CN109981481A - A kind of design method of more scene protections - Google Patents

A kind of design method of more scene protections Download PDF

Info

Publication number
CN109981481A
CN109981481A CN201910040657.3A CN201910040657A CN109981481A CN 109981481 A CN109981481 A CN 109981481A CN 201910040657 A CN201910040657 A CN 201910040657A CN 109981481 A CN109981481 A CN 109981481A
Authority
CN
China
Prior art keywords
message
sequence number
scene
link
initial message
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
Application number
CN201910040657.3A
Other languages
Chinese (zh)
Other versions
CN109981481B (en
Inventor
陈俊来
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Polytron Technologies Inc
Original Assignee
Beijing Polytron Technologies Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Polytron Technologies Inc filed Critical Beijing Polytron Technologies Inc
Priority to CN201910040657.3A priority Critical patent/CN109981481B/en
Publication of CN109981481A publication Critical patent/CN109981481A/en
Application granted granted Critical
Publication of CN109981481B publication Critical patent/CN109981481B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/02Topology update or discovery
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/32Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/50Queue scheduling
    • H04L47/62Queue scheduling characterised by scheduling criteria
    • H04L47/6245Modifications to standard FIFO or LIFO

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Communication Control (AREA)
  • Small-Scale Networks (AREA)

Abstract

The present invention relates to a kind of design methods of more scene protections, wherein includes: that each message of file is divided into initial message, midamble and completes message;The message received is judged, and according to whether, if it is initial message, passes through link number management, sequence number maintenance, the initial Message processing process of scene protection mechanism execution for initial message;After receiving non-initial message, start the flow chart of data processing of non-initial message by link number management, sequence number maintenance, scene protection mechanism.A kind of design method of more scene protections of the invention, entire design introduce, and the network transmission concurrent for multi-link can eliminate the insecurity factor of transmission, it is ensured that the zero defect and high efficiency of transmission of data.

