CN103384225A - Flow control method based on input intersection cache fast packet switched network - Google Patents
Flow control method based on input intersection cache fast packet switched network Download PDFInfo
- Publication number
- CN103384225A CN103384225A CN2013100841657A CN201310084165A CN103384225A CN 103384225 A CN103384225 A CN 103384225A CN 2013100841657 A CN2013100841657 A CN 2013100841657A CN 201310084165 A CN201310084165 A CN 201310084165A CN 103384225 A CN103384225 A CN 103384225A
- Authority
- CN
- China
- Prior art keywords
- output
- input
- buffer
- crosspoint
- voq
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000013256 coordination polymer Substances 0.000 claims abstract description 29
- 230000005540 biological transmission Effects 0.000 claims abstract description 12
- 230000015654 memory Effects 0.000 claims description 49
- 230000004075 alteration Effects 0.000 claims description 17
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000011017 operating method Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 102100031476 Cytochrome P450 1A1 Human genes 0.000 description 3
- 101000941690 Homo sapiens Cytochrome P450 1A1 Proteins 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Provided is a flow control method based on an input intersection cache fast packet switched network. An operation method comprises the steps that input data packets of each input end are written in corresponding virtual output queue VOQ in real time according to addresses of a destination end, an output scheduler module of the input end reads data packets from the VOQ in real time according to scheduling disciplines and writes the data packets in an intersection cache CP of an interchange switch, then an output scheduler module of the output end reads and sends data packets in the CP corresponding to the output end in real time according to the scheduling disciplines, when the data packets of the cache of the CP change, the CP sends status update information to the corresponding input end and the corresponding output end in real time, the output end and the input end update stored status values of themselves in real time respectively according to the status update information, and meanwhile the output scheduler module of the input end and the output scheduler module of the output end carry out scheduling on the data packets respectively according to the updated status values to complete transmission of the corresponding data packets.
Description
Technical field
The present invention relates to a kind of flow control methods based on inputting crosspoint buffer memory fast packet switching network, belong to the technical field of fast packet switching.
Background technology
The legacy packets switching technology was succeeded in developing at late 1960s to the beginning of the seventies.Transmission medium used was mainly copper cash (comprising coaxial cable and twisted-pair cable) at that time, and transmission rate is low, and transmission error rates is high, thereby the bottleneck of network performance is transmission link, rather than switch.
Through the development of many decades, the main transmission medium of link has become optical fiber, and its transmission rate is up to several Gbits, and transmission error rates is low to 10
-9Below.The bottleneck of network performance is no longer transmission link, but the switch of node.This just orders about people and goes to study the fast packet switching technology.
The basic ideas of fast packet switching are to simplify the protocol processes of packet switch, push complexity to network terminal system.The concrete technical measures that realize fast packet switching have following several:
(1) cancel error control and the flow control of link layer, solve the mistake of few generation by error control end to end.
(2) adopt hardware to realize forwarding of packets and parallel processing technique.
(3) adopt the little grouping of regular length.Because shorten the time delay that block length is conducive to reduce forwarding of packets, the even now processing can cause the increase of the packet count that arrives in the unit interval, still, and its advantage or main.The grouping of employing regular length is conducive to the hardware designs of packet switch, is convenient to arrange the port input from different to divide into groups, and carries out parallel the exchange.
The structure of the switching network that fast packet switching adopts and the setting of packet buffer are indivisible.The main purpose that buffer is set is to solve the output collision problem.The whole switching system of fast packet switching is comprised of input, output and alteration switch.
Form based on the switch network architecture of crossbar switching matrix crossbar and be divided into according to arranging mainly of buffer: input-buffer IQ(Input Queued), output buffer memory OQ(Output Queued), input into/output from cache CIOQ(Combined Input and Output Queued), crosspoint buffer memory CQ(Crosspoint Queued) and input crosspoint buffer memory CICQ(Combined Input and Crosspoint Queued) etc.Wherein, the input crosspoint buffer memory switching network based on crossbar is prior art.Flow control methods of the present invention proposes on the architecture basics based on the input crosspoint buffer memory switching network of crossbar.
