CN109639595A - A kind of CDN dynamic priority scheduling algorithm based on time delay - Google Patents
A kind of CDN dynamic priority scheduling algorithm based on time delay Download PDFInfo
- Publication number
- CN109639595A CN109639595A CN201811123772.9A CN201811123772A CN109639595A CN 109639595 A CN109639595 A CN 109639595A CN 201811123772 A CN201811123772 A CN 201811123772A CN 109639595 A CN109639595 A CN 109639595A
- Authority
- CN
- China
- Prior art keywords
- priority
- timeslice
- queue
- grouping
- time
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/622—Queue service order
- H04L47/6235—Variable service order
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
- H04L47/6275—Queue scheduling characterised by scheduling criteria for service slots or service orders based on priority
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/6285—Provisions for avoiding starvation of low priority queues
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of CDN dynamic priority scheduling algorithm based on time delay, grouping in each queue has the limitation of service time, a higher priority is distributed it when grouping is close to its limiting time, it is i.e. higher closer to limiting time priority, simultaneously, in order to avoid the process limited of system becomes larger, grouping is set to have disconnection phenomenon, timeslice is combined with priority, with the increase of timeslice, not the stopping to improve of low priority may also seize service time of high priority packet, the present invention is by carrying out control to process switching opportunity and determining for timeslice size, most of interactive sessions are enable to complete in a timeslice, it is hereby achieved that the response time of very little;Simultaneously this invention also solves when new being grouped into comes, how to be avoided the problem that and existing packet collisions when being written and read to new grouping.
Description
Technical field
The present invention relates to Internet technical fields, and in particular to a kind of CDN dynamic priority scheduling algorithm based on time delay.
Background technique
In CDN business platform, when CDN node is when handling network flow, it is frequently encountered multiple tasks needs
The case where occupying resource simultaneously, and total Internet resources are limited.Needing to determine priority processing in this case, which is appointed
Business, which task can be with delay disposal.In the present circumstance, scheduling is solve the problems, such as multiple service competition shared resources effective
Means.
Existing dispatching algorithm can be divided into following a few classes according to scheduling rule: based on differentiated service priority service, be based on
Dynamic bidirectional priority, ratio distinguish algorithm and based on discrete particle clusters etc..Queue scheduling algorithm now is all adjustment queue
The grouping of middle storage calculates trigger order, and as the control that benchmark is scheduled.
However, existing dispatching technique, can solve the problems, such as that Internet resources distribute, but there is also one to a certain extent
Fixed defect, this is mainly manifested in following aspect, after each task trigger order in the queue calculates, according to certain suitable
Sequence storage, and these sequences will not change after once arranging, and in practical business, the preferential of some tasks is substantially
Dynamic change, the priority level of some tasks task processing in rigid start is higher, it comes by front position, and with the time
Variation, the priority of these tasks may reduce, need to be come rearward position.And existing technology lacks automatic adjust
Energy saving power cannot make corresponding adjustment according to the variation of the priority level of task, thus technically lack corresponding flexible
Property and accuracy.
Summary of the invention
In view of the deficiencies of the prior art, the invention discloses a kind of CDN dynamic priority scheduling algorithm based on time delay, often
Grouping in a queue has the limitation of service time, it is distributed when grouping is close to its limiting time one it is higher preferential
Grade is that is, higher closer to limiting time priority.
The present invention is achieved by the following technical programs:
A kind of CDN dynamic priority scheduling algorithm based on time delay, the grouping in each queue have the limit of service time
System distributes it higher priority when grouping is close to its limiting time, i.e., higher closer to limiting time priority,
Meanwhile to avoid the process limited of system from becoming larger, grouping is made to have a disconnection phenomenon, this algorithm by timeslice with it is excellent
First grade combines, and determines timeslice using robin scheduling algorithm, specifically: all ready processes is i.e. first first by FCFS
The strategy of service lines up a ready queue.System is settable just to generate primary interruption at regular intervals, deactivates process tune
Degree program is scheduled, and gives head of the queue process resource allocation, and it is enabled to execute a timeslice;After it runs, and handle
Multi-frame system executed a timeslice by also to be allowed it, guaranteed in ready queue with this to head of the queue process new in ready queue
All processes can obtain the processor time of a timeslice within the determining period.