Description

A kind of design method of more scene protections
Technical field
The present invention relates to network communication technology, in particular to a kind of design method of more scene protections.
Background technique
In network communication, in order to guarantee the safety of network data transmission, it usually needs to network data to be transmitted into Row processing.Due to the limitation and requirement of data processing module performance, the network packet of input data processing module must be connected Continuous.The objective condition such as security mechanism is unsound due to Ethernet transmission, cache resources are limited in logic, larger handling capacity, The data of large bandwidth transmit occasion, unavoidably will appear certain messages and lose in transmission process, this will result in data Processing gets muddled, and whole system is caused to be paralysed.
Summary of the invention
The purpose of the present invention is to provide a kind of design methods of more scene protections, for solving asking for the above-mentioned prior art Topic.
A kind of design method of more scene protections of the present invention, wherein include: by each message of file be divided into initial message, Midamble and completion message;The message received is judged, and according to whether executing below step for initial message: After receiving initial message, start the flow chart of data processing of initial message, comprising: S1, query link number manage table, are confirmed whether There is available link number, if it is not, giving up message, if so, then executing S2;S2, access link number management table, are file Each message application link number;S3, the link number of application is write in sequence number Maintenance Table;Whether S4, inquiry system can be located Initial message is managed, if not all right, by initial packet buffer, until being capable of handling initial message, if it is possible to the initial report of processing Text thens follow the steps S5;S5, initial message data processing is carried out;S6, it is extracted from the processed initial message data of output Scene, and scene protection table is written;After receiving non-initial message, start the flow chart of data processing of non-initial message, comprising: A1, the link index sequence Maintenance Table with non-initial message;A2, judge whether the sequence number of non-initial message is equal to or compares Sequence number in Maintenance Table is big by 1, if it is, executing A3, otherwise gives up non-initial message;A3, by current non-initial message Sequence number writes sequence number Maintenance Table, and renewal sequence number;A4, it is indexed now with the sequence number and link number of current non-initial message Field protection table, and read field data;Whether A5, inquiry are capable of handling non-initial message, if it is then step A6 is executed, If not all right, by non-initial packet buffer, until can handle non-initial message, then execute A6, A6, by non-initial message and Field data carries out data processing together;A7, " scene " is extracted from treated data, and scene protection table is written;A8, sentence Whether the non-initial message that breaks is to complete message, if so, thening follow the steps A9, otherwise, executes step A10;A9, will link number can Link management table is write, link number recycling is carried out;A10, return, and wait new message to be received.
One embodiment of the design method of more scene protections according to the present invention, wherein the manager of link number management table Method includes: the caching FIFO for storing the structure of link management table as first in first out;Initialization link number management table: by data 1 It is sequentially written in FIFO to n, wherein n is caching depth;When receiving initial message application hyphen, then from link number management Table takes out link number;The link number of application is distributed to initial message, midamble and the end message of file;Work as reception To the end message for carrying link number, the link number of end message is extracted, and be written to FIFO again, completes link number recycling.
One embodiment of the design method of more scene protections according to the present invention, wherein sequence number Maintenance Table includes: input The same link number fragment message sequence number be it is continuous, the current sequence number for receiving message is in sequence number Maintenance Table Or it is bigger than the sequence number of current sequence number Maintenance Table by 1;In the case of receiving sequence of message number in Maintenance Table, then Maintenance Table is not Become;Sequence of message number is bigger than Maintenance Table sequence number by 1, the last one sequence number of sequence number Maintenance Table is given up.
One embodiment of the design method of more scene protections according to the present invention, wherein the structure packet of sequence number Maintenance Table It includes: link number, each corresponding sequence number of link number.
One embodiment of the design method of more scene protections according to the present invention, wherein sequence number Maintenance Table working mechanism It include: that the link number using distribution be to index the sequence number Maintenance Table being written to using RAM as framework after receiving initial message;It receives To midamble and message is completed, reads the sequence number of maintenance for indexed addressing RAM with link number;If the intermediate report received Text and the sequence number for completing message then continue in sequence number Maintenance Table;If the midamble received and completion message Sequence number+1 then continued in sequence number Maintenance Table, while the sequence number for receiving current slice message being written to Sequence number Maintenance Table;If the midamble received is dropped with completion message because sequence number is unsatisfactory for condition, sequence Number Maintenance Table is constant.
One embodiment of the design method of more scene protections according to the present invention, wherein if scene protection includes: current The data or operation result of fragment message before same file are used in the processing of fragment message, then are delayed to fragment message before It deposits, after current slice message arrives, extracts the data or operation result of prime fragment message, handled.
