CN110138686A - A kind of Ethernet design scheme based on dynamic secondary feedback scheduling - Google Patents
A kind of Ethernet design scheme based on dynamic secondary feedback scheduling Download PDFInfo
- Publication number
- CN110138686A CN110138686A CN201910422802.4A CN201910422802A CN110138686A CN 110138686 A CN110138686 A CN 110138686A CN 201910422802 A CN201910422802 A CN 201910422802A CN 110138686 A CN110138686 A CN 110138686A
- Authority
- CN
- China
- Prior art keywords
- data
- queue
- time
- real
- real 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.)
- Granted
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
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
-
- 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
- H04L47/6225—Fixed service order, e.g. Round Robin
-
- 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
- H04L47/6245—Modifications to standard FIFO or LIFO
-
- 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
Abstract
The present invention relates to a kind of design scheme of industry ethernet networked-induced delay, especially a kind of Ethernet Real-time Design scheme based on dynamic secondary feedback scheduling.It mainly include Dynamic Scheduling Strategy and self-adaptive secondary feedback scheduling strategy two parts, ethernet network system response mode uses dynamic dispatching, it not only contains the real-time status of the existing real time data of storehouse, further comprise the state after new real time data reaches, ensure that ethernet network system can make rapid reaction to the environment of variation, enable ethernet network to monitor the dynamic requirements of real time data in real time, increases its overall real-time；Self-adaptive secondary feedback scheduling strategy combines dual cycle queue priority policy and storehouse hierarchical policy, it is ensured that the priority of real time data can be met.The program can be used for solving the problems, such as ethernet network real time data ethernet network system real time due to caused by collision, decaying, loss etc., ensure the transmission of real time data.
Description
Technical field
The present invention relates to a kind of design schemes of industry ethernet networked-induced delay, especially a kind of anti-based on dynamic secondary
Present the Ethernet Real-time Design scheme of scheduling.This scheme can be used for solving ethernet network real time data due to collision, decaying,
Ethernet network system real time problem caused by loss etc. provides energy-saving square for industry ethernet network application and its research
Case.
Background technique
Ethernet (Ethernet) is the most general communication protocol standard that current existing local area network uses, transmission rate
There are tetra- kinds of 10Mbit/s, 100Mbit/s, 1Gbit/s, 10Gbit/s, is able to satisfy quick, real-time, the reliable control of Ethernet and wants
It asks, has obtained extensive fortune in the higher industrial control field of the requirement of real-time such as EMU, subway train and city rail vehicle
With.However Ethernet uses traditional CSMA/CD (csma/collision detection) media access control mode, this
Determine that common Ethernet has uncertainty in real-time.
In order to improve the real-time of ethernet network data transmission, guarantee its reliability, and then reach and improve Ethernet net
The real-time of network system needs to analyze data transmission policies and theory, while influence factor and property to real-time
Energy index is analyzed.Existing key technology such as redundancy, reliability engineering, bus-powered, Industrial Ethernet real-time
The research of security mechanism, functional safety etc. is all based on this research background.
Ethernet network real-time has particularly significant influence to the optimization of network performance and system design.At present to ether
The research of net real-time common are following several method: 1) improving its real-time based on topological structure, such as change ether
Network topology structure increases number of nodes and interchanger number to improve the real-time of ethernet network；2) based on tune in real time
Degree strategy is to improve its real-time, such as Packet scheduling and mixing sleep pattern etc.；3) it is protected based on purely logical circuit
Demonstrate,prove its real-time etc..However there is respective limitations for existing method, such as can not support gigabit Ethernet, shake compared with
Greatly, cross-communication etc. is not supported, these all reduce the real-time of data transmission.Therefore, the reality of ethernet network how is improved
When property becomes Ethernet field problem to be solved.
Summary of the invention
The technical problem to be solved by the present invention is to provide one kind for real time problems existing for existing ethernet network
Ethernet design scheme based on dynamic secondary feedback scheduling is made of two parts: Dynamic Scheduling Strategy and adaptive two
Grade feedback scheduling strategy.Ethernet network system response mode uses dynamic dispatching, it not only contains the existing real-time number of storehouse
According to real-time status, the state after new real time data reaches is further comprised, so that it is guaranteed that ethernet network system can be to change
The environment of change makes rapid reaction, and ethernet network is enabled to monitor the dynamic requirements of real time data in real time, and it is total to increase it
Body real-time；Self-adaptive secondary feedback scheduling strategy combines dual cycle queue priority policy and storehouse hierarchical policy, it is ensured that
The priority of real time data can be met.