In the above-mentioned technical solutions, when the switching of process should be being carried out, is being divided into two kinds:
If 1, a timeslice is not yet finished, the process being currently running just has been completed, and just activates scheduler program immediately, will
It is deleted from ready queue, the process operation of head of the queue in scheduling ready queue, and starts a new timeslice;
2, when a timeslice is finished, timer interrupt handling routine is activated.If process not yet runs and finishes, adjust
Degree program will be sent to it at the end of ready queue.
In the above-mentioned technical solutions, a timeslice size preferably is slightly greater than required for primary typical interaction
Time, so that most of interactive sessions is completed in a timeslice, it is hereby achieved that the response time of very little.
In the above-mentioned technical solutions, the formulation of priority specifically:
In IP datagram head, a buffer area is divided into several queues.Because the subqueue in buffer area is from excellent
The different data packet of first grade, so recording corresponding priority state, the low sub- team of script priority when grouping enters queue
Column improve the priority serviced in itself over time, i.e., the priority of queue be not it is constant always but with when
Prolong dynamic change;
In order to realize dynamic priority algorithm, counter is respectively set to record the queue for entering buffer area, with its institute
Belong to initial value of the priority of data packet as counter;Every to pass through a timeslice, Counter Value increases by 1;Choose counter
Highest subqueue team head grouping is forwarded.
In the above-mentioned technical solutions, it is solution when new being grouped into comes, how is avoided when being written and read to new grouping
The problem of with existing packet collisions, it is preferred that this algorithm uses dual-ported memory, mutual comprising two groups of parallel independent operations
Independent Read-write Catrol route is only stored into blank area when being newly grouped into and, and has grouping address in the buffer not
It is identical, so being written and read on the two ports, will not centainly clash.In one timeslice, what is be newly grouped is arrived
Come can be with parallel processing with the forwarding for having grouping.
In the above-mentioned technical solutions, in order to solve the problems, such as that Counter Value is too sensitive to timeslice, the present invention is used and is set
The method that threshold value distinguishes is set, number namely subqueue length are grouped present in each subqueue, the head grouping of subqueue team
The priority height that threshold size is grouped according to subqueue and the actual flow of each priority packet are configured, at one
Between queue counter after piece only more than threshold value just increase by 1.
The invention has the benefit that
The present invention relates to a kind of CDN dynamic priority scheduling algorithm based on time delay, grouping in each queue
There is the limitation of service time, distribute it higher priority when grouping is close to its limiting time, i.e., closer to limiting
Time priority is higher, meanwhile, in order to avoid the process limited of system becomes larger, so that grouping is had disconnection phenomenon, by timeslice
It is combined with priority, with the increase of timeslice, not the stopping to improve of low priority may also seize the clothes of high priority packet
It is engaged in the time, the present invention makes most of interactive sessions by carrying out control to process switching opportunity and determining for timeslice size
It can be completed in a timeslice, it is hereby achieved that the response time of very little;This invention also solves work as new grouping simultaneously
When arrival, how to be avoided the problem that and existing packet collisions when being written and read to new grouping.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of process step signal of CDN dynamic priority scheduling algorithm based on time delay of the embodiment of the present invention
Figure;
Fig. 2 is a kind of CDN dynamic priority scheduling algorithm based on time delay of the embodiment of the present invention using setting threshold value
Distinguish each subqueue count device situation of change schematic diagram in the buffer area of method.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
As shown in Figs. 1-2, to improve CDN network service quality, the invention proposes a kind of, and the CDN dynamic based on time delay is excellent
First grade dispatching algorithm.Grouping in the algorithm in each queue has the limitation of service time, when grouping is close to its limiting time
When distribute it higher priority, i.e., it is higher closer to limiting time priority.But the algorithm is easy to cause system
Process limited becomes larger, and grouping is made to have disconnection phenomenon.The present invention is on the basis of the dispatching algorithm based on time delay by timeslice
It is combined with priority, proposes a kind of new dynamic priority scheduling algorithm based on time delay.
It is the determination of timeslice first, the invention patent uses robin scheduling algorithm to determine timeslice.