One embodiment of the design method of more scene protections according to the present invention, wherein scene protection working mechanism includes: After receiving initial message, according to the link number of acquisition and Serial No. index write-in scene, and handled, by processing result It is synchronously written into same scene;In-situ FTIR spectroelectrochemitry includes: to receive non-initial message, and parsing link number is with sequence number and with call number It reads and restoring scene, live by reading is handled with the non-initial message received;Locale Holding include: receive it is non- Initial message, if the sequence number of non-initial message does not need Locale Holding in sequence number Maintenance Table;If non-initial report The sequence number+1 of text in sequence number Maintenance Table, then need by scene and treated result synchronize be saved in scene, oldest shows Field is dropped;If the sequence number of the non-initial message received is not in sequence number Maintenance Table, while non-initial message+1 Not in sequence number Maintenance Table, then the message received is invalid packet.
A kind of design method of more scene protections of the invention, entire design introduces link number management, sequence number is safeguarded, The big security mechanism of scene protection three, the network transmission concurrent for multi-link can eliminate the insecurity factor of transmission, it is ensured that The zero defect and high efficiency of transmission of data.
Detailed description of the invention
Fig. 1 show link number management and indicates to be intended to;
Fig. 2 is the structural schematic diagram of sequence number Maintenance Table;
Fig. 3 is more scene protection schematic diagrames;
Fig. 4 show the flow chart of data processing figure of initial message;
Fig. 5 show midamble and completes the flow chart of data processing figure of message.
Specific embodiment
To keep the purpose of the present invention, content and advantage clearer, with reference to the accompanying drawings and examples, to of the invention Specific embodiment is described in further detail.
In Ethernet transport protocol, data are packaged into message and are transmitted.Since network transmission agreement is to message Length, which has, strictly to be limited, for biggish file, it usually needs is divided into multiple messages and is transmitted respectively.File First transmitting message be known as initial message, the message finally transmitted is known as completing message, and intermediate message is referred to as centre Message.In logical design, different file messages is distinguished with link number, i.e., logic can distribute unique link for each file Number, all messages for being under the jurisdiction of this document have identical link number.And the different fragment messages of each file can be by software It distributes a sequence number, for characterizing order of the message in the fragment message of file.Under normal conditions, same file The sequence number of all messages is necessarily continuously arranged.Logic receives the data message that software issues, and needs with link number and sequence Row number judges the integrality of message transmission procedure, and it is continuous that guarantee, which is input to the message of the same link of data processing module, 's.
Fig. 1 show link number management and indicates to be intended to, as shown in Figure 1, link management method includes:
The structure for storing the table is the caching FIFO (hereinafter referred to as FIFO) of first in first out, and cache size, which need to meet, entirely to be set Required maximum link number in meter.
The working method of link management table includes:
Initialization: being sequentially written in FIFO for data 1 to n, and wherein n is caching depth;
The course of work includes:
Cpu sends " the initial message " of file, does not carry link number (there are no distribution);
When receiving " initial message ", access link number management table applies for a link number;
It sends " returned packet ", the link number of application is sent to cpu;Cpu will link number " the centre report for distributing to file Text " and " end message ";
Cpu sends " midamble " of file, and carries link number;
Cpu sends " end message " of file, and carries link number;
When receiving " end message " for carrying link number, the link number for being somebody's turn to do " end message " is extracted, and be written to again FIFO completes link number recycling;
It is only stored in the link number of FIFO, other file messages could apply for the link number, avoid text in this way The link collision problem of part message during processing.
Further, cpu can parse " returned packet " after " returned packet " for receiving " initial message ".Therefrom take " link number " out, and by the corresponding field of " link number " write-in " midamble " and " completing message ", a text is belonged in this way All messages of part just have one identical " link number ".
2 sequence number Maintenance Tables include:
The sequence number of the fragment message of the same link number of input must be continuous, the current sequence number for receiving message Should in sequence number Maintenance Table or it is bigger than the sequence number of current sequence number Maintenance Table by 1 (i.e. currently receive sequence of message number+1 exist In sequence number Maintenance Table).Situation of the sequence of message number in Maintenance Table is received, message is typically occurred in and is lost in transmission process It loses, when notifications software retransmits the message.The measure that the undesirable message procedure of those sequence numbers is taken: it loses It abandons;Receive sequence of message number it is bigger than Maintenance Table sequence number 1 the case where be to there is new sequence number to arrive, sequence number maintenance need to be written to The last one sequence number of table, sequence number Maintenance Table is given up accordingly, and it is slow that the size of sequence number Maintenance Table need to generally meet scene The size of stored messages.
Fig. 2 is the structural schematic diagram of sequence number Maintenance Table, as shown in Fig. 2, in figure " session0, session1 ..., Session n " represents " link number ", and " seq_num_i, seq_num_j, seq_num_k " represent " sequence number ".Inquire " sequence Number Maintenance Table " use " link number " indexes.