The real-time status that monitoring available data is focused on for traditional ethernet, has ignored the monitoring to data of newly arriving, from
And the data that cause to newly arrive conflict with legacy data generation, seize behavior so that should the data of real-time Transmission can not be preferential
Transmission can only wait and seize successful non real-time data transmission and finish.The invention proposes a kind of Ethernet Dynamic Scheduling Strategies
Mode, enable ethernet network system in the case that dispatch environment and real-time property there are when unpredictable disturbance to variation
Environment makes rapid reaction.Different generate that this method could receive and transmit data for ethernet network has more
There is the decision scheme of reliability and operability, and the existing situation of new data and legacy data can be monitored in real time, i.e., when new
When the states such as data arrival, data transmission, data transmission success, data transmission fails occur, system can pass through dynamic dispatching
It learns Various types of data standing state, to reduce the probability of data collision, guarantees the transmission of real time data, improve the real-time of system
Property.
Simultaneously in order to guarantee that the stabilization of ethernet network, the data in network must carry out fast and reliable transmission.Therefore
Guarantee the fast and reliable transmission of data the invention proposes a kind of self-adaptive secondary feedback scheduling strategy.All data types
Real-time period data, in real time aperiodic data and non-real-time data three types can be divided into, in prerequisite variable (First-
Come-First-Service) in queue, it is successively etc. to be forwarded that these three data types can all be considered as equal priority.However
If long data frame is being sent always, real time data cannot forward for a long time, will generate strong influence to Time Delay of Systems.
Single cycle queue in FCFS has been changed to by the present invention has the other dual cycle queue of different priorities, by high-priority queue and low
Both priority queries combine, and using dual cycle queue scheduling strategy, propose that different data classification accounts for the heap there are two storehouse
The self-adaptive secondary feedback scheduling strategy that stack hierarchical policy and dual cycle queue priority policy combine.Dual cycle queue is preferential
Grade strategy solves the problems, such as that non-real-time data seizes channel and cannot obtain real-time Transmission so as to cause real time data.Storehouse classification
Strategy takes different strategies to real time data and non-real-time data packet, solves data forwarding sequence in priority query
Problem.
In the present invention, we consider the underlying protocol of Ethernet, including physical layer and data link layer, wherein data
Link layer includes logic link control (LLC) sublayer and medium access control (MAC) sublayer again.We first proposed one kind
Dynamic Scheduling Strategy, when data terminal equipment (DTE) input data to physical layer, dynamic dispatching first detects coming for new data
Arrive, and start Real-time Monitoring Data dynamic, with ensure data mode and whether transmission success.Then, we have proposed another
Kind scheduling strategy and rule, the strategy and rule consider the mode of data stack classification and data priority.In order to realize
This strategy and rule, the present invention devise storehouse hierarchical policy and dual cycle queue priority policy, and wherein storehouse is classified plan
To it in real time with non real-time classification when slightly entering for realizing data, dual cycle queue priority policy then uses priority
It can also first be sent with the real time data for ensuring to reach earliest, all real time datas can be sent before non-real-time data.
Compared with the prior art, the present invention is based on Dynamic Scheduling Strategies, in self-adaptive secondary feedback scheduling strategy and rule
Basis under, propose a kind of real-time scheme of ethernet network bottom, that is, be based on Dynamic Scheduling Strategy, monitor in real time it is existing
The real-time status of data and new data, to ensure data Successful transmissions, if transmission failure, is passed again using retransmission mechanism
It is defeated, until success.Herein on basis, using storehouse hierarchical policy and dual cycle queue priority policy, in data from data
First data are classified after terminal device (DTE) is incoming, real-time data accounts for a storehouse, and non real-time nature data account for a heap
Stack, setting real-time data storehouse enters in circulation one, rather than real-time data enters in circulation two, this solves channel competition
When assignment problem, so that guarantee requirement of real-time, solves the problems, such as the channel right to use.
Detailed description of the invention
Fig. 1 is network bottom layer schematic diagram of the invention.
