CN108228329A - A kind of packet scheduling system - Google Patents
A kind of packet scheduling system Download PDFInfo
- Publication number
- CN108228329A CN108228329A CN201810107836.XA CN201810107836A CN108228329A CN 108228329 A CN108228329 A CN 108228329A CN 201810107836 A CN201810107836 A CN 201810107836A CN 108228329 A CN108228329 A CN 108228329A
- Authority
- CN
- China
- Prior art keywords
- scheduling
- grouping
- control module
- job
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of packet scheduling system, including:Grouping management module, schedule job control module, scheduling perform control module;Operation with identical scheduling rule is abstracted into a complete grouping by grouping management module;Schedule job control module relies on the function of grouping, provides a variety of executive modes;Scheduling performs control module according to the executing rule being configured in schedule job execution module, creates a job execution pond, scheduling performs control module module and performed from the job execution pond according to scheduling time and dependence timing.Packet scheduling system provided by the invention, optimize existing scheduling implementation strategy, by building the packing of grouping by the operation progress of multiple identical scheduling rules in logic, so as to simplify the configuration of operation, the unified management of operation is facilitated, at the same time aspect maintenance work in later stage etc..
Description
Technical field
The present invention relates to a kind of scheduling systems, and in particular to a kind of by building grouping by multiple identical scheduling rules
Operation carries out packing in logic, so as to simplify the configuration of operation, facilitates the packet scheduling system of the unified management of operation.
Background technology
Scheduling main function of system is to start correct operation at correct time point, it is ensured that operation is relied on according to correct
The execution of relationship promptly and accurately.It is played a crucial role in business data platform.In the big data epoch, due to data
Source is numerous and diverse, and data class is various, and data dependence relation is complicated, and schedule job is more, how quickly to improve allocative efficiency, and simplification is matched
Put the problem of being our urgent need to resolve.
Dispatching technique at this stage in predominantly dispatching each operation a scheduler is configured, i.e., no matter operation how much, often
All there are one independent schedulers periodically to perform operation for a operation, and at the same time there may be certain dependence passes between operation
System when scheduling rule meets simultaneously simultaneously for operation, successively performs according to dependence.It staggers the time as retribution for sins, needs individually weight
New configuration schedules mode, performs schedule job.
Mainstream scheduling tool is broadly divided into two kinds at this stage, and one kind is CS frameworks, provides visualization interface, is dragged by dilatory
Mode configuration schedules operation, another is B/S framework, and based on Web configuration schedules operations, the two utilizes timer timing
Operation is performed, schedule job is then performed by certain dependence.Each operation suffers from oneself independent scheduling rule,
No matter whether the rule repeats.So when multiple and different operations are configured, and these job schedulings rule it is identical when, Jiu Huicun
Regular allocation problem is being repeated several times, is greatly reducing the efficiency of configuration.Simultaneously when job patch reports an error in scheduling, weight is needed
Compounding puts the rule of multiple schedule jobs to realize that operation is run again, this also increases the workload of later maintenance.
Invention content
It is in view of the above-mentioned problems, a kind of by building grouping by multiple identical scheduling it is a primary object of the present invention to provide
The operation of rule carries out packing in logic, so as to simplify the configuration of operation, facilitates the grouping tune of the unified management of operation
Degree system.
The present invention is to solve above-mentioned technical problem by following technical proposals:A kind of packet scheduling system, described point
Group scheduling system includes:Grouping management module, schedule job control module, scheduling perform control module.
Operation with identical scheduling rule is abstracted into a complete grouping by the grouping management module.
The schedule job control module relies on the function of grouping, provides a variety of executive modes.
The scheduling performs control module according to the executing rule being configured in schedule job execution module, creates an operation
Pond is performed, scheduling performs control module module and performed from the job execution pond according to scheduling time and dependence timing.
In specific embodiments of the present invention, the grouping management module include configuration packet basic information unit (BIU), to
Operation unit, configuration job dependence relation unit are added in grouping, logging in system by user is matched in configuration packet basic information unit (BIU)
Put grouping essential information;Essential information is persisted to the grouping that user's option and installment is good in database and then adds schedule job,
Operation in grouping will have identical clocking discipline;After operation has been configured, the dependence being separately configured between operation is closed
System, dependence are persisted in database and are preserved.
