CN109450817B - Mixed scheduling method for time-triggered Ethernet multi-service message transmission - Google Patents
Mixed scheduling method for time-triggered Ethernet multi-service message transmission Download PDFInfo
- Publication number
- CN109450817B CN109450817B CN201811492235.1A CN201811492235A CN109450817B CN 109450817 B CN109450817 B CN 109450817B CN 201811492235 A CN201811492235 A CN 201811492235A CN 109450817 B CN109450817 B CN 109450817B
- Authority
- CN
- China
- Prior art keywords
- data
- transmission
- message
- time
- sending
- 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.)
- Active
Links
Images
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
- H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
-
- 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
-
- 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 invention belongs to the network communication technology. Aiming at the transmission requirements of various service messages in the AS6802 standard, a mixed scheduling method for transmitting various service messages of the time-triggered Ethernet and a transmission scheduling message interaction mechanism suitable for the method are formulated, the time-triggered characteristic of TT messages, the event-triggered characteristic of PCF messages, the rate limiting characteristic of RC messages and the best effort transmission characteristic of BE messages are met, the problem that the transmission characteristics of various service messages in the time-triggered Ethernet are difficult to BE considered is solved, and technical support is provided for designing and realizing an end system supporting various service flows.
Description
Technical Field
The invention belongs to the network communication technology, and relates to a hybrid scheduling method for sending multiple messages of a time-triggered Ethernet and a related data structure.
Background
In the ethernet conforming to the IEEE802.3 standard, a network device uses an event-triggered ethernet transmission principle, and any network device can access the network at any time point, using a first-come-first-serve principle. Communication between network devices over a shared medium necessarily results in some messages causing indeterminate jitter and cumulative delay in transmission.
The AFDX network conforming to the ARINC664Part7 standard is based on the Ethernet, and Virtual Links (VL), bandwidth Allocation Gaps (BAG), jitter management and other mechanisms are added on a data link layer, so that the conflict in the message transmission process is effectively avoided, the message has instantaneity and certainty, and the requirements of reliability and certainty of avionic system information transmission are met.
TTE networks conforming to the AS6802 standard, compatible Ethernet networks and AFDX networks, use network synchronization time based on coordination among network devices to reduce transmission delay and jitter, and support TT messages of a time trigger mechanism, RC messages (namely AFDX messages) and BE messages of an event trigger mechanism (Ethernet messages), thereby meeting application requirements of different real-time and security levels in a single network.
The AS6802 standard supports three service messages, i.e., TT, RC, and BE, at a network level, and does not integrate them into one end system, nor proposes a transmission scheduling mechanism among the three messages.
The invention designs a sending scheduling message interaction mechanism meeting the message transmission characteristics of PCF, TT, RC, BE and the like, formulates a non-preemptive time triggering and priority mixed scheduling method of multiple service messages, realizes sending scheduling according with the time triggering characteristic of TT message, the event triggering characteristic of PCF message, the rate limiting characteristic of RC message and the best effort transmission characteristic of BE message, solves the problem that the transmission characteristics of multiple service messages are difficult to BE considered in time-triggered Ethernet, and provides technical support for designing and realizing a terminal system supporting multiple service flows.
Disclosure of Invention
The purpose of the invention is as follows:
the invention designs a sending and scheduling message interaction mechanism meeting the message transmission characteristics of PCF, TT, RC, BE and the like, establishes a non-preemptive time triggering and priority mixed scheduling method of multiple service messages, solves the problem that the transmission characteristics of the multiple service messages are difficult to BE considered in a time triggering Ethernet, and provides technical support for designing and realizing a terminal system supporting multiple service flows.