As shown in Fig. 2, sequence number Maintenance Table working mechanism includes:
Program receives " initial message ", and " link number " using distribution is to index the sequence number being written to using RAM as framework Maintenance Table, and be transferred to next stage and need to receive the module of message and handled;
" midamble " and " completing message " are received, can be the sequence that indexed addressing RAM reads maintenance with " link number " Number;
If the sequence number of " midamble " and " completing message " that receives is in sequence number Maintenance Table, message transmissions are arrived Next stage needs to receive the resume module of message;
(sequence number+1 of " midamble " and " complete message ") if that receives in sequence number Maintenance Table, message It is transferred to the resume module that next stage needs to receive message, while the sequence number for receiving current slice message is written to sequence Number Maintenance Table;If packet loss, sequence number Maintenance Table is constant not in sequence number Maintenance Table.
3 scene protections include:
If the processing of current slice message is used prime fragment message and (is split complete message, network is primary Transmitting message be no more than 2KB, if it is greater than 2KB then need carry out message segmentation, be divided into the message of suitable network transmission) data Or operation result, prime fragment message must just be cached, can cache with the RAM inside program, piece can also be cached to Outer cache chip, such as DDR/SRAM/FLASH, cache by taking RAM as an example below.After i.e. current slice message arrives, need to extract Then the data or operation result of prime fragment message are transferred to next stage module together and are handled, this process is also named scene Protection.
Fig. 3 is more scene protection schematic diagrames, as shown in figure 3, { session_i/j/k, seq_num } is " link number " in figure The call number of the access scene caching RAM of " sequence number " composition, Scene i is the scene (fragment message) saved.
Scene protection working mechanism includes:
Program receives " initial message ", according to " link m " and " sequence number 0 " (" initial message " sequence number of acquisition 0) to be index write-in scene scene 0, and it is transferred to next stage and needs to receive the module of message and handled, and will processing knot Fruit is synchronously written into same scene scene 0;
In-situ FTIR spectroelectrochemitry: program receives " midamble " and " completing message ", and parsing " link m " and " sequence number k " is simultaneously Read with call number { m, k-i (1≤i≤n) } and restoring scene, by the scene of reading and the message that receives synchronize be input to down Previous module is handled;
Locale Holding: program receives " midamble " and " completing message ", if " sequence number " is in sequence number Maintenance Table It is interior, then Locale Holding is not needed, because scene has been saved in RAM;(if " sequence number "+1) in sequence number Maintenance Table, Then need by scene with the result after next stage resume module synchronize be saved in scene, oldest scene is dropped;
Invalid packet: if " sequence number " that receives not in sequence number Maintenance Table, while sequence number (" sequence number "+ 1) not in sequence number Maintenance Table, then the message received is invalid packet, is abandoned.
Fig. 4 show the flow chart of data processing figure of initial message, as shown in figure 4, the flow chart of data processing packet of initial message It includes:
S1, the new message that software issues is received;
Whether S2, the message for judging channel are " initial message ", if it is not, then dropping packets, and report is received again Text, if it is, continuing to execute S3;
S3, query link number manage table, are confirmed whether there is available link number, if it is not, give up message, if Have, then executes S4;
S4, access link number management table, are each message application link number of file;
S5, the sequence number of message is write in sequence number Maintenance Table;
Whether S6, inquiry can handle new message, if not all right, by packet buffer, until can handle new report Text executes S7, if it is then executing step S7;
S7, message data processing is carried out;
S8, " scene " is extracted from the processed message data of output, and " scene protection table " is written;
S9, end.
Fig. 5 show midamble and completes the flow chart of data processing figure of message, as shown in figure 5, midamble and completion The data processing of message includes:
A1, software receive the returned packet of initial message, issue after link number to be write to the message of rear class;
A2, judge type of message, if it is midamble, execute step A3, if it is message is completed, then return link number Link management table is write, as link number recycling;
A3, with message " link number " index sequence Maintenance Table;
A4, judge whether the sequence number of current message is equal to or bigger by 1 than the sequence number in Maintenance Table, if it is, executing Otherwise A5 gives up message;
A5, the sequence number of current message is write to sequence number Maintenance Table, and renewal sequence number;
A6, with " sequence number " and " link number " index " scene protection table ", and read field data;
Whether A7, inquiry can handle message, if not all right, by packet buffer, until can handle message executes A8, If it is then executing step A8;
A8, " message " and " scene " is subjected to data processing together;
A9, " scene " is extracted from treated data, and scene protection table is written;It returns and receives new message.
Compared with prior art, technical solution proposed by the present invention passes through building " link number management table ", " sequence number maintenance Table ", " scene protection " three big security mechanism realize the safe transmission of network, designer by saving the scene of multiple messages It can be in conjunction with actual conditions flexible configuration, so the present invention has certain engineering practical value using this mechanism.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art For member, without departing from the technical principles of the invention, several improvement and deformations can also be made, these improvement and deformations Also it should be regarded as protection scope of the present invention.