Fig. 2 is network stack hierarchical policy process schematic of the invention.
Fig. 3 is network dual cycle queue priority policy process schematic of the invention.
Fig. 4 is FCFS scheduling flow figure in dual cycle queue priority policy of the invention.
Fig. 5 is that simulation result of the invention compares figure.
Specific embodiment
The present invention is further detailed with reference to the accompanying drawing.Fig. 1 is the fabric of ethernet network,
It is the structure that the present invention is optimized, respectively physical layer and data link layer, wherein physical layer is object required for transmission data
Link establishment, maintenance, dismounting are managed, and characteristic that have machinery, is provided.Briefly, physics
Layer ensures that original data can be transmitted on various physical mediums.And data link layer is divided into logic link control (LLC)
Layer and medium access control (MAC) sublayer, Dynamic Scheduling Strategy bring into operation from MAC, and the state of Real-time Monitoring Data is to protect
Card data can be accurately successfully transmitted, and self-adaptive secondary feedback scheduling then in LLC to real time data and non-real-time data into
Row classification guarantees real time data with high priority transfer.
In Dynamic Scheduling Strategy, data real-time status can use data free time --- and the data redundancy time, this was empty
Between model indicate that and change in location of the data in this space has different meanings: (whether too single Dynamic Scheduling Strategy is
It is thin) introduce it is very little.
1) data start to transmit: data move down, and remaining time becomes smaller；2) data are without transmission: data are to moving to left
Dynamic, free time becomes smaller；3) data do not reach: data are motionless.It has been confirmed that so these data exist when some data do not reach
Reserved location on space diagram is activated when its arrival；4) new data reaches: according to the calculating time of data and free time,
It is set in the position in space；5) data end of transmission: data transmission success, remaining time is 0 at this time；6) data transmission is lost
Lose: data fall in the remaining time axis left side, at this time free time < 0.
As shown in Fig. 2, when data terminal equipment (DTE) be passed to data when, network bottom layer by real time data with it is non real-time
Data classification will be in original prerequisite variable (First-Come-First-Service) strategy using dual stack classification policy
All data account for that a storehouse is changed to real time data and non-real-time data respectively accounts for a storehouse altogether.When data start transmission when pair
Stack space where its is answered, empties its corresponding stack space after data transmission success, and it is reassigned to newly arrived
Data.
As shown in figure 3, when data by successful classification into respective storehouse after, network will use dual cycle queue, will be original
One single cycle queue of all data sharings in prerequisite variable (First-Come-First-Service) strategy is changed to double
Round-robin queue.The data of strong real-time are come into first queue from the storehouse where real time data, from non real-time nature data
Non real-time nature data are placed in second queue in the storehouse at place and are transmitted.The Min life of data is set as the team in queue
The minimum time period of column data transmission.Data scheduler dynamically adjusted between two queues according to the queue minimum time period
Data switching time.Its scheduling strategy is as follows:
1, real time data initially enters the high queue 1 of priority and waits when entering queue to be scheduled waiting.
2, the real time data in high-priority queue is transmitted first.If do not dispatched in queue in high priority into
Journey, the then process in schedules low priority queue.The present invention shares 2 two queue 1, queue queues, and if only if in queue 1
Do not have just to go the data of 2 awaiting transmission of scheduling queue when data latency transmission.
3, for each data in the same queue, time slice scheduling is distributed according to FCFS.Such as the timeslice of queue 1
Enter queue 2 then the operation in queue 1 experienced N number of timeslice (i.e. S1) if having not been completed afterwards for N and wait, if
The timeslice of queue 2 is finished (i.e. S2), and data can't be transmitted afterwards, then be again introduced into queue 2, until completing.
4, each data in the last one queue distribute time slice scheduling according to round-robin.
5, the data in the queue of low priority and have newly arrived real time data in transmission, at this time must immediately
The data transmitted put back to the tail of the queue of current queue, then transmit newly arrived real time data in high-priority queue.Change and
Yan Zhi, any moment can just go the data (preemptive type) of transmission queue 2 only when queue 1 is all empty.It illustrates, when
When being transferred to the data of current queue once again, only distribute last time also unfinished timeslice, the queue of not reallocating is corresponding
Full time piece.