In specific embodiments of the present invention, the schedule job control module includes following several job execution sides
Formula:It is grouped in all job executions, specified in packets job execution, grouping operation and is performed according to some operation for starting point, it is above-mentioned
Executive mode will directly generate executable operation, be pushed in operating pool.
In specific embodiments of the present invention, the scheduling performs control module from operating pool according between operation
Dependence obtains executable operation;Schedule job control module can skip clocking discipline and create executable operation, scheduling
The grouping creation time that performing control module will be subordinate to according to the operation in operating pool sequentially performs executable operation successively.
The positive effect of the present invention is:Packet scheduling system provided by the invention optimizes existing scheduling and performs
Strategy, by building the packing of grouping by the operation progress of multiple identical scheduling rules in logic, so as to simplify operation
Configuration, facilitates the unified management of operation, at the same time aspect maintenance work in later stage etc..
Description of the drawings
Fig. 1 is the overall schematic of the present invention.
Specific embodiment
Present pre-ferred embodiments are provided below in conjunction with the accompanying drawings, with the technical solution that the present invention will be described in detail.
Mainstream scheduling tool is broadly divided into two kinds at this stage, and one kind is CS frameworks, provides visualization interface, is dragged by dilatory
Mode configuration schedules operation, another is B/S framework, and based on Web configuration schedules operations, the two utilizes timer timing
Operation is performed, schedule job is then performed by certain dependence.Each operation suffers from oneself independent scheduling rule,
No matter whether the rule repeats.So when multiple and different operations are configured, and these job schedulings rule it is identical when, Jiu Huicun
Regular allocation problem is being repeated several times, is greatly reducing the efficiency of configuration.Simultaneously when job patch reports an error in scheduling, weight is needed
Compounding puts the rule of multiple schedule jobs to realize that operation is run again, this also increases the workload of later maintenance.
The present invention overcomes existing dispatching techniques there are shortcoming, packet scheduling system provided by the invention, including:Grouping
Management module, schedule job control module, scheduling perform control module;Grouping management module is by the work with identical scheduling rule
Industry is abstracted into a complete grouping;Schedule job control module relies on the function of grouping, provides a variety of executive modes;Scheduling
Control module is performed according to the executing rule being configured in schedule job execution module, creates a job execution pond, scheduling performs
Control module module is performed from the job execution pond according to scheduling time and dependence timing.
Fig. 1 is the overall schematic of the present invention, as shown in Figure 1, grouping management module is mainly believed substantially including configuration packet
Interest statement member adds operation unit, configuration job dependence relation unit into grouping, wherein logging in system by user configuration packet first
Essential information, such as group names, grouping clocking discipline, grouping number of concurrent etc., this partial information will be persisted to database
In;Then schedule job is added in the good grouping of user's option and installment, the operation in grouping will have identical clocking discipline, this
In do not need to repeat be configured;After operation has been configured, the dependence between operation can be separately configured, this partial dependency closes
System, which can be also persisted in database, to be preserved, and is simplified later stage repetition and is configured.So far grouping module basic function is completed, master
If for managing the dependence between operation and operation.Simultaneity factor will create one according to the timing information of grouping
Scheduler, but meet grouping and perform time thing, scheduler will generate multiple executable operations and be pushed in operating pool, for below
Module used.
Schedule job control module mainly provides several job execution modes, such as is grouped all job executions, grouping
Middle assignment performs, is performed in grouping operation according to some operation for starting point.Above-mentioned configuration will directly generate executable work
Industry is pushed in operating pool, is used for module below.Such mode is not belonging to clocking discipline scope, mainly makes up
When job execution when something goes wrong, simplify scheduling flow.Such as when occurring Partial Jobs failure in schedule job, Ke Yitong
The operation for crossing functional configuration execution re-executes, can be by performing grouping simultaneously when re-executing Historical Jobs
In all operations carry out meet demand.The problem of operation fails, needs to re-execute all operations is avoided, brings not keep away and want
The wasting of resources and maintenance cost.