Referring to Fig. 1, the structure of introducing a kind of input crosspoint buffer memory switching network based on crossbar forms: this switching network all is provided with buffer in the crosspoint of input and alteration switch.Be provided with N output virtual queue at each input, corresponding different outputs, be used for the packet that buffer memory mails to corresponding output end respectively.Each output virtual queue VOQ and each crosspoint buffer CP of input are one to one.For example, input 1 receives the packet that mails to output 1, and it is cached in output virtual queue VOQ11, by output scheduling module 1, this packet is write the crosspoint buffer CP11 corresponding with VOQ11.Again by the output scheduling module in output 1 with the packet reading and sending in CP11.
Input crosspoint buffer memory switching network based on crossbar shown in Figure 1 is also application scenarios or the basis of flow control methods of the present invention.The structure of following this packet switching network of brief description forms, and it is provided with: a plurality of inputs, a plurality of output and alteration switch.Wherein:
N input is used for completing the buffer memory of input packet and the output scheduling of packet.Each input is provided with: N output virtual queue VOQ is used for buffer memory input packet.An output scheduling module, be used for to receive the state information of the crosspoint buffer that alteration switch sends, and complete the packet output scheduling of input output virtual queue according to the state information of the buffer status information of the output virtual queue of input and crosspoint buffer.
An alteration switch is used for completing buffer memory and the exchange of packet.Comprising: the crosspoint buffer, be used for data cached bag, and monitor the seizure condition of each crosspoint buffer, send respectively crosspoint buffer status lastest imformation to input and output; The switching network connecting line is used for connecting respectively crosspoint buffer and input and output.
N output, the crosspoint buffer status information that sends according to switching network is completed the output scheduling of packet, read data packet from the buffer of crosspoint, and outwards forward.It is corresponding one by one with crosspoint buffer in alteration switch that each output is provided with N output virtual queue, and each packet of exporting in virtual queue of output is written into the crosspoint buffer of correspondence in alteration switch by the output scheduling module.For example, the crosspoint buffer in the alteration switch that the output virtual queue VOQ11 in input 1 is corresponding is CP11.
Because packet is all isometric, therefore, the duration of following four times equates: namely input time, input of receiving a packet to the crosspoint buffer write the time of a packet, time and the output that output outwards sends a packet reads the time of a packet from the crosspoint buffer, four numerical value are identical, and are called as a time slot.Input can only write a packet by the output scheduling module to the crosspoint buffer at a time slot, and similarly, output also can only read a packet from the crosspoint buffer memory at a time slot.
The below introduces the current situation of the flow control of fast packet switching:
The purpose of the flow control of fast packet switching is that attempt reduces, even avoids lost data packets in exchange process.The flow control methods that adopts at present is normally based on the switching network of input into/output from cache, adopt flow control methods end to end, namely carry out the data packet dispatching of input and output by transmit flow control information between input and output, thereby reduce or avoid lost data packets in exchange process.
At present, the flow control methods of prior art mainly contains two kinds:
First method (shown in Figure 2): output is according to the state independent decision-making of output buffer, to the control information of input transmitted traffic, when the data packet number in the output formation surpasses certain threshold value of setting, close or reduce the flow control information of data flow traffic to the output transmitted traffic, after input is received flow control information, close or reduce the corresponding data flow of flow control information indication.On the contrary, when another threshold value that the data packet number in the output formation is set, open or increase the flow control information of data flow traffic to the output transmitted traffic, after input is received flow control information, open or increase the corresponding data flow of flow control information indication.
Second method (shown in Figure 3): be a kind of flow control methods based on trusting, input sends buffer status information to output, output is controlled token according to the buffer status of input and output to the input transmitted traffic, and input is obtaining could to send packet after token.
Current flow control basically all is based on the input into/output from cache switching network and proposes corresponding flow control methods.Up to now, still find no the flow control methods that is exclusively used in based on inputting crosspoint buffer memory switching network.
Summary of the invention
In view of this, the objective of the invention is existing based on inputting on the buffer memory fast packet switching network foundation of crosspoint, a kind of flow control methods based on inputting crosspoint buffer memory fast packet switching network is provided, and the method can reduce even to be avoided in the fast packet switching process because the congested data-bag lost that causes.