Wherein in self-adaptive secondary feedback scheduling, when data i is completed from submission, reading, caching, scheduling, execution, defeated
Out, to a complete cycle process of release, then the data turnaround time of data i is obtained, T is usedciIt indicates.When data i passes through heap
When stack arrival queue etc. is to be transmitted, queue is known as data latency time to the processing delay of data i, uses TwiIt indicates.When data i is logical
When crossing queue transmission, queue is known as data transmission period to the transmission time of data i, uses TtiIt indicates.When data i end of transmission
When, the locating time is known as the data deadline, uses TfiIt indicates.Data submission time then uses TsiIt indicates.The then data turnaround time
TciIt should calculate as follows:
Tci=Twi+Tti*(Tfi-Tsi) (1)
Wherein TacIt is averaged the turnaround time, can be calculated with following formula for data:
Tac=(Tc1+Tc2+...+Tcn)/n (2)
And the with of data weighs turnaround time WciAnd the average with power turnaround time is then calculated with following formula:
Wci=Tci/Tti (3)
Wac=(Wc1+Wc2+...+Wcn)/n (4)
As shown in figure 4, when data successfully assigned to corresponding queue etc. it is to be transmitted when, scheduling strategy employed in queue is then
For prerequisite variable (First-Come-First-Service) strategy.Wherein queue 1 is high-priority queue, and queue 2 is low
Priority query, therefore first the data in queue 1 are transmitted.Its scheduling strategy is as follows:
1, the real time data in high-priority queue 1 is transmitted first.Queue 1 is initialized, by the real time data in queue 1 by successive
Sequence arranges, and time quantum T is zero at this time.
2, start sequentially to transmit the real time data in queue 1.
3, the data deadline T of data i in queue 1 is calculatedfi, data turnaround time TciAnd with weighs turnaround time Wci:
Tfi=Tsti+Tti, TstiStart transmission time (5) for data
Tci=Tfi-Tsi (6)
4, when data i end of transmission, time quantum T value is changed:
T=T+Tti (7)
If 5, still having data latency transmission in queue 1, repeatedly 2-5；With power turnover is calculated if data end of transmission in queue 1
Time WciAnd data are averaged turnaround time Tac, start to transmit the data in queue 2 after calculating.
The average value E (X) that wherein data reach in the unit time calculates (x is data amount check) as follows:
Due to
Data arithmetic average value E (i) calculates (ρ is to have probability existing for data in queue) as follows:
Queue average response time R calculates (λ is data average arrival rate) as follows:
In order to verify a kind of validity of the Ethernet design scheme based on dynamic secondary feedback scheduling of the present invention, by present invention side
System schema is compared case with simple prerequisite variable (First-Come-First-Service), as shown in figure 5, this hair
It is bright to be better than FCFS scheme in terms of networked-induced delay.The present invention in addition to can real-time monitoring data state, there is also additional real-time
Property promoted.Such as during data are incoming, real time data and non-real-time data can be classified before entering queue in advance,
And the storehouse classified is corresponded to the queue of respective priority.In contrast, FCFS system can only sequentially transmit data,
The transmission of real time data can not be shifted to an earlier date, not can guarantee the real-time of real time data.Therefore, real-time of the invention is higher than FCFS
The real-time of system
The present invention is a kind of Ethernet design scheme based on dynamic secondary feedback scheduling, in order to prevent non-real-time data
Transmission time is too long to be caused to occupy real-time data channel, and then guarantees the real-time Transmission of real time data, improves networked-induced delay.This
Invention is made of two parts, including Dynamic Scheduling Strategy and second level feedback scheduling strategy.Dynamic Scheduling Strategy ensures data
It can be monitored in real-time after being passed to from data terminal equipment (DTE).For the latter, second level feedback scheduling strategy is dual stack classification plan
Slightly with the combination of dual cycle queue strategy, it can be provided safeguard for the transmission of real time data, improve the reality of entire ethernet network
Shi Xing.