It is mainly to obtain executable work according to the dependence between operation from operating pool that scheduling, which performs control module,
Industry.Executable operation is created since schedule job control module can skip clocking discipline, then scheduling performs control module
The grouping creation time that will be subordinate to according to the operation in operating pool sequentially performs executable operation successively, avoids identical point
Group job is performed simultaneously the situation for leading to job execution confusion.The module additionally provides a variety of implementation strategies simultaneously, including automatic
Mistake reruns function, i.e., be when mistake occurs in operation, module can active rerun the operation, reduce operation to the greatest extent
There is mistake so as to influence operation below.
The basic principles, main features and the advantages of the invention have been shown and described above.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention, the claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (4)
1. a kind of packet scheduling system, it is characterised in that:The packet scheduling system includes:Grouping management module, schedule job
Control module, scheduling perform control module;
Operation with identical scheduling rule is abstracted into a complete grouping by the grouping management module;
The schedule job control module relies on the function of grouping, provides a variety of executive modes;
The scheduling performs control module according to the executing rule being configured in schedule job execution module, creates a job execution
Pond, scheduling perform control module module and are performed from the job execution pond according to scheduling time and dependence timing.
2. packet scheduling system according to claim 1, it is characterised in that:The grouping management module includes configuration packet
Basic information unit (BIU) adds operation unit into grouping, job dependence relation unit is configured, in configuration packet basic information unit (BIU)
Logging in system by user configuration packet essential information;Then essential information is persisted to user's option and installment is good in database grouping
Schedule job is added, the operation in grouping there will be identical clocking discipline;After operation has been configured, operation is separately configured
Between dependence, dependence is persisted in database and preserved.
3. packet scheduling system according to claim 1, it is characterised in that:The schedule job control module includes following
Several job execution modes:It is grouped in all job executions, specified in packets job execution, grouping operation and is according to some operation
Starting point performs, and above-mentioned executive mode will directly generate executable operation, be pushed in operating pool.
4. packet scheduling system according to claim 1, it is characterised in that:The scheduling performs control module from operating pool
In executable operation is obtained according to the dependence between operation;Schedule job control module can skip clocking discipline establishment can
The operation of execution, the grouping creation time sequence that scheduling execution control module will be subordinate to according to the operation in operating pool is come successively
Perform executable operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810107836.XA CN108228329A (en) | 2018-02-02 | 2018-02-02 | A kind of packet scheduling system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810107836.XA CN108228329A (en) | 2018-02-02 | 2018-02-02 | A kind of packet scheduling system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108228329A true CN108228329A (en) | 2018-06-29 |
Family
ID=62669443
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810107836.XA Pending CN108228329A (en) | 2018-02-02 | 2018-02-02 | A kind of packet scheduling system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108228329A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11847194B2 (en) | 2020-06-29 | 2023-12-19 | Baidu Online Network Technology (Beijing) Co., Ltd. | Scheduling method and apparatus, device and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101038559A (en) * | 2006-09-11 | 2007-09-19 | 中国工商银行股份有限公司 | Batch task scheduling engine and dispatching method |
CN101241567A (en) * | 2008-03-04 | 2008-08-13 | 浪潮集团山东通用软件有限公司 | Background automatic business accomplishing method based on Windows service technology |
CN101271296A (en) * | 2007-03-20 | 2008-09-24 | 京瓷美达株式会社 | Job scheduler, job scheduling method |
US20110145829A1 (en) * | 2009-12-11 | 2011-06-16 | Ingo Molnar | Performance counter inheritance |