In order to achieve the above object, the invention provides a kind of flow control methods based on inputting crosspoint buffer memory fast packet switching network, it is characterized in that: described method comprises following operating procedure:
(1) the input packet of each input is write corresponding output virtual queue VOQ in real time according to the destination interface address, waits for and being read by the output scheduling module of this input;
(2) the output scheduling module of input is dispatched the packet in output virtual queue VOQ according to scheduling rule, and from VOQ read data packet in real time, then write crosspoint buffer in alteration switch;
(3) packet in the output scheduling module of output reading and sending real-time according to the scheduling rule crosspoint buffer corresponding with it;
(4) data packet number of crosspoint buffer Real-Time Monitoring self buffer memory, in case find that this data packet number changes, input and the output to correspondence sends respectively corresponding buffer status lastest imformation immediately;
(5) output and input receive respectively the buffer status lastest imformation that the crosspoint buffer sends in real time, and the state value self stored corresponding with this buffer status information is updated to the in-cache status value in the buffer status lastest imformation that receives;
(6) the output scheduling module of input and output is dispatched packet according to the state value of self storing after each self refresh respectively, completes the transmission of corresponding data bag.
Current flow control methods is all to propose on the basis of input into/output from cache switching network, and the basis of flow control methods of the present invention is to input crosspoint buffer memory fast packet switching network, is a kind of flow control methods of innovation.The present invention is based on the current flow control methods advantage based on inputting crosspoint buffer memory fast packet switching network of generally using is: operating procedure is simple, practical, can effectively reduce the Loss Rate of packet, improve the utilance of crosspoint buffer in switching network, be conducive to popularize.
Description of drawings
Fig. 1 is the application scenarios of the inventive method: the structure of inputting crosspoint buffer memory fast packet switching network forms and the method for operation schematic diagram.
Fig. 2 is that the first is based on the flow control methods schematic diagram of the switching network of input into/output from cache.
Fig. 3 is that the second is based on the flow control methods schematic diagram of the switching network of input into/output from cache.
Fig. 4 (A), (B) are two buffer status memory schematic diagrames of input in network of the present invention.
Fig. 5 is output buffer status memory schematic diagram in network of the present invention.
Fig. 6 is the operating procedure flow chart of flow control methods of the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with accompanying drawing.
Referring to Fig. 1 and Fig. 6, introduce the following concrete operation step that the present invention is based on the flow control methods of inputting crosspoint buffer memory fast packet switching network:
Step 2, the output scheduling module of input is dispatched the packet in output virtual queue VOQ according to scheduling rule, and from VOQ read data packet in real time, then write crosspoint buffer in alteration switch.This step comprises following content of operation:
(21) the output scheduling module of each input i is provided with two memories: one is output virtual queue status register (shown in Fig. 4 (A)), is used for each VOQ of this input of storage
ijState value, i.e. its buffer memory h that has taken
jAnd the cache resources total amount H of each output virtual queue buffer
jKnown; Another is input crosspoint buffer status memory (shown in Fig. 4 (B)), is used for storage each crosspoint buffer CP corresponding with this input
ijState value, i.e. its buffer memory q that has taken
j, and each crosspoint buffer CP
ijCache resources total amount Q
jKnown; Each VOQ in input i
ijThe occupation proportion of buffer
Each CP corresponding with input i
ijOccupation proportion
In formula, natural number i is the sequence number of input, and natural number j is VOQ in input i and the sequence number of the crosspoint buffer CP corresponding with this VOQ, and the maximum of i and j is N.
(22) output scheduling module is dispatched packet according to state value and the configuration scheduling rule of two memory stores in step (21), and it is write crosspoint buffer CP in alteration switch
ij
The scheduling rule of this step is: at first the output scheduling module of input i checks its input crosspoint buffer status memory, and judges whether each CP is filled with;
If CP
ijFull, i.e. its corresponding state value q
jEqual its cache resources total amount Q
j, the output scheduling module of input i reads VOQ with regard to suspending
ijMiddle packet;
If CP
ijBe discontented with, i.e. its corresponding state value q
jLess than its cache resources total amount Q
j, the output scheduling module of input i just continues to read VOQ
ijMiddle packet;
Then, the occupation proportion η of each VOQ buffer of output scheduling module check of input i
j, never be suspended in the VOQ that reads the VOQ that selects the occupation proportion maximum, read data packet therefrom, and write crosspoint buffer in alteration switch.