Claims (3)
1. a kind of Ethernet design scheme based on dynamic secondary feedback scheduling, is made of two parts: Dynamic Scheduling Strategy and
Self-adaptive secondary feedback scheduling strategy；Ethernet network system response mode uses dynamic dispatching, and it is existing that it not only contains storehouse
There is the real-time status of real time data, the state after new real time data reaches is further comprised, so that it is guaranteed that ethernet network system
Rapid reaction can be made to the environment of variation, ethernet network is enabled to monitor the dynamic requirements of real time data in real time,
Increase its overall real-time；Self-adaptive secondary feedback scheduling strategy combines dual cycle queue priority policy and storehouse classification plan
Slightly, classify to real time data and non-real-time data, it is ensured that the priority of real time data can be met.
2. a kind of Ethernet design scheme based on dynamic secondary feedback scheduling according to claim 1, it is characterised in that:
The Dynamic Scheduling Strategy brings into operation from MAC, and the state of Real-time Monitoring Data is to guarantee that data accurately can successfully be sent out
Sending, data real-time status is with data free time --- this spatial model of data redundancy time indicates, while according to data bit
The variation set splits data into 6 kinds: 1) data start to transmit；2) data are without transmission；3) data do not reach；4) new data
It reaches；5) data end of transmission；6) data transmission fails；When data terminal equipment (DTE) is passed to data, incited somebody to action in network bottom layer
Real time data and non-real-time data are classified, using dual stack classification policy, by original prerequisite variable (First-Come-
First-Service) all data in strategy account for that a storehouse is changed to real time data and non-real-time data respectively accounts for a heap altogether
Stack corresponds to stack space where it when data start transmission, its corresponding stack space is emptied after data transmission success, and will
It is reassigned to newly arrived data.
3. a kind of Ethernet design scheme based on dynamic secondary feedback scheduling according to claim 1, it is characterised in that:
The self-adaptive secondary feedback scheduling is the combination of dual stack classification policy and dual cycle queue strategy, when data are successfully divided
Class is into after respective storehouse, and network is by all numbers in original prerequisite variable (First-Come-First-Service) strategy
It is changed to dual cycle queue according to a shared single cycle queue, comes the data of strong real-time from the storehouse where real time data
First queue is transmitted, queue from non real-time nature data are placed in second queue in the storehouse where non real-time nature data
The Min life of middle data is set as the minimum time period of queuing data transmission, and data scheduler is according to queue minimum time
Period dynamically adjusts the data switching time between two queues, and scheduling strategy is as follows:
(1), real time data initially enters the high queue 1 of priority and waits when entering queue to be scheduled waiting；
(2), the real time data in high-priority queue is transmitted first；
The process if process that do not dispatched in queue in high priority, in schedules low priority queue；The present invention is shared
2 two queue 1, queue queues, just go scheduling queue 2 medium to be passed when not having data latency transmission in queue 1
Defeated data；
(3), for each data in the same queue, time slice scheduling is distributed according to FCFS；
For example the timeslice of queue 1 is N, then the operation in queue 1 experienced N number of timeslice (i.e. S1) if complete not yet afterwards
At then into the waiting of queue 2, if the timeslice of queue 2 is finished (i.e. S2), data can't be transmitted afterwards, be again introduced into team
Column 2, until completing；
(4), each data in the last one queue distribute time slice scheduling according to round-robin；
(5), the data in the queue of low priority are in transmission, and have newly arrived real time data, at this time must be immediately just
The tail of the queue of current queue is put back in the data of transmission, then transmits newly arrived real time data in high-priority queue；For changing
It, any moment can just go the data (preemptive type) of transmission queue 2, illustrate, when again only when queue 1 is all empty
Degree is when being transferred to the data of current queue, only distributes last time also unfinished timeslice, and the queue of not reallocating is corresponding complete