In robin scheduling algorithm, all ready processes are lined up one by the strategy of FCFS (prerequisite variable) by system
Ready queue.System settable at regular intervals (such as 30ms) just generates primary interruption, deactivates process dispatcher and carries out
Scheduling, gives head of the queue process resource allocation, and it is enabled to execute a timeslice.After it runs, and multi-frame system
To head of the queue process new in ready queue, it is also allowed to execute a timeslice.This way it is secured that all in ready queue
Process can obtain the processor time of a timeslice within the determining period.In robin scheduling algorithm, there are two close
Key factor: the determination on process switching opportunity and timeslice size.
Robin scheduling algorithm process switching opportunity:
In robin scheduling algorithm, when the switching of process should be being carried out, can be divided into two kinds of situations: if 1, timeslice
Not yet it is finished, the process being currently running just has been completed, and just activates scheduler program immediately, it is deleted from ready queue,
The process operation of head of the queue in ready queue is dispatched, and starts a new timeslice.2, when a timeslice is finished, timer
Interrupt handling routine is activated.If process not yet runs and finishes, scheduler program will be sent to it at the end of ready queue.
The determination of robin scheduling algorithm timeslice size:
In robin scheduling algorithm, the size of timeslice has very big influence to system performance.If select very little when
Between inner sheet, short operation is beneficial to, because it can be completed in the timeslice.But timeslice is small, it is meant that can be frequently executed
The switching of process scheduling and process context, this undoubtedly will increase the expense of system.Conversely, if event piece selection it is too long, and
To complete each process can in a timeslice, it is unable to satisfy the demand of short operation and oolhiu interactive user.One more
Desirable timeslice size is slightly greater than the time required for primary typical interaction, enables most of interactive sessions at one
It is completed in timeslice, it is hereby achieved that the response time of very little
Next the present invention discusses the formulation of priority.In IP datagram head, usually use TOS (service type) as
The label of service priority, TOS are 3 bit, range 0-7, indicate 8 kinds of priority level triggering types, set minimum triggering time
Sequence is 0, is successively increased, and 7 be highest trigger order, and setting when counter maximum increases to 8 indicates subqueue as sky.Cause
One buffer area is divided into 8 queues by this.Because of subqueue in the buffer area data packet different from priority,
Grouping records corresponding priority state when entering queue.The low subqueue of script priority improves itself over time
The priority serviced, the i.e. priority of queue are not constant always but with time delay dynamic changes.
In order to realize dynamic priority algorithm, counter is respectively set to record the queue for entering buffer area, with its institute
Belong to initial value of the priority of data packet as counter;Every to pass through a timeslice, Counter Value increases by 1 (if Counter Value
Greater than 7, illustrate that queue for sky, does not increase);The highest subqueue team head grouping of counter is chosen to be forwarded.Assumed priority
Position I (I=0,1 ... 7) in.Such as have 5 queues etc. to be serviced in a buffer area, the Counter Value of individual queue is followed successively by 6,
4,3,2,0, by a timeslice, forwarding Counter Value is 6 grouping, and it is 5,4,3 that the Counter Value of residual queue, which increases,
1. if when that is, queue 5 is identical as the priority of queue 4, how to be forwarded, this is just to draw there is new grouping to reach in queue at this time 5
The problem of implementation of heap is entered.Next the present invention discusses the relationship of priority and heap.
Since the dispatching algorithm that the present invention uses always is selected from CDN buffer area and removes the team of priority maximum value
Column, i.e., will packets forwarding.The present invention selects " heap " as data structure priority query storage and realized.It is inserted into node and deletes
Except the average time cost of heap top element and worst time cost are all 0.
Firstly, each node in heap saves the priority value of dynamic adjustment, the i.e. Counter Value of individual queue.In addition, by
In after selecting the maximum node of priority, needing the buffer area where it to forward the corresponding queue of this priority, so
Each heap node also needs to save the subqueue number that priority packet is entered.A label is added for each heap in we herein
Position, becomes the heap of tape label.It is convenient for programming, flag bit is put behind.Each node consists of two parts in heap: front
Dividing content is the value of priority;Rear portion is divided into marker bit, records subqueue number corresponding to the priority.