Step 3, the packet in the output scheduling module of output reading and sending real-time according to the scheduling rule crosspoint buffer corresponding with it.This step comprises following content of operation: because each output k is provided with output crosspoint buffer status memory (shown in Figure 5), and, be used for each crosspoint buffer CP corresponding to storage this output k
bkState value, i.e. its buffer memory m that has taken
bAnd each crosspoint buffer CP
bkCache resources total amount M
bKnown, therefore each crosspoint buffer CP
bkOccupation proportion
In formula, natural number k and b are respectively the sequence numbers of output sequence number and this output k corresponding crosspoint buffer, and the maximum of k and b is N; Then, the packet reading and sending in the CP of the output scheduling module of this output k selection occupation proportion maximum.
Step 4, the data packet number of crosspoint buffer Real-Time Monitoring self buffer memory, in case find that this data packet number changes, input and the output to correspondence sends respectively corresponding buffer status lastest imformation immediately.
This step comprises following content of operation: the N * N that arranges in alteration switch crosspoint buffer separately respectively the data packet number of Real-Time Monitoring self buffer memory whether change, i.e. the operating position of himself cache resources.In case find that quantity changes, this crosspoint buffer sends respectively the state updating information of its buffer immediately to its corresponding input and output; If do not find number change, just do not send state updating information.
Step 5, output and input receive respectively the buffer status lastest imformation that the crosspoint buffer sends in real time, and the state value self stored corresponding with this buffer status information is updated to the in-cache status value in the buffer status lastest imformation that receives.
Block diagram referring to Fig. 1 lower-left side: the state updating information of crosspoint buffer comprises two fields: crosspoint buffer sequence number and in-cache status value thereof.Wherein, crosspoint buffer sequence number is the unique identification of each crosspoint buffer, distinguishes this state updating information from which crosspoint buffer for input and output, and the in-cache status value is the occupied buffer memory of this crosspoint buffer.
Step 6, the output scheduling module of input and output are dispatched packet according to the state value of self storing after each self refresh respectively, complete the transmission of corresponding data bag.
The present invention has carried out repeatedly implementing test, and the result of experiment is successfully, has realized goal of the invention.
Claims (6)
1. flow control methods based on input crosspoint buffer memory fast packet switching network, it is characterized in that: described method comprises following operating procedure:
(1) the input packet of each input is write corresponding output virtual queue VOQ in real time according to the destination interface address, waits for and being read by the output scheduling module of this input;
(2) the output scheduling module of input is dispatched the packet in output virtual queue VOQ according to scheduling rule, and from VOQ read data packet in real time, then write crosspoint buffer in alteration switch;
(3) packet in the output scheduling module of output reading and sending real-time according to the scheduling rule crosspoint buffer corresponding with it;
(4) data packet number of crosspoint buffer Real-Time Monitoring self buffer memory, in case find that this data packet number changes, input and the output to correspondence sends respectively corresponding buffer status lastest imformation immediately;
(5) output and input receive respectively the buffer status lastest imformation that the crosspoint buffer sends in real time, and the state value self stored corresponding with this buffer status information is updated to the in-cache status value in the buffer status lastest imformation that receives;
(6) the output scheduling module of input and output is dispatched packet according to the state value of self storing after each self refresh respectively, completes the transmission of corresponding data bag.