CN101502149B (en) * | 2006-11-14 | 2012-05-30 | 中兴通讯股份有限公司 | A network managing system for completing customized operation automatically and a method thereof |
CN103838621A (en) * | 2012-11-27 | 2014-06-04 | 中国电信股份有限公司 | Method and system for scheduling routine work and scheduling nodes |
US20150121387A1 (en) * | 2013-10-30 | 2015-04-30 | Mediatek Inc. | Task scheduling method for dispatching tasks based on computing power of different processor cores in heterogeneous multi-core system and related non-transitory computer readable medium |
-
2018
- 2018-02-02 CN CN201810107836.XA patent/CN108228329A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101038559A (en) * | 2006-09-11 | 2007-09-19 | 中国工商银行股份有限公司 | Batch task scheduling engine and dispatching method |
CN101502149B (en) * | 2006-11-14 | 2012-05-30 | 中兴通讯股份有限公司 | A network managing system for completing customized operation automatically and a method thereof |
CN101271296A (en) * | 2007-03-20 | 2008-09-24 | 京瓷美达株式会社 | Job scheduler, job scheduling method |
CN101241567A (en) * | 2008-03-04 | 2008-08-13 | 浪潮集团山东通用软件有限公司 | Background automatic business accomplishing method based on Windows service technology |
US20110145829A1 (en) * | 2009-12-11 | 2011-06-16 | Ingo Molnar | Performance counter inheritance |
CN103838621A (en) * | 2012-11-27 | 2014-06-04 | 中国电信股份有限公司 | Method and system for scheduling routine work and scheduling nodes |
US20150121387A1 (en) * | 2013-10-30 | 2015-04-30 | Mediatek Inc. | Task scheduling method for dispatching tasks based on computing power of different processor cores in heterogeneous multi-core system and related non-transitory computer readable medium |
Non-Patent Citations (2)
Title |
---|
JINGHUI ZHANG 等: "An Adaptive QoS Group Guided Grid Scheduling Algorithm with Task Replicas", 《2007 11TH INTERNATIONAL CONFERENCE ON COMPUTER SUPPORTED COOPERATIVE WORK IN DESIGN》 * |
巴巍 等: "一种新的组优先级动态实时调度算法", 《计算机科学》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11847194B2 (en) | 2020-06-29 | 2023-12-19 | Baidu Online Network Technology (Beijing) Co., Ltd. | Scheduling method and apparatus, device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8473951B2 (en) | Method and system for traversing in reverse chronological order along a critical path of a plurality of jobs, and reducing time gaps between jobs until an estimated end time of the last job is less than or equal to a target end time | |
CN102663552B (en) | Dynamic workflow engine supporting online self-evolution | |
Fohler | Flexibility in statically scheduled hard real-time systems | |
CN102591759A (en) | Clock precision parallel simulation system for on-chip multi-core processor | |
CN102279766A (en) | Method and system for concurrently simulating processors and scheduler | |
CN103197960A (en) | Scheduling method and scheduling system for batch job system | |
CN107316124B (en) | Extensive affairs type job scheduling and processing general-purpose system under big data environment | |
Fernández-Cerero et al. | Limiting global warming by improving data-centre software | |
CN108415740A (en) | A kind of workflow schedule method applied to data analysis task | |
Mikida et al. | Towards pdes in a message-driven paradigm: A preliminary case study using charm++ | |
CN104461471A (en) | Unified instruction scheduling and register allocating method on clustering VLIW processor | |
CN102567856A (en) | Method for realizing time management in work flow system | |
Wang et al. | Colored petri net model with automatic parallelization on real-time multicore architectures | |
Lowinski et al. | Splitting tasks for migrating real-time automotive applications to multi-core ecus | |
CN108228329A (en) | A kind of packet scheduling system | |
Sorkhpour et al. | Meta-scheduling techniques for energy-efficient robust and adaptive time-triggered systems | |
Zheng et al. | Optimizing MapReduce framework through joint scheduling of overlapping phases | |
Bettati et al. | Algorithms for end-to-end scheduling to meet deadlines | |
Kraemer et al. | Reducing the number of response time service level objective violations by a cloud‐HPC convergence scheduler | |
CN104834571B (en) | A kind of data prefetching method applied to cloud workflow schedule | |
Tumanov et al. | Tetrisched: Space-time scheduling for heterogeneous datacenters | |
Rumi et al. | Optimization techniques within the hadoop eco-system: A survey | |
Wang et al. | Slo-driven task scheduling in mapreduce environments | |
Chang et al. | Schedulability analysis for a real-time multiprocessor system based on service contracts and resource partitioning | |
Bunde et al. | PReMAS: Simulator for resource management |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180629 |
|
RJ01 | Rejection of invention patent application after publication |