The technical scheme is as follows:
the mixed scheduling method for time-triggered Ethernet multi-service message transmission is characterized by specifically comprising the following steps of:
step 1: the host application and the end system synchronization function are used as data sources to generate initial data to be sent; data generated by the host application respectively passes through a TT transmitter, an RC transmitter and a BE transmitter according to types and is packaged into a TT message triggered by time, an RC message triggered by an event and a BE message; data generated by the end system synchronization function is packaged into event-triggered PCF messages through PCF transmitters, and the four transmitters TT, PCF, RC and BE provide the following scheduling information:
(1) The four transmitters respectively maintain corresponding transmission buffers;
(2) The TT sender maintains a sending schedule which comprises one or more sending time points and corresponding buffer areas, and the schedule is organized according to the time sequence;
(3) The PCF transmitter maintaining a transmission request signal and a fixed buffer;
(4) The RC transmitter maintains a transmission schedule containing the status of all VLs;
(5) The BE transmitter maintains a transmit queue containing whether BE data is ready, the address and length of the BE data;
step 2: the scheduler works in parallel with all transmitters, sets a first item in the TT plan table as a current transmission plan, and skips to the step 3;
and step 3: if the end system is in a synchronous state, reading a transmission plan, recording the length of a data frame which can be transmitted before a scheduled time point as an allowable length, changing with the change of a local clock, and otherwise, setting the allowable length to be fixed as the maximum frame length and 1518 bytes.
If the allowable length is less than 64 bytes, jumping to step 4; otherwise, jumping to step 6;
and 4, step 4: the scheduler prohibits data transmission until a scheduled time point, judges whether scheduled data is ready or not, and jumps to the step 5 after transmission if the scheduled data is ready; otherwise, directly jumping to the step 5.
And 5: setting the next item, wherein the next item of the table tail item is a table head item, and jumping to the step 3 for sending the plan;
step 6: and judging whether the length of the data frame is less than or equal to the allowed length according to the priority sequence of PCF, RC and BE from high to low, simultaneously judging the data frames of all VLs of the RC service, if so, sending a frame, and jumping to the step 3 after completion, otherwise, directly jumping to the step 3.
In step 1, the state of VL includes whether RC data is ready, data address, and data length.
Has the advantages that:
based on three service messages and transmission characteristics thereof provided by the AS6802 standard, a scheduling method for sending multiple service messages in a time-triggered Ethernet is formulated, the time-triggered characteristic of TT messages, the rate limiting characteristic of RC messages and the best effort transmission characteristic of BE messages are met, and technical support is provided for designing and realizing an end system supporting multiple service flows.
The method adopts non-preemptive scheduling, avoids a large number of incomplete fragments on the network caused by preemptive scheduling, and saves the bandwidth occupied by sending the incomplete fragments.
In order to ensure the time-triggered characteristic of the TT frame, the method presets the transmission time point for the transmission of the TT frame in advance, all messages which can occupy the time point are not allowed to be transmitted, and the accurate timing transmission of the TT frame is ensured, as shown in fig. 2. When TT is scheduled to the ith item, if TT frames are transmitted, non-TT frames in the area 2 are transmitted according to priority; if no TT frame is transmitted, the non-TT frames in regions 1 and 2 are transmitted according to priority.
When RC scheduling is carried out, when one VL prepares to send data frames with the length exceeding the allowed length and the other VL prepares to send data frames with the length lower than the allowed length, the method simultaneously controls the sending of all VLs of the RC service, ensures that the former does not block the latter, and improves the use efficiency of network bandwidth.
Drawings
Fig. 1 is a flow chart of a scheduling method.