2. method according to claim 1, it is characterized in that: described step (2) comprises following content of operation:
(21) the output scheduling module of each input i is provided with two memories: one is output virtual queue status register, is used for each VOQ of this input of storage
ijState value, i.e. its buffer memory h that has taken
jAnd the cache resources total amount H of each output virtual queue buffer
jKnown; Another is input crosspoint buffer status memory, is used for storage each crosspoint buffer CP corresponding with this input
ijState value, i.e. its buffer memory q that has taken
j, and each crosspoint buffer CP
ijCache resources total amount Q
jKnown; Each VOQ in input i
ijThe occupation proportion of buffer
Each CP corresponding with input i
ijOccupation proportion
In formula, natural number i is the sequence number of input, and natural number j is VOQ in input i and the sequence number of the crosspoint buffer CP corresponding with this VOQ, and the maximum of i and j is N;
(22) output scheduling module is dispatched packet according to state value and the configuration scheduling rule of two memory stores in step (21), and it is write crosspoint buffer CP in alteration switch
ij
3. method according to claim 1, it is characterized in that: the scheduling rule in described step (22) is: at first the output scheduling module of input i checks its input crosspoint buffer status memory, and judges whether each CP is filled with;
If CP
ijFull, i.e. its corresponding state value q
jEqual its cache resources total amount Q
j, the output scheduling module of input i reads VOQ with regard to suspending
ijMiddle packet;
If CP
ijBe discontented with, i.e. its corresponding state value q
jLess than its cache resources total amount Q
j, the output scheduling module of input i just continues to read VOQ
ijMiddle packet;
Then, the occupation proportion η of each VOQ buffer of output scheduling module check of input i
j, never be suspended in the VOQ that reads the VOQ that selects the occupation proportion maximum, read data packet therefrom, and write crosspoint buffer in alteration switch.
4. method according to claim 1, it is characterized in that: described step (3) comprises following content of operation: because each output k is provided with output crosspoint buffer status memory, be used for each crosspoint buffer CP corresponding to storage this output k
bkState value, i.e. its buffer memory m that has taken
bAnd each crosspoint buffer CP
bkCache resources total amount M
bKnown, therefore each crosspoint buffer CP
bkOccupation proportion
In formula, natural number k and b are respectively the sequence numbers of output sequence number and this output k corresponding crosspoint buffer, and the maximum of k and b is N; Then, the packet reading and sending in the CP of the output scheduling module of this output k selection occupation proportion maximum.
(4) data packet number of crosspoint buffer Real-Time Monitoring self buffer memory, in case find that this data packet number changes, input and the output to correspondence sends respectively corresponding buffer status lastest imformation immediately.
5. method according to claim 1, it is characterized in that: described step (4) comprises following content of operation: the N * N that arranges in alteration switch crosspoint buffer separately respectively the data packet number of Real-Time Monitoring self buffer memory whether change, in case find that quantity changes, this crosspoint buffer just sends respectively the state updating information of its buffer immediately to its corresponding input and output; If do not find number change, just do not send state updating information.
6. method according to claim 1, it is characterized in that: the state updating information of described crosspoint buffer comprises two fields: crosspoint buffer sequence number and in-cache status value thereof; Wherein, crosspoint buffer sequence number is the unique identification of each crosspoint buffer, distinguishes this state updating information from which crosspoint buffer for input and output, and the in-cache status value is the occupied buffer memory of this crosspoint buffer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310084165.7A CN103384225B (en) | 2013-03-15 | 2013-03-15 | Based on the flow control methods of input intersection cache fast packet switching network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310084165.7A CN103384225B (en) | 2013-03-15 | 2013-03-15 | Based on the flow control methods of input intersection cache fast packet switching network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103384225A true CN103384225A (en) | 2013-11-06 |
| CN103384225B CN103384225B (en) | 2016-04-27 |
Family
ID=49491923
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310084165.7A Expired - Fee Related CN103384225B (en) | 2013-03-15 | 2013-03-15 | Based on the flow control methods of input intersection cache fast packet switching network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103384225B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104717160A (en) * | 2013-12-17 | 2015-06-17 | 华为技术有限公司 | Interchanger and scheduling algorithm |
| CN106453137A (en) * | 2016-09-20 | 2017-02-22 | 北京航空航天大学 | Dynamic multicast enqueue algorithm for relieving HOL Blocking |
| CN106453134A (en) * | 2016-10-31 | 2017-02-22 | 北京航空航天大学 | Packet scheduling algorithm of CICQ structure switches based on virtual queue length coordination single multicast competition |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1859279A (en) * | 2005-08-02 | 2006-11-08 | 华为技术有限公司 | Method for dispatching variable length data packet queue in crossbar switching matrix |
| US20110170558A1 (en) * | 2009-12-10 | 2011-07-14 | Shunyuan Ye | Scheduling, including distributed scheduling, for a buffered crossbar switch |
| CN102970249A (en) * | 2012-12-25 | 2013-03-13 | 武汉烽火网络有限责任公司 | Routing switching device and method |
-
2013
- 2013-03-15 CN CN201310084165.7A patent/CN103384225B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1859279A (en) * | 2005-08-02 | 2006-11-08 | 华为技术有限公司 | Method for dispatching variable length data packet queue in crossbar switching matrix |
| US20110170558A1 (en) * | 2009-12-10 | 2011-07-14 | Shunyuan Ye | Scheduling, including distributed scheduling, for a buffered crossbar switch |
| CN102970249A (en) * | 2012-12-25 | 2013-03-13 | 武汉烽火网络有限责任公司 | Routing switching device and method |
Non-Patent Citations (3)
| Title |
|---|
| KENJI YOSHIGOE等: "《An Evolution to Crossbar Switches with Virtual Output Queuing and Buffered Cross Points》", 《IEEE NETWORK》 * |
| 扈红超 等: "一种公平服务的动态轮询调度算法", 《软件学报》 * |
| 王斌 等: "一种基于PRIRR-PRIRR算法的CICQ交换机", 《南京邮电大学学报（自然科学版）》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104717160A (en) * | 2013-12-17 | 2015-06-17 | 华为技术有限公司 | Interchanger and scheduling algorithm |
| CN106453137A (en) * | 2016-09-20 | 2017-02-22 | 北京航空航天大学 | Dynamic multicast enqueue algorithm for relieving HOL Blocking |
| CN106453137B (en) * | 2016-09-20 | 2019-04-05 | 北京航空航天大学 | It is a kind of to be joined the team method using alleviating HOL Blocking as the dynamic multicast of target |
| CN106453134A (en) * | 2016-10-31 | 2017-02-22 | 北京航空航天大学 | Packet scheduling algorithm of CICQ structure switches based on virtual queue length coordination single multicast competition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103384225B (en) | 2016-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101873253A (en) | Buffered crossbar switch system | |
| CN106537858B (en) | A kind of method and apparatus of queue management | |
| CN103955436B (en) | A kind of data processing equipment and terminal | |
| CN110109872B (en) | Unified storage management device for remote sensing satellite heterogeneous data | |
| CN102185833A (en) | Fiber channel (FC) input/output (I/O) parallel processing method based on field programmable gate array (FPGA) | |
| CN108366111A (en) | A kind of data packet low time delay buffer storage and method for switching equipment | |
| CN110401558A (en) | Safety level multi-point communication network equipment, system and safety level network communication method | |
| CN209149287U (en) | Big data operation acceleration system | |
| CN109716311A (en) | For executing the systems, devices and methods of distributed arbitration program | |
| CN109861931A (en) | A kind of storage redundant system of high speed Ethernet exchange chip | |
| CN103384225B (en) | Based on the flow control methods of input intersection cache fast packet switching network | |
| CN100421420C (en) | Method for dispatching variable length data packet queue in crossbar switching matrix | |
| CN104486258B (en) | A kind of switched circuit based on interchange channel | |
| CN106027397B (en) | A kind of distributed measurement device network communication method of star-like expansion | |
| CN104407992B (en) | A kind of four port stores based on dual ported register array | |
| CN101247663B (en) | Considerable routing system and its forwarding table generation method | |
| CN106059927B (en) | A kind of automatic reconstructed network routing device of the limited cascade of hub-and-spoke configuration and its network | |
| KR102384685B1 (en) | Centralized scheduling apparatus and method considering non-uniform traffic | |
| CN110336756A (en) | Crossbar flow back-pressure control method with trunking function | |
| CN106372013B (en) | Long-distance inner access method, device and system | |
| CN108462653A (en) | The quick sending method of Protocol Control frame based on TTE | |
| CN105871761A (en) | High order matrix switch, network on chip and communication method | |
| CN103414660A (en) | Node device of sensing network capable of supporting sensing access and optical packet transmission | |
| CN105912273B (en) | A kind of message shares the FPGA implementation method of storage management | |
| TW202236104A (en) | Message communication between integrated computing devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160427 |