The present invention will also handle a technical problem, i.e., when new being grouped into comes, such as when being written and read to new grouping
What is avoided the problem that and existing packet collisions.Buffer area provided by the invention uses dual-ported memory, can be effectively avoided
This problem.The method of dual-ported memory is the mutually independent reading in the same memory comprising two groups of parallel independent operations
Write line road is only stored into blank area when being newly grouped into and, and the address of grouping in the buffer is not identical with having, so
It is written and read on the two ports, will not centainly clash.In one timeslice, the arrival that is newly grouped and have point
The forwarding of group can be with parallel processing, and the process of the dynamic priority scheduling algorithm based on time delay is as shown in Figure 1.
The present invention is also optimized the dynamic priority scheduling algorithm based on time delay.It is calculated in scheduling priority-based
In method, a large amount of high-priority queues successively arrive be likely to result in other Low Priority Queuings fall into a long wait and cannot be normal
Service;And in the dynamic priority algorithm based on time delay, with the increase of timeslice, low priority do not stop improve may also
Seize the service time of high priority packet.To ensure that different priorities grouping can obtain normal service, it is fair to improve algorithm
Property, the dynamic priority algorithm based on time delay is optimized in the present invention, the method is as follows:
In order to solve the problems, such as that Counter Value is too sensitive to timeslice, the present invention, which uses, is arranged the side that threshold value distinguishes
Method.The threshold size of subqueue team head grouping is according to the subqueue priority height being grouped and the practical stream of each priority packet
Amount (number namely subqueue length are grouped present in each subqueue) is configured.By this method, buffer area is combined
The actual use situation of middle queue, the queue counter after a timeslice only more than threshold value just increase by 1.Set priority
0,1,2 be low priority, and threshold value is set as 30;Priority 3,4,5 are middle priority, threshold value 20;Priority 6,7 is Gao Youxian
Grade, threshold value 10.Threshold value, subqueue priority and corresponding the packet count such as threshold value of following table and packet count institute in respective queue
Show:
Subqueue | Priority | Threshold value | Packet count in subqueue |
0 | 0 | 30 | 32 |
1 | 1 | 30 | 28 |
2 | 2 | 30 | 29 |
3 | 3 | 20 | 21 |
4 | 4 | 20 | 19 |
5 | 5 | 20 | 17 |
6 | 6 | 10 | 11 |
7 | 7 | 10 | 9 |
Grouping enters in corresponding subqueue I (I=0,1 ... .7) according to priority, and wherein packet count is more than preset threshold
Subqueue there was only 1,3,6, then pass through a timeslice, the queue heads grouping in subqueue 7 due to Counter Value it is maximum and by
Forwarding, subqueue 1, subqueue 3, team's head group technology device value of subqueue 6 increase by 1, grouping present in remaining subqueue
Number is not above threshold value, thus the Counter Value of its team head grouping remains unchanged.Each subqueue count device in this process buffers area
Situation of change is as shown in Fig. 2, when through variation after a period of time all to change into highest excellent for the priority of each subqueue team head grouping
When first grade (Counter Value 7), from high serial number queue, i.e. high priority packet, start to dispatch, to ensure high priority packet
Service quality.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features;And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (6)
1. a kind of CDN dynamic priority scheduling algorithm based on time delay, it is characterised in that: the grouping in each queue has service
The limitation of time distributes it higher priority when grouping is close to its limiting time, i.e., excellent closer to limiting time
First grade is higher,
Meanwhile to avoid the process limited of system from becoming larger, grouping is made to have a disconnection phenomenon, this algorithm is by timeslice and priority
It combines, timeslice is determined using robin scheduling algorithm, specifically: all ready processes are pressed into FCFS, that is, prerequisite variable
Strategy line up a ready queue.System is settable just to generate primary interruption at regular intervals, deactivates process scheduling journey
Sequence is scheduled, and gives head of the queue process resource allocation, and it is enabled to execute a timeslice;After it runs, and processing
Machine distributes to head of the queue process new in ready queue, it is also allowed to execute a timeslice, is guaranteed with this all in ready queue
Process can obtain the processor time of a timeslice within the determining period.
2. a kind of CDN dynamic priority scheduling algorithm based on time delay according to claim 1, which is characterized in that Ying
The switching for when carrying out process, is divided into two kinds:
If 1) timeslice is not yet finished, the process being currently running just has been completed, and just activates scheduler program immediately, by it from
It is deleted in ready queue, the process operation of head of the queue in scheduling ready queue, and starts a new timeslice;
2) when a timeslice is finished, timer interrupt handling routine is activated.If process not yet runs and finishes, journey is dispatched
Sequence will be sent to it at the end of ready queue.