FIG. 2 is a relationship between TT transmission time point and non-TT frame transmission
Detailed Description
The implementation steps of the present invention are further described below with reference to the accompanying drawings, and as shown in fig. 1, the detailed implementation steps of the technical solution of the present invention include:
step 1: the host application and end-system synchronization function act as a data source to generate the initial data to be sent. Data generated by the host application respectively passes through the TT transmitter, the RC transmitter and the BE transmitter according to types and is packaged into a TT message triggered by time, an RC message triggered by an event and a BE message. The data generated by the end-system synchronization function is packaged into an event-triggered PCF message by the PCF transmitter. All transmitters and schedulers work in parallel, the transmitters providing the following information:
(1) TT sender, PCF sender, RC sender and BE sender maintain the corresponding sending buffer separately, wherein PCF sends the buffer and only includes a unit;
(2) The TT sender maintains a sending schedule table ttSchTbl, each item in the table is a binary group formed by { ttPit, ttStatus }, wherein:
a) ttPit represents a statically configured sending time point, and the items of ttSchTbl are arranged according to the sequence of ttPit;
b) ttStatus represents the state information of the data corresponding to the transmission time point ttPit, and when the value of ttStatus is between 0 and the TT maximum transmission buffer number ttMaxBufNo, the state information represents the buffer number and is directly mapped to the position of the data to be transmitted in the buffer; otherwise, the data is not ready;
(3) The PCF sender maintains a send request signal pcfReq;
(4) The RC transmitter maintains an RC transmission schedule rcSchTbl, each entry in the table representing status information of a transmitting virtual link rcVL of an RC data, represented by a triplet of { rcBuf, rcBufLen, rcReq }:
a) rcBuf represents the memory address of the data frame to be sent by the rcVL;
b) rcBufLen represents the frame length of the data frame to be transmitted by the rcVL;
c) rcReq indicates whether the rcVL has data to send;
d) The data to be sent refers to data passing through a flow regulating function of an RC (remote control) transmitter;
(5) The BE transmitter maintains a BE data transmission queue, each entry in the queue is composed of a transmission buffer address beBuf and a length beBufLen of the BE data, and whether the queue is empty is represented by beReq:
and 2, step: the scheduler works in parallel with all the transmitters, sets a first item in the TT plan table as a current plan item ttSchItem, and skips to the step 3;
and step 3: if the end system is in the stable synchronization state Sync, reading the ttSchItem, recording the sending time point ttPit, and jumping to the step 4; otherwise, setting the value of the allowed sending frame length txLenAllowed as a value not less than the maximum frame length, and jumping to the step 8;
and 4, step 4: reading local clock localcock, calculating its time difference timeDiff to ttPit, considering localcock takes cluster cycle duration as the period, so the timeDiff calculation method is as follows:
timeDiff = ttPit-localClock if localClock is less than ttPit;
otherwise timeDiff = ClusterCycleDuration + ttPit-localClock;
then jumping to step 5;
and 5: calculating a value of allowed transmission frame length txLenAllowed:
txLenAllowed＝(timeDiff-t shift )*netspeed-ifsLen；
where netspeed represents the network rate; ifsLen represents the length of a preamble, a start delimiter and a frame interval, and is fixed to 20 bytes; t is t shift The switching time is not less than the time difference between the permission of sending the data and the arrival of the data at the physical layer. Due to the difference of addressing and reading time of PCF, RC and BE data frames, t shift May be different, and thus the allowed frame length values of the three data frames are txLenAllowed respectively PCF 、txLenAllowed RC And txLenAllowed BE ；
If the txLenAllowed is lower than 64, waiting until localClock is equal to ttPit, and jumping to step 6; otherwise, jumping to the step 8;
and 6: reading the data state ttStatus of ttSchItem, if ttStatus indicates that the data is ready, reading and sending TT data from the TT sending buffer according to the position indicated by ttStatus, and jumping to the step 7 after the sending is finished; otherwise, directly jumping to the step 7;
and 7: if ttSchItem is the last item of the TT schedule table, setting the first item to be ttSchItem; otherwise, setting the next item to be ttSchItem. Then jumping to the step 3;
and 8: responding to the sending request of messages such as PCF, RC, BE and the like:
pcfGnt＝pcfReq AND(64≤txLenAllowed PCF )
rcvlGnt = satisfies { rcReq } i AND rcBufLen i ≤txLenAllowed RC ) Minimum value of i }
beGnt＝NOT(pcfGnt AND rcGnt)AND beReq AND(beBufLen≤
txLenAllowed BE )
Then jumping to step 9;
and step 9: and processing the transmission of messages such as PCF, RC, BE and the like:
(1) If pcfGnt is equal to 1, sending PCF data frame in pcfBuf, and entering step 3 after completion;
(2) If rcGnt is equal to 1, sending an RC data frame specified by { rcbuf, rcBufLen } of the rcvlGnt item, and entering step 3 after completion;
(3) If beGnt is equal to 1, the BE data specified by { bebuf, beBufLen } is transmitted, and step 3 is entered after completion.