Whole timeslice.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910422802.4A CN110138686B (en) | 2019-05-21 | 2019-05-21 | Ethernet design method based on dynamic secondary feedback scheduling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910422802.4A CN110138686B (en) | 2019-05-21 | 2019-05-21 | Ethernet design method based on dynamic secondary feedback scheduling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110138686A true CN110138686A (en) | 2019-08-16 |
| CN110138686B CN110138686B (en) | 2022-12-27 |
Family
ID=67571801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910422802.4A Active CN110138686B (en) | 2019-05-21 | 2019-05-21 | Ethernet design method based on dynamic secondary feedback scheduling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110138686B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010021661A1 (en) * | 2000-03-13 | 2001-09-13 | Mitsubishi Denki Kabushiki Kaisha | Method of transmitting between a base station in an access network and an access network controller of a telecommunications system |
| CN101553769A (en) * | 2005-10-11 | 2009-10-07 | 努雅公司 | Method and system for tracking and monitoring computer applications |
| US20100020689A1 (en) * | 2007-01-29 | 2010-01-28 | Bob Tang | Immediate ready implementation of virtually congestion free guaranteed service capable network : nextgentcp/ftp/udp intermediate buffer cyclical sack re-use |
| CN103685064A (en) * | 2013-12-10 | 2014-03-26 | 中国人民解放军理工大学 | Heterogeneous cascade network QoS closed-loop mutual coupling system and method |
-
2019
- 2019-05-21 CN CN201910422802.4A patent/CN110138686B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010021661A1 (en) * | 2000-03-13 | 2001-09-13 | Mitsubishi Denki Kabushiki Kaisha | Method of transmitting between a base station in an access network and an access network controller of a telecommunications system |
| CN101553769A (en) * | 2005-10-11 | 2009-10-07 | 努雅公司 | Method and system for tracking and monitoring computer applications |
| US20100020689A1 (en) * | 2007-01-29 | 2010-01-28 | Bob Tang | Immediate ready implementation of virtually congestion free guaranteed service capable network : nextgentcp/ftp/udp intermediate buffer cyclical sack re-use |
| CN103685064A (en) * | 2013-12-10 | 2014-03-26 | 中国人民解放军理工大学 | Heterogeneous cascade network QoS closed-loop mutual coupling system and method |
Non-Patent Citations (3)
| Title |
|---|
| 宋刚: ""嵌入式实时操作系统任务调度算法研究与改进"", 《中国优秀博硕士学位论文全文数据库(硕士)•信息科技辑》 * |
| 林恺等: "一种基于TinyOS的自适应双环调度策略", 《东北大学学报(自然科学版)》 * |
| 蔡杨湧 等: ""基于工业以太网的列车通信网络调度算法优化"", 《计算机仿真》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110138686B (en) | 2022-12-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103823780B (en) | A kind of Real-time Fieldbus controls dispatching method | |
| CN106789726B (en) | A kind of high concurrent service resource allocation method based on service priority | |
| CN101411134B (en) | Memory management for high speed media access control | |
| CN101621460B (en) | Packet scheduling method and device | |
| AU3778499A (en) | Method and apparatus for supplying requests to a scheduler in an input-buffered multiport switch | |
| CN109697122A (en) | Task processing method, equipment and computer storage medium | |
| CN109104373A (en) | The processing method of network congestion, apparatus and system | |
| CN105409170A (en) | Packet output controller and method for dequeuing multiple packets from one scheduled output queue and/or using over- scheduling to schedule output queues | |
| GB2483737A (en) | Workflow scheduler for server cluster with blocking assignment of order sensitive flows and non blocking assignment of other flows | |
| CN106878132A (en) | Bus dispatching method based on cycle and priority | |
| CN105025524A (en) | A multi-path parallel-transmitted data scheduling method and a transmission control protocol | |
| US8588239B2 (en) | Relaying apparatus and packet relaying apparatus | |
| CN108093486A (en) | The dispatching method of mixed data flow transmitting in industrial wireless sensor network | |
| CN109783225A (en) | A kind of tenant's priority management method and system of multi-tenant big data platform | |
| CN101414974A (en) | Data flux management method based on communication terminal | |
| CN105530157B (en) | The multiple virtual links of the AFDX network switch share the framework and method of credit | |
| CN102957625B (en) | A kind of system and method for realizing end-to-end multilevel service quality | |
| CN110138686A (en) | A kind of Ethernet design scheme based on dynamic secondary feedback scheduling | |
| CN106603215A (en) | Unfair network channel resource sharing method based on ZigBee | |
| CN102281202A (en) | Dispatching method, server and system | |
| CN107070695A (en) | A kind of bus-network loaded self-adaptive communication means | |
| CN108462650A (en) | A kind of output unit based on TTE interchangers | |
| CN102594670B (en) | Multiport multi-flow scheduling method, device and equipment | |
| Shi | Real-time communication services for networks on chip | |
| Wang et al. | Randomized load-balanced routing for fat-tree networks |
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 |