Claims (7)

1. a kind of design method of more scene protections characterized by comprising by each message of file be divided into initial message, in Between message and complete message;
The message received is judged, and according to whether executing below step for initial message:
After receiving initial message, start the flow chart of data processing of initial message, comprising:
S1, query link number manage table, are confirmed whether there is available link number, if it is not, giving up message, if so, then Execute S2;
S2, access link number management table, are each message application link number of file;
S3, the link number of application is write in sequence number Maintenance Table;
Whether S4, inquiry system are capable of handling initial message, if not all right, by initial packet buffer, until being capable of handling just Beginning message, if it is possible to handle initial message, then follow the steps S5;
S5, initial message data processing is carried out;
S6, scene is extracted from the processed initial message data of output, and scene protection table is written;
After receiving non-initial message, start the flow chart of data processing of non-initial message, comprising:
A1, the link index sequence Maintenance Table with non-initial message;
A2, judge whether the sequence number of non-initial message is equal to or bigger by 1 than the sequence number in Maintenance Table, if it is, A3 is executed, Otherwise give up non-initial message;
A3, the sequence number of current non-initial message is write into sequence number Maintenance Table, and renewal sequence number;
A4, sequence number and link number index scene protection table with current non-initial message, and read field data;
Whether A5, inquiry are capable of handling non-initial message, if it is then step A6 is executed, if not all right, by non-initial report Text caching, until can handle non-initial message, then executes A6,
A6, non-initial message and field data are subjected to data processing together;
A7, " scene " is extracted from treated data, and scene protection table is written;
A8, judge whether non-initial message is to complete message, if so, thening follow the steps A9, otherwise, execute step A10;
A9, the link number of link write-back is managed into table, carries out link number recycling;
A10, return, and wait new message to be received.
2. the design method of more scene protections as described in claim 1, which is characterized in that the management method of link number management table Include:
The structure of storage link number management table is the caching FIFO of first in first out;
Initialization link number management table: being sequentially written in FIFO for data 1 to n, and wherein n is caching depth;
When receiving initial message application hyphen, then from link number management table, link number is taken out;
The link number of application is distributed to initial message, midamble and the end message of file;
When receiving the end message for carrying link number, the link number of end message is extracted, and be written to FIFO again, completes chain The number of connecing recycling.
3. the design method of more scene protections as described in claim 1, which is characterized in that sequence number Maintenance Table includes:
The sequence number of the fragment message of the same link number of input is that continuously, the current sequence number for receiving message is in sequence number It is in Maintenance Table or bigger than the sequence number of current sequence number Maintenance Table by 1;In the case of receiving sequence of message number in Maintenance Table, then Maintenance Table is constant;Sequence of message number is bigger than Maintenance Table sequence number by 1, the last one sequence number of sequence number Maintenance Table is given up.
4. the design method of more scene protections as described in claim 1, which is characterized in that the structure packet of sequence number Maintenance Table It includes: link number, each corresponding sequence number of link number.
5. the design method of more scene protections as described in claim 1, which is characterized in that sequence number Maintenance Table working mechanism packet Include: after receiving initial message, the link number using distribution is to index the sequence number Maintenance Table being written to using RAM as framework;It receives Midamble and completion message, read a sequence number for maintenance with link number for indexed addressing RAM;If the midamble received It is then continued in sequence number Maintenance Table with the sequence number for completing message;If the midamble received and completion message Sequence number+1 then continues in sequence number Maintenance Table, while the sequence number for receiving current slice message is written to sequence Row number Maintenance Table;If the midamble received is dropped with completion message because sequence number is unsatisfactory for condition, sequence number Maintenance Table is constant.
6. the design method of more scene protections as described in claim 1, which is characterized in that if scene protection includes: current The data or operation result of fragment message before same file are used in the processing of fragment message, then are delayed to fragment message before It deposits, after current slice message arrives, extracts the data or operation result of prime fragment message, handled.
7. the design method of more scene protections as described in claim 1, which is characterized in that scene protection working mechanism includes:
After receiving initial message, according to the link number of acquisition and Serial No. index write-in scene, and is handled, will be handled As a result it is synchronously written into same scene;
In-situ FTIR spectroelectrochemitry includes: to receive non-initial message, and parsing link number is read with sequence number and with call number and restoring scene, The scene of reading and the non-initial message received are handled;
Locale Holding include: receive non-initial message, if the sequence number of non-initial message in sequence number Maintenance Table, no Need Locale Holding;If the sequence number+1 of non-initial message in sequence number Maintenance Table, need by scene and treated knot Fruit synchronizes the scene of being saved in, and oldest scene is dropped;If the sequence number of the non-initial message received is not tieed up in sequence number It protects in table, while non-initial message+1 also not in sequence number Maintenance Table, then the message received is invalid packet.
CN201910040657.3A 2019-01-16 2019-01-16 Multi-site protection design method Active CN109981481B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910040657.3A CN109981481B (en) 2019-01-16 2019-01-16 Multi-site protection design method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910040657.3A CN109981481B (en) 2019-01-16 2019-01-16 Multi-site protection design method