The technical principle and the advantages of the technical scheme of the invention comprise:
advantages of non-preemptive scheduling:
by adopting preemptive scheduling, high-priority data frames can be generated to preempt a physical link, so that the transmission of low-priority data frames is not finished, a large number of fragments appear on a network, and meanwhile, the low-priority data frames need to be transmitted again, and the bandwidth utilization rate is reduced.
Resource reservation of TT data:
due to the time-triggered nature of TT data, and the requirement for non-preemptive scheduling, if a non-TT data frame cannot be transmitted before the TT transmission time point arrives, its transmission is prohibited. Therefore, the method defines the txLenAllowed parameter to control the transmission of the non-TT data frame. The txLenAllowed curve is shown in fig. 2 according to the above steps.
When localClock is at reserved time points i and ttPit i In the meantime, txLenAllowed is smaller than 64, the control logic in step 8 ensures that the non-TT data frames are forbidden to be sent, and ensures that TT data frames arriving on time can be sentSending on time; when localClock = ttPiti, if the transmission data at the time point is ready, switching to the next TT transmission point ttPitj after the transmission is completed; otherwise, switching immediately. The non-TT data frame is controlled by ttPitj.
TT data in the time-triggered Ethernet occupies part of bandwidth, and the bandwidth of all data does not exceed the interface bandwidth, so that enough time is available between two adjacent TT sending time points for sending non-TT data frames.
Priority scheduling of non-TT data is implemented in step 8.
Claims (2)
1. The mixed scheduling method for time-triggered Ethernet multi-service message transmission is characterized by specifically comprising the following steps of:
step 1: the host application and the end system synchronization function are used as data sources to generate initial data to be sent; data generated by the host application respectively passes through a TT transmitter, an RC transmitter and a BE transmitter according to types and is packaged into a TT message triggered by time, an RC message triggered by an event and a BE message; data generated by the end system synchronization function is packaged into event-triggered PCF messages through PCF transmitters, and the four transmitters TT, PCF, RC and BE provide the following scheduling information:
(1) Four transmitters respectively maintain corresponding transmission buffers;
(2) The TT sender maintains a sending schedule which comprises one or more sending time points and corresponding buffer zones, and the schedule is organized according to the time sequence;
(3) The PCF transmitter maintaining a transmission request signal and a fixed buffer;
(4) The RC transmitter maintains a transmission schedule containing the status of all VLs;
(5) The BE transmitter maintains a transmit queue containing whether BE data is ready, BE
The address and length of the data;
and 2, step: the scheduler works in parallel with all the transmitters, sets the first item in the TT plan table as the current transmission plan, and skips to the step 3;
and step 3: if the end system is in a synchronous state, reading a sending plan, recording the length of a data frame which can BE sent before a scheduled time point as an allowable length, and changing along with the change of a local clock, wherein the allowable length comprises an allowable sending frame length value of a PCF message, an allowable sending frame length value of an RC message and an allowable sending frame length value of a BE message, otherwise, setting the allowable length to BE fixed as the maximum frame length, and 1518 bytes;
if the frame length value allowed to BE sent of PCF message, the frame length value allowed to BE sent of RC message and the frame length value allowed to BE sent of BE message in the allowed length are all less than 64 bytes, skipping to step 4; otherwise, jumping to step 6, calculating the value of the allowed transmission frame length by the following formula:
txLenAllowed＝(timeDiff-tshift)*netspeed-ifsLen，
wherein txLenAllowed is a value of a frame length allowed to be sent, netspeed represents a network rate, ifsLen represents a preamble, a start delimiter and a frame interval, tshift represents switching time, and timeDiff is a time difference from a local clock to a statically configured sending time point of TT;
and 4, step 4: the scheduler prohibits data transmission until a scheduled time point, judges whether scheduled data is ready or not, and jumps to the step 5 after transmission if the scheduled data is ready; otherwise, directly jumping to the step 5;
and 5: setting the next item, wherein the next item of the table tail item is a table head item, and jumping to the step 3 for sending the plan;
and 6: and judging whether the data frame length of each message is less than or equal to the corresponding transmission allowed frame length value according to the priority sequence of PCF, RC and BE from high to low, judging simultaneously the data frames of all VLs of the RC service, if so, transmitting a frame, and jumping to the step 3 after finishing, otherwise, directly jumping to the step 3.