3. a kind of CDN dynamic priority scheduling algorithm based on time delay according to claim 2, which is characterized in that one
Timeslice size preferably is slightly greater than the time required for primary typical interaction, and most of interactive sessions is enable to exist
It is completed in one timeslice, it is hereby achieved that the response time of very little.
4. a kind of CDN dynamic priority scheduling algorithm based on time delay according to claim 1, which is characterized in that preferential
The formulation of grade specifically:
In IP datagram head, a buffer area is divided into several queues.Because the subqueue in buffer area comes from priority
Different data packets, so record corresponding priority state when grouping enters queue, the low subqueue of script priority with
The passage of time improve priority in itself that serviced, i.e. the priority of queue is not constant always but as time delay is dynamic
State variation;
In order to realize dynamic priority algorithm, counter is respectively set to record the queue for entering buffer area, with number belonging to it
Initial value according to the priority of packet as counter;Every to pass through a timeslice, Counter Value increases by 1;Choose counter highest
Subqueue team head grouping be forwarded.
5. a kind of CDN dynamic priority scheduling algorithm based on time delay according to claim 1, which is characterized in that for solution
Certainly when new being grouped into comes, how to be avoided the problem that and existing packet collisions when being written and read to new grouping, it is preferred that this
Algorithm uses dual-ported memory, and the mutually independent Read-write Catrol route comprising two groups of parallel independent operations is grouped into when newly
It is only stored into blank area when coming, the address of grouping in the buffer is not identical with having, so being read on the two ports
Write operation will not centainly clash.In one timeslice, the arrival being newly grouped can be located parallel with the forwarding for having grouping
Reason.
6. a kind of CDN dynamic priority scheduling algorithm based on time delay according to claim 1, which is characterized in that in order to
Solve the problems, such as that Counter Value is too sensitive to timeslice, the method that the present invention is distinguished using setting threshold value, each subqueue
Present in be grouped number namely subqueue length, the threshold size of subqueue team head grouping is grouped preferential according to subqueue
Grade height and the actual flow of each priority packet are configured, only more than the queue count of threshold value after a timeslice
Device just increases by 1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811123772.9A CN109639595A (en) | 2018-09-26 | 2018-09-26 | A kind of CDN dynamic priority scheduling algorithm based on time delay |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811123772.9A CN109639595A (en) | 2018-09-26 | 2018-09-26 | A kind of CDN dynamic priority scheduling algorithm based on time delay |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109639595A true CN109639595A (en) | 2019-04-16 |
Family
ID=66066307
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811123772.9A Pending CN109639595A (en) | 2018-09-26 | 2018-09-26 | A kind of CDN dynamic priority scheduling algorithm based on time delay |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109639595A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111737385A (en) * | 2020-06-02 | 2020-10-02 | 通号城市轨道交通技术有限公司 | Electronic map data error detection method and device |
CN112243253A (en) * | 2019-10-24 | 2021-01-19 | 北京大学 | Communication equipment |
CN112585914A (en) * | 2020-11-27 | 2021-03-30 | 新华三技术有限公司 | Message forwarding method and device and electronic equipment |
CN113010273A (en) * | 2021-03-23 | 2021-06-22 | 河北冀联人力资源服务集团有限公司 | Human resource data distributed task processing method and system |
WO2021139560A1 (en) * | 2020-01-07 | 2021-07-15 | 苏州大学 | Task scheduling processing system and method for embedded real-time operating system |
CN117032993A (en) * | 2023-10-08 | 2023-11-10 | 麒麟软件有限公司 | Task scheduling method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090161684A1 (en) * | 2007-12-21 | 2009-06-25 | Juniper Networks, Inc. | System and Method for Dynamically Allocating Buffers Based on Priority Levels |
CN102014052A (en) * | 2010-11-05 | 2011-04-13 | 中国科学院声学研究所 | Virtual dynamic priority packet dispatching method |
-
2018
- 2018-09-26 CN CN201811123772.9A patent/CN109639595A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090161684A1 (en) * | 2007-12-21 | 2009-06-25 | Juniper Networks, Inc. | System and Method for Dynamically Allocating Buffers Based on Priority Levels |
CN102014052A (en) * | 2010-11-05 | 2011-04-13 | 中国科学院声学研究所 | Virtual dynamic priority packet dispatching method |
Non-Patent Citations (2)
Title |
---|
张登银等: "基于时延的动态优先级调度算法", 《计算机技术与发展》 * |
汤小丹等编著: "《计算机操作系统(第四版)高等学校计算机类"十二五"规划教材 部级优秀教材》", 31 May 2014 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112243253A (en) * | 2019-10-24 | 2021-01-19 | 北京大学 | Communication equipment |
CN112243253B (en) * | 2019-10-24 | 2022-07-08 | 北京大学 | Communication equipment |
WO2021139560A1 (en) * | 2020-01-07 | 2021-07-15 | 苏州大学 | Task scheduling processing system and method for embedded real-time operating system |
CN111737385A (en) * | 2020-06-02 | 2020-10-02 | 通号城市轨道交通技术有限公司 | Electronic map data error detection method and device |
CN112585914A (en) * | 2020-11-27 | 2021-03-30 | 新华三技术有限公司 | Message forwarding method and device and electronic equipment |
WO2022109986A1 (en) * | 2020-11-27 | 2022-06-02 | 新华三技术有限公司 | Message forwarding method and apparatus, and electronic device |
CN112585914B (en) * | 2020-11-27 | 2024-05-28 | 新华三技术有限公司 | Message forwarding method and device and electronic equipment |
CN113010273A (en) * | 2021-03-23 | 2021-06-22 | 河北冀联人力资源服务集团有限公司 | Human resource data distributed task processing method and system |
CN113010273B (en) * | 2021-03-23 | 2022-07-19 | 河北冀联人力资源服务集团有限公司 | Human resource data distributed task processing method and system |
CN117032993A (en) * | 2023-10-08 | 2023-11-10 | 麒麟软件有限公司 | Task scheduling method |
CN117032993B (en) * | 2023-10-08 | 2024-04-05 | 麒麟软件有限公司 | Task scheduling method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109639595A (en) | A kind of CDN dynamic priority scheduling algorithm based on time delay | |
US6085215A (en) | Scheduling mechanism using predetermined limited execution time processing threads in a communication network | |
US6477562B2 (en) | Prioritized instruction scheduling for multi-streaming processors | |
CN102317917B (en) | Hot field virtual machine cpu dispatching method and virtual machine system (vms) | |
US7660251B2 (en) | Method and apparatus for hierarchial scheduling of virtual paths with underutilized bandwidth | |
US8539501B2 (en) | Managing access to a shared resource in a data processing system | |
CN107579926A (en) | The QoS methods to set up of Ceph cloud storage systems based on token bucket algorithm | |
US20110061058A1 (en) | Task scheduling method and multi-core system | |
JP5050601B2 (en) | Resource allocation method for job, resource allocation method for job, and program | |
CN112506634B (en) | Fairness operation scheduling method based on reservation mechanism | |
CN101923487A (en) | Comprehensive embedded type real-time period task scheduling method | |
CN104243349B (en) | Method for dispatching message and device | |
US8640133B2 (en) | Equal duration and equal fetch operations sub-context switch interval based fetch operation scheduling utilizing fetch error rate based logic for switching between plurality of sorting algorithms | |
US20130061233A1 (en) | Efficient method for the scheduling of work loads in a multi-core computing environment | |
EP2562644A1 (en) | Systems and methods for bounding processing times on multiple processing units | |
CN101145112A (en) | Real-time system task scheduling method | |
US8295305B2 (en) | Dual scheduling of work from multiple sources to multiple sinks using source and sink attributes to achieve fairness and processing efficiency | |
CN112306904B (en) | Cache data disk refreshing method and device | |
CN107454014A (en) | A kind of method and device of Priority Queuing | |
GB2322718A (en) | Dynamic classification and dispatch of instructions out of order | |
CN103685053A (en) | Network processor load balancing and scheduling method based on residual task processing time compensation | |
WO2013167973A2 (en) | Hybrid dataflow processor | |
CN104360902A (en) | Sliding window-based multi-priority metadata task scheduling method | |
US8879578B2 (en) | Reducing store and forward delay in distributed systems | |
CN114500401A (en) | Resource scheduling method and system for dealing with burst traffic |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190416 |