Publications (2)

Publication Number Publication Date
CN109981481A true CN109981481A (en) 2019-07-05
CN109981481B CN109981481B (en) 2022-11-11

Family

ID=67076666

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910040657.3A Active CN109981481B (en) 2019-01-16 2019-01-16 Multi-site protection design method

Country Status (1)

Country Link
CN (1) CN109981481B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101291194A (en) * 2008-05-20 2008-10-22 华为技术有限公司 Method and system for keeping sequence of report
CN103986647A (en) * 2014-05-21 2014-08-13 大唐移动通信设备有限公司 Message transmission method and device
WO2016145764A1 (en) * 2015-03-18 2016-09-22 中兴通讯股份有限公司 Method and apparatus for packet loss control
WO2017000836A1 (en) * 2015-06-29 2017-01-05 华为技术有限公司 Message transmission method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101291194A (en) * 2008-05-20 2008-10-22 华为技术有限公司 Method and system for keeping sequence of report
CN103986647A (en) * 2014-05-21 2014-08-13 大唐移动通信设备有限公司 Message transmission method and device
WO2016145764A1 (en) * 2015-03-18 2016-09-22 中兴通讯股份有限公司 Method and apparatus for packet loss control
WO2017000836A1 (en) * 2015-06-29 2017-01-05 华为技术有限公司 Message transmission method and device

Also Published As

Publication number Publication date
CN109981481B (en) 2022-11-11

Similar Documents

Publication Publication Date Title
US9444737B2 (en) Packet data processor in a communications processor architecture
US9485200B2 (en) Network switch with external buffering via looparound path
US8248945B1 (en) System and method for Ethernet per priority pause packet flow control buffering
US6414961B1 (en) ATM switching with virtual circuit FIFO buffers
CN102014067A (en) Message fragment sending method, device and network equipment
CN109246036A (en) A kind of method and apparatus handling fragment message
CA3094300C (en) Message cache management in a mesh network
KR20110025118A (en) Receiver for error-protected packet-based frame
CN108829613A (en) Date storage method and storage equipment
US8990422B1 (en) TCP segmentation offload (TSO) using a hybrid approach of manipulating memory pointers and actual packet data
US8824468B2 (en) System and method for parsing frames
CN110535827A (en) Method and system for realizing TCP (Transmission control protocol) full-unloading IP (Internet protocol) core of multi-connection management
CN107508828B (en) A kind of very-long-range data interaction system and method
CN103685060A (en) Data packet sending method and data packet sending device
CN101803303A (en) Method and system for managing transmission of fragmented data packets
CN105357148A (en) Method and system for preventing output message of network exchange chip from being disordered
CN109981481A (en) A kind of design method of more scene protections
CN101150525A (en) Release method, system and logic module for buffered address
CN112565105A (en) Method, device and equipment for reducing time-sensitive frame forwarding time delay
CN102790663A (en) Full-hardware network interface applied to very long baseline interferometry (VLBI) hardware related processor
CN108259483A (en) On-line monitoring equipment communication data packet generic reception method
CN110865891B (en) Asynchronous message arrangement method and device
CN105450543B (en) Voice data transmission method
CN107592361A (en) A kind of data transmission method based on double IB networks, device, equipment
CN105263023A (en) Network code stream real-time receiving method based on high-speed decoding platform

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
CB03 Change of inventor or designer information

Inventor after: Chen Junlai

Inventor after: Zhang Wei

Inventor after: Ma Xuezhen

Inventor after: Sun Haiyang

Inventor before: Chen Junlai

CB03 Change of inventor or designer information
GR01 Patent grant
GR01 Patent grant