2. The mixed scheduling method for time-triggered ethernet multi-service message transmission according to claim 1, wherein in step 1, the status of VL comprises whether RC data is ready, data address, data length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811492235.1A CN109450817B (en) | 2018-12-07 | 2018-12-07 | Mixed scheduling method for time-triggered Ethernet multi-service message transmission |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811492235.1A CN109450817B (en) | 2018-12-07 | 2018-12-07 | Mixed scheduling method for time-triggered Ethernet multi-service message transmission |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109450817A CN109450817A (en) | 2019-03-08 |
| CN109450817B true CN109450817B (en) | 2023-04-07 |
Family
ID=65556911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811492235.1A Active CN109450817B (en) | 2018-12-07 | 2018-12-07 | Mixed scheduling method for time-triggered Ethernet multi-service message transmission |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109450817B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111092712B (en) * | 2019-11-28 | 2022-04-05 | 中国航空工业集团公司西安航空计算技术研究所 | Preemptive scheduling method for time-triggered Ethernet |
| CN114448546B (en) * | 2021-12-29 | 2024-01-23 | 中国航空工业集团公司西安航空计算技术研究所 | Scheduling method of time-triggered network message based on timer |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8473563B2 (en) * | 2005-04-22 | 2013-06-25 | GM Global Technology Operations LLC | Extensible scheduling of messages on time-triggered busses |
| US9762501B2 (en) * | 2015-04-01 | 2017-09-12 | Honeywell International Inc. | Systematic hybrid network scheduling for multiple traffic classes with host timing and phase constraints |
| CN108462650B (en) * | 2016-12-12 | 2021-09-14 | 中国航空工业集团公司西安航空计算技术研究所 | Output unit based on TTE switch |
| CN108449285B (en) * | 2018-03-01 | 2022-05-17 | 中国航空无线电电子研究所 | Split scheduling method applied to shared Ethernet |
-
2018
- 2018-12-07 CN CN201811492235.1A patent/CN109450817B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109450817A (en) | 2019-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112105080B (en) | Time-sensitive network data transmission system and transmission method | |
| US10218631B2 (en) | Method for traffic shaping of data frames in network and device and computer program product therefor | |
| CN1668020B (en) | Method, apparatus and system for guaranteed packet delivery times in asynchronous networks | |
| Hong et al. | Bandwidth allocation scheme in CAN protocol | |
| US8315274B2 (en) | System and method for supporting synchronous system communications and operations | |
| US20070127521A1 (en) | Interface between network data bus application and avionics data bus | |
| EP2453613B1 (en) | Method and apparatus for allocating and prioritizing data transmission | |
| US7792153B2 (en) | Sequencing multi-source messages for delivery as partial sets to multiple destinations | |
| CN109450817B (en) | Mixed scheduling method for time-triggered Ethernet multi-service message transmission | |
| CN104717149A (en) | Data frame scheduling method based on virtual links | |
| US20200322180A1 (en) | Method for Transmitting Data Packets, Controller and System Having a Controller | |
| CN114424507A (en) | Method for transmitting data packets and device for carrying out said method | |
| CN102347877A (en) | Bus dispatching method and device | |
| EP3487132A1 (en) | Packet processing method and router | |
| Rom et al. | Message-based priority functions in local multiaccess communication systems | |
| Cena et al. | Performance analysis of Byteflight networks | |
| CN110492960A (en) | Synchronized communication method based on Ethernet | |
| CN109525315B (en) | Optical fiber channel network end system based on time triggering | |
| CN110809012B (en) | Train network communication data scheduling control method | |
| CN113014465A (en) | Bus transmission interface system based on quality of service characteristics and design method | |
| CN114221911A (en) | Message hybrid scheduling method and device, electronic equipment and storage medium | |
| CN114448546B (en) | Scheduling method of time-triggered network message based on timer | |
| Lea et al. | A channel access protocol for integrated voice/data applications | |
| Cena et al. | A Fair Access Mechanism to Support Fragmented Data Transfers in CAN | |
| CN110601744B (en) | Dispatching method for sending virtual link of AFDX (avionics full Duplex switched Ethernet) network |
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 |