data storage system |
| CN108848150A (en) * | 2018-06-06 | 2018-11-20 | 郑州云海信息技术有限公司 | A kind of plan target method and system in cloud management platform |
| CN110795222A (en) * | 2019-10-25 | 2020-02-14 | 北京浪潮数据技术有限公司 | Multithreading task scheduling method, device, equipment and readable medium |
| CN110795222B (en) * | 2019-10-25 | 2022-03-22 | 北京浪潮数据技术有限公司 | Multithreading task scheduling method, device, equipment and readable medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105607956B (en) | 2019-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8595732B2 (en) | Reducing the response time of flexible highly data parallel task by assigning task sets using dynamic combined longest processing time scheme | |
| CN105607956A (en) | Task allocation method and system in computer | |
| Razaque et al. | Task scheduling in cloud computing | |
| Khorasani et al. | Scalable simd-efficient graph processing on gpus | |
| US9619430B2 (en) | Active non-volatile memory post-processing | |
| Farzanyar et al. | Efficient mining of frequent itemsets in social network data based on MapReduce framework | |
| CN102708009B (en) | Method for sharing GPU (graphics processing unit) by multiple tasks based on CUDA (compute unified device architecture) | |
| CN104462302A (en) | Distributed data processing coordination method and system | |
| US8458136B2 (en) | Scheduling highly parallel jobs having global interdependencies | |
| US20160034528A1 (en) | Co-processor-based array-oriented database processing | |
| CN102831102A (en) | Method and system for carrying out matrix product operation on computer cluster | |
| US10275289B2 (en) | Coexistence of message-passing-like algorithms and procedural coding | |
| CN103685492A (en) | Dispatching method, dispatching device and application of Hadoop trunking system | |
| US10740317B2 (en) | Using message-passing with procedural code in a database kernel | |
| Liu et al. | KubFBS: A fine‐grained and balance‐aware scheduling system for deep learning tasks based on kubernetes | |
| CN101969402B (en) | Data exchanging method based on parallel processing | |
| US10268727B2 (en) | Batching tuples | |
| CN107038072A (en) | Method for scheduling task and device based on Hadoop system | |
| Khalil et al. | Survey of Apache Spark optimized job scheduling in Big Data | |
| CN114896295A (en) | Data desensitization method, desensitization device and desensitization system in big data scene | |
| Sun et al. | Job Performance Optimization Method Based on Data Balance in the Wireless Sensor Networks. | |
| Babu et al. | Dynamic colocation algorithm for Hadoop | |
| CN111930495A (en) | Big data platform storage isolation and calculation isolation method in software as a service (SaaS) mode | |
| US9760372B2 (en) | Parallel processing in plural processors with result register each performing associative operation on respective column data | |
| Zhao et al. | TDWS: A job scheduling algorithm based on MapReduce |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |