Dual-redundancy FC network transmission system

文档序号:786595 发布日期:2021-04-09 浏览:18次 中文

阅读说明:本技术 一种双冗余fc网络传输系统 (Dual-redundancy FC network transmission system ) 是由 陈旭辉 卫铁锤 季培培 许永建 王永星 陆亚南 于 2020-12-14 设计创作,主要内容包括:一种双冗余FC网络传输系统,包括发送端和接收端;发送端包括两个缓冲区,分别与两个缓冲区串联的两个FC端口,用于将数据帧同时发送至两个缓冲区的发送端控制器;还包括当其中一个FC端口发生堵塞时用于接收该FC端口发出的堵塞信号的发送控制模块;接收端包括分别用于接收两个FC端口传输的数据并对SN的连续性进行检测和过滤的序号检查模块,与序号检查模块相串联的进行重复帧的剔除的冗余剔除模块,与冗余剔除模块相串联的对接收到的乱序帧重新排序并上报数据帧的乱序重排模块;本发明提出的双冗余FC网络传输系统,解决现有FC冗余网络方法中存在的丢帧、乱序等风险,对提高安全关键应用中FC网络的数据传输可靠性有明显的效果。(A dual-redundancy FC network transmission system comprises a sending end and a receiving end; the sending end comprises two buffer areas, two FC ports which are respectively connected with the two buffer areas in series and a sending end controller which is used for sending the data frame to the two buffer areas simultaneously; the system also comprises a sending control module used for receiving a blocking signal sent by one FC port when the FC port is blocked; the receiving end comprises a serial number checking module, a redundant eliminating module and a disorder rearranging module, wherein the serial number checking module is respectively used for receiving data transmitted by two FC ports and detecting and filtering the continuity of SN (serial number), the redundant eliminating module is connected with the serial number checking module in series and is used for eliminating repeated frames, and the disorder rearranging module is connected with the redundant eliminating module in series and is used for reordering the received disorder frames and reporting the data frames; the dual-redundancy FC network transmission system provided by the invention solves the risks of frame loss, disorder and the like in the conventional FC redundancy network method, and has an obvious effect of improving the data transmission reliability of the FC network in safety key application.)

1. A dual redundant FC network transmission system, characterized by: the system comprises a sending end and a receiving end;

the sending end comprises two buffer areas, two FC ports respectively connected with the two buffer areas in series and a sending end controller used for sending the data frames sent by the system to the two buffer areas simultaneously; the sending control module receives the blocking signal and sends a command that the sending end controller does not send data frames to the blocked FC port after the blocking signal is received by the sending control module;

the receiving end comprises a serial number checking module which is respectively used for receiving data transmitted by two FC ports and detecting and filtering the continuity of SN, a redundancy eliminating module which is connected with the serial number checking module in series and eliminates repeated frames, and a disorder rearranging module which is connected with the redundancy eliminating module in series and reorders the received disorder frames and reports the data frames.

2. The dual redundant FC network transport system of claim 1, wherein: the FC ports adopt a redundancy strategy of 'first receiving and then discarding' and carry out frame sequence identification by using SN (service number) based on S _ ID and D _ ID, thereby realizing that two FC ports simultaneously transmit and receive data.

3. The dual redundant FC network transport system of claim 1, wherein: the sending end controller sends data to the network where the two buffers are located at the same time, and controls the network with large transmission delay as a reference.

4. The dual redundant FC network transport system of claim 1, wherein: when one link is disconnected, emptying the buffer memory in the middle of the channel after the time of the link disconnection reaches LDTOV, otherwise, keeping the buffer memory content.

5. The dual redundant FC network transport system of claim 1, wherein: the function of the transmission control module is as follows:

(1) the sending end of the sending control module takes the SN port of the lagged FC port as a reference to synchronously send and control the two buffer areas, and the output SN of the two buffer areas is ensured to be the same;

(2) when the link is disconnected and exceeds the LDTOV and then is recovered, no packet is sent in the HDLDR, so that the cache data of the normal link is emptied; and the cache in the new reply link is full of data; avoid causing disorder; HDLDR is more than or equal to port quantity and maximum frame length transmission time and buffer zone depth under the lowest rate;

(3) synchronously writing data sent from a system into a buffer area, and carrying out back pressure on the system side under the condition of a lagging SN port;

(4) control generates the SN.

6. The dual redundant FC network transport system of claim 5, wherein: the method for controlling the generation of the SN comprises the following steps:

(1) ensuring that SN of two copied frames sent by two FC ports are the same;

(2) each terminal sending port independently controls SN generation aiming at each DID;

(3) starting from '0' after the system is reset, and in other cases, once a frame is sent, the system is '+' 1;

(4) for broadcast and multicast frames, each destination DID requires "+" 1; a device list of the whole network is required to be maintained at each node;

(5) the DID space is not less than 256.

7. The dual redundant FC network transport system of claim 1, wherein: the filtering rule of the sequence number checking module is as follows:

(1) the receiving range is: PSN '+' 1, PSN '+' 2, PSN, the previous normal received frame SN of the link, the frame in this range is sent to the redundancy eliminating module; meanwhile, the missing SN under the condition of discontinuous SN is recorded, wherein the SN received by any FC port and missing by any FC port is also sent to a redundancy eliminating module through a serial number checking module;

(2) when the receiving end is reset, SN is equal to 0 or the link disconnection time exceeds LDTOV, the IC checks and receives all the arriving frames;

(3) and discarding the frames which do not meet the conditions, and counting and reporting the frames.

8. The dual redundant FC network transport system of claim 1, wherein: the method for eliminating the repeated frames by the redundancy eliminating module comprises the following steps:

(1) each terminal node, except the node of the terminal node, establishes FIT for all SIDs on the network;

(2) adopting a removing strategy of 'receiving before and discarding after' that is, the ER module reserves the use of the data frame received firstly and discards the data frame received later;

(3) performing repeated redundant frame elimination by using FITS (fitt), wherein FIT mainly stores fields of SID (security identification) and exchange ID (OXID for short) and the fields are 5B; FIT depth is 512;

(4) comparing OXID with FIT of corresponding SID through frame filtered by sequence number checking module, if FIT has the information, discarding the frame; otherwise, storing the OXID information into the FIT;

5) and updating the FIT in a circulating covering mode, starting covering the FIT again after the FIT is completely written, and emptying the FIT after the terminal system is reset.

9. The dual redundant FC network transport system of claim 1, wherein: the sorting method of the disorder rearrangement module comprises the following steps:

(1) setting a buffer area for each SID, filling the buffer area with later received frames in sequence and then reporting;

(2) setting out-of-order rearrangement overtime time threshold, abandoning reordering when the overtime or the number of frames is exceeded, and reporting the data frames of the buffer area;

(3) setting a jump SN table and aging time for each SID, and clearing the table entry after receiving a data frame corresponding to SN or reaching the aging time;

(4) the reordering buffering time and the buffering frame number can be configured or closed, and the reordering function and the delay characteristic are considered.

10. The dual redundant FC network transport system of claim 1, wherein: after the fault network channel is recovered, the sending end controller actively controls the sending end not to send the data frame within a period of time T, and the buffer area in the other normal network channel transfers and empties the buffered data within the period of time T, so that the states of the two redundant networks are consistent when the two redundant networks are sent again.

Technical Field

The invention belongs to the technical field of redundant transmission of FC (fiber channel) optical fiber channels, and particularly relates to a dual-redundancy FC network transmission system.

Background

FC (fibre channel) networks are very popular for use in safety critical information transmission applications and in order to further increase the reliability of the network, dual redundant networks are generally used, i.e. signals are sent, transmitted and received simultaneously over the same two networks, and a signal can be transmitted successfully as long as one network is normal, see fig. 1 for details.

Then the following disadvantages exist in the existing FC redundancy method:

(1) there is a risk of misordering:

if the network A fails, the network B is normal, after a period of time, the network A recovers from failure, and the network B is always normal; there is data in the buffer at each position in the channel of the B network, and there is no data in the buffer of the A network, so the data transmitted through the A network will arrive at the receiving end earlier than the B network, which is easy to cause the receiving out of order.

(2) There is a risk of frame loss:

if the data transmission delay of the network A is larger than that of the network B and the difference of a plurality of frames exists, when the network B loses one frame of data, the existing redundancy method cannot complete the data after passing through the network A, and finally a frame loss is formed.

Disclosure of Invention

Aiming at the problems of disorder and frame loss in the existing method and the problem that the existing redundancy method cannot be used for high-reliability information transmission, the invention aims to provide a dual-redundancy FC network transmission system.

The invention aims to realize the following technical scheme, and provides a dual-redundancy FC network transmission system, which comprises a sending end and a receiving end; the sending end comprises two buffer areas, two FC ports respectively connected with the two buffer areas in series and a sending end controller used for sending the data frames sent by the system to the two buffer areas simultaneously; the sending control module receives the blocking signal and sends a command that the sending end controller does not send data frames to the blocked FC port after the blocking signal is received by the sending control module;

the receiving end comprises a serial number checking module which is respectively used for receiving data transmitted by two FC ports and detecting and filtering the continuity of SN, a redundancy eliminating module which is connected with the serial number checking module in series and eliminates repeated frames, and a disorder rearranging module which is connected with the redundancy eliminating module in series and reorders the received disorder frames and reports the data frames.

Preferably, the FC ports adopt a redundancy policy of "receive first and discard second", and perform frame sequence identification using SNs based on the S _ ID and the D _ ID, thereby implementing that two FC ports transmit and receive data simultaneously.

Preferably, the sending-end controller simultaneously sends data to the network where the two buffers are located, and controls the network with the large transmission delay as a reference.

Preferably, when one of the links is disconnected, the buffer in the middle of the channel is emptied after the time of the link disconnection reaches the LDTOV, otherwise, the buffer content is maintained.

Preferably, the transmission control module functions as follows:

(1) the sending end of the sending control module takes the SN port of the lagged FC port as a reference to synchronously send and control the two buffer areas, and the output SN of the two buffer areas is ensured to be the same;

(2) when the link is disconnected and exceeds the LDTOV and then is recovered, no packet is sent in the HDLDR, so that the cache data of the normal link is emptied; and the cache in the new reply link is full of data; avoid causing disorder; HDLDR is more than or equal to port quantity and maximum frame length transmission time and buffer zone depth under the lowest rate;

(3) synchronously writing data sent from a system into a buffer area, and carrying out back pressure on the system side under the condition of a lagging SN port;

(4) control generates the SN.

Preferably, the method for controlling the generation of SN comprises:

(1) ensuring that SN of two copied frames sent by two FC ports are the same;

(2) each terminal sending port independently controls SN generation aiming at each DID;

(3) starting from '0' after the system is reset, and in other cases, once a frame is sent, the system is '+' 1;

(4) for broadcast and multicast frames, each destination DID requires "+" 1; a device list of the whole network is required to be maintained at each node;

(5) the DID space is not less than 256.

Preferably, the filtering rule of the sequence number checking module is as follows:

(1) the receiving range is: PSN '+' 1, PSN '+' 2, PSN, the previous normal received frame SN of the link, the frame in this range is sent to the redundancy eliminating module; meanwhile, the missing SN under the condition of discontinuous SN is recorded, wherein the SN received by any FC port and missing by any FC port is also sent to a redundancy eliminating module through a serial number checking module;

(2) when the receiving end is reset, SN is equal to 0 or the link disconnection time exceeds LDTOV, the IC checks and receives all the arriving frames;

(3) and discarding the frames which do not meet the conditions, and counting and reporting the frames.

Preferably, the method for rejecting the repeated frames by the redundant rejection module comprises the following steps:

(1) each terminal node, except the node of the terminal node, establishes FIT for all SIDs on the network;

(2) adopting a removing strategy of 'receiving before and discarding after' that is, the ER module reserves the use of the data frame received firstly and discards the data frame received later;

(3) performing repeated redundant frame elimination by using FITS (fitt), wherein FIT mainly stores fields of SID (security identification) and exchange ID (OXID for short) and the fields are 5B; FIT depth is 512;

(4) comparing OXID with FIT of corresponding SID through frame filtered by sequence number checking module, if FIT has the information, discarding the frame; otherwise, storing the OXID information into the FIT;

5) and updating the FIT in a circulating covering mode, starting covering the FIT again after the FIT is completely written, and emptying the FIT after the terminal system is reset.

Preferably, the sorting method of the out-of-order rearrangement module is as follows:

(1) setting a buffer area for each SID, filling the buffer area with later received frames in sequence and then reporting;

(2) setting out-of-order rearrangement overtime time threshold, abandoning reordering when the overtime or the number of frames is exceeded, and reporting the data frames of the buffer area;

(3) setting a jump SN table and aging time for each SID, and clearing the table entry after receiving a data frame corresponding to SN or reaching the aging time;

(4) the reordering buffering time and the buffering frame number can be configured or closed, and the reordering function and the delay characteristic are considered.

Preferably, after the failed network channel is recovered, the sending-end controller actively controls not to send the data frame within a period of time T, and within the period of time T, the buffer area in the other normal network channel transfers and clears the buffered data, so as to ensure that the states of the two redundant networks are consistent when the data is sent again.

The invention has the following advantages:

1. the dual-redundancy FC network transmission system provided by the invention effectively solves the risks of frame loss, disorder and the like in the conventional FC redundancy network method, has an obvious effect of improving the data transmission reliability of the FC network in safety key application, can improve the quality of the whole network, and avoids various accidents possibly caused by data transmission errors.

The foregoing is a summary of the present invention, and for the purpose of making clear the technical means of the present invention, the present invention can be implemented according to the content of the description, and for the purpose of making the above and other objects, features, and advantages of the present invention more comprehensible, the following preferred embodiments are described in detail:

drawings

Fig. 1 is a schematic diagram of an application of a conventional redundant FC.

Fig. 2 is a block diagram of a transmitting end in the present embodiment.

Fig. 3 is a block diagram of a receiving end in the present embodiment.

Detailed Description

To further illustrate the technical means and effects of the present invention adopted to achieve the intended purpose, the following detailed description is given to a dual redundant FC network transmission system according to the present invention, with reference to the accompanying drawings and preferred embodiments, and the detailed description thereof.

The invention provides a dual-redundancy FC network transmission system which comprises a sending end and a receiving end.

Referring to fig. 2, the transmitting end includes a transmitting end controller, a first buffer area having an input end connected to an output end of the transmitting end controller, a first FC port having an input end connected to an output end of the first buffer area, a second buffer area having an input end connected to an output end of the transmitting end controller, a second FC port having an input end connected to an output end of the second buffer area, and output ends of the first FC port and the second FC port are both connected to the receiving end; the sending end further comprises a sending control module, the input end of the sending control module is respectively connected with the input end of the first FC port and the input end of the second FC port, and the output end of the sending control module is respectively connected with the input ends of the first buffer area, the second buffer area and the sending end controller.

The first buffer area or the second buffer area receives the data sent by the sending end controller and sends the data to the first FC port or the second FC port according to the control of the sending flow control module, meanwhile, the follow-up compatible double-channel independent function is facilitated, and the depth of the buffer area is designed according to the prior art.

Referring to fig. 3, the receiving end includes a redundancy elimination module, a disorder rearrangement module, and sequence number checking modules respectively located on two links; a first sequence number checking module and a second sequence number checking module are respectively designed on the two redundant channels, the input end of the first sequence number checking module is connected with the output end of the first F C port, the input end of the second sequence number checking module is connected with the output end of the second FC port, the output ends of the first sequence number checking module and the second sequence number checking module are both connected with the input end of the redundant eliminating module, and the input end of the disorder rearrangement module is connected with the output end of the redundant eliminating module.

The sending end controller sends the data frame sent by the system to a first buffer area on the A link and a second buffer area on the B link simultaneously; the first buffer area transmits the received data frame to a first sequence number checking module of the receiving end through a first FC port, and the second buffer area transmits the received data frame to a second sequence number checking module through a second FC port.

If the first FC port on the A link is blocked, the first FC port sends a signal that the port is blocked to the sending control module, the sending control module sends a command that the data frame is not sent to the first FC port to the first buffer when receiving the blocking signal, and the sending control module simultaneously sends a command that the sending end controller does not send the data frame to the first buffer when the first buffer is full to the sending end controller.

If the second FC port on the B link is transmitting a congestion, the second FC port sends a signal of the congestion to the transmission control module, and the transmission control module sends a command that the data frame is no longer transmitted to the second FC port to the second buffer when receiving the congestion signal, and the transmission control module simultaneously sends a command that the data frame is no longer transmitted to the second buffer by the transmission controller when the second buffer is full to the transmission controller.

The first FC port and the second FC port both adopt a redundancy strategy of 'first come receiving and then discard', and utilize SN (serial number for short hereinafter) based on S _ ID and D _ ID to carry out frame sequence identification, thereby realizing that the first FC port and the second FC port simultaneously transmit and receive data.

According to a general method, using the FC frame OXID field as SN, the available range: 0-65535, wherein 0 represents the first frame after the terminal equipment (module, board card) is reset, and SN circulates from 1-65535 to 1 in every frame under other conditions, and loops back to 1 after reaching 65535.

The first FC port and the second FC port need to realize the following functions in addition to the functions specified by the standard (described above).

(1) The number of the line is sent at a high speed after the link is disconnected and overtime, and then the RDY is returned, the overtime time (LDTOV for short) of the link disconnection is configurable, and the LDTOV can be calculated and tested according to the scale of the network, the network load and the number of nodes;

(2) under the condition that the link is disconnected and does not exceed the LDTOV, the FC port cannot cache clearly and does not send data, back pressure is continuously carried out, for example, error detection overtime time is overtime, error detection overtime time is calculated and tested according to the scale of the network, the network load and the number of nodes, namely, the error detection overtime time is shorter, the overtime time of the link is not as long as the overtime time of the link, the FC port which has the error detection overtime also cannot cache clearly and does not send data, and therefore the link can be recovered quickly.

The sending control module mainly realizes the following functions:

(1) a sending end of the sending control module takes an SN (serial number) port of a lagged FC port as a reference, namely, an FC port with slow sending as a reference, and synchronously sends control to the first buffer area and the second buffer area to ensure that the SN output by the first buffer area and the SN output by the second buffer area are the same;

(2) when the link is disconnected and exceeds the LDTOV and then recovered, pressing HDLDR for a period of time (namely keeping time after the disconnection is recovered) without sending packets, and emptying the cache data of the normal link; and the cache in the new reply link is full of data; avoid causing disorder; the HDLDR should be configurable, suggest the maximum frame length transmission time under the port number and the minimum rate and the buffer depth, and can be automatically matched according to the rate configuration;

(3) synchronously writing data sent from a system into a buffer zone, and carrying out back pressure on the system side under the condition of a lagging SN port, namely, a sending control module feeds back to a sending end controller and the buffer zone positioned on the offline link;

(4) the control method for generating SN comprises the following steps:

1) ensuring that SN of two copied frames sent by two FC ports are the same;

2) a transmission port of each FC port individually controls SN generation for each DID;

3) starting from '0' after the system is reset, and in other cases, once a frame is sent, the system is '+' 1;

4) for broadcast and multicast frames, each destination DID requires "+" 1; a device list of the whole network is required to be maintained at each node;

5) the DID space is not less than 256.

The first sequence number checking module and the second sequence number checking module mainly detect and filter SN continuity, and the main filtering rules are as follows:

(1) the reception range is typically: PSN '+' 1, PSN '+' 2, PSN (i.e. previous SN), the previous normal received frame SN of the link, the frame in this range is sent to the redundancy elimination module; meanwhile, the missing SN under the condition of discontinuous SN is recorded, the missing SN is also sent to the ER module when the A or B network receives the missing SN, namely the SN received by the FC port on the A or B network is also sent to the redundancy eliminating module through the serial number checking module;

(2) when the receiving end of the serial number checking module is reset, and SN is equal to 0 or the link disconnection time exceeds LDTOV, the IC (integrity checking) checks and receives all the arriving frames;

(3) and discarding the frames which do not meet the conditions, and counting and reporting the frames.

The redundant eliminating module mainly eliminates the repeated frames according to a strategy, and the main method is as follows:

(1) each terminal node, except its own node, establishes a frame information table (FIT for short) for all source addresses (SID for short) on the network;

(2) adopting a removing strategy of 'receiving before and discarding after' that is, the ER module reserves the use of the data frame received firstly and discards the data frame received later;

(3) adopting a frame information table strategy (FITS for short) to remove repeated redundant frames, wherein FIT mainly stores SID and OXID fields, and the total number is 5B; DPFIT (i.e., depth of FIT) 512;

(4) comparing OXID with FIT of corresponding SID through frame filtered by sequence number checking module, if FIT has the information, discarding the frame; otherwise, storing the OXID information into the FIT;

(5) and updating the FIT in a circulating covering mode, starting covering the FIT again after the FIT is completely written, and emptying the FIT after the terminal system is reset.

The disorder rearrangement module is mainly used for rearranging and reporting the data frames aiming at the received disorder frames, and the arranging method comprises the following steps:

(1) setting a buffer area for each SID, filling the buffer area with later received frames in sequence and then reporting;

(2) setting out-of-order rearrangement overtime time threshold, abandoning reordering when the overtime time or the number of frames is exceeded, and reporting the data frames of the buffer area, wherein the overtime time threshold can be set according to the requirements of system application;

(3) setting a jump SN table and aging time for each SID, and clearing the table entry after receiving a data frame corresponding to SN or reaching the aging time;

(4) the reordering buffering time and the buffering frame number can be configured or closed, and the reordering function and the delay characteristic are considered.

The network channel fault recovery sending control mechanism of the dual-redundancy FC network transmission method comprises the following steps:

after the fault network channel is recovered, the controller at the sending end actively controls not to send data frames within a period of time T, and a buffer area in another normal network channel transmits and clears the cache data within the period of time T; the time T can be flexibly configured according to the network condition; ensuring that the states of the two redundant networks of network A, B are consistent when re-transmitting; otherwise reception out of order will result.

The lag reference sending control mechanism of the dual-redundancy FC network transmission method comprises the following steps:

the sending end controller sends data to A, B two networks at the same time, and controls the networks with large transmission delay as a reference; otherwise, in the case of A, B network load imbalance, the link transmission delay difference does not converge.

Configurable lane intermediate buffer flush time: this time is defined as the link drop timeout (LDTOV for short). The link is disconnected due to the link problem, the buffer memory in the middle of the channel can be emptied after the time of the link disconnection reaches LDTOV, and otherwise, the buffer memory content is kept and the method is mainly used for quickly recovering the link. The method is different from the conventional method that the cache is emptied immediately after the error detection is overtime, and the link recovery time is prolonged.

Designing a disorder rearrangement mechanism, setting configurable disorder recovery time, and according to a 7-layer model of the network, the disorder recovery is generally realized by a transmission layer, and some standards of FC require that the sequence of data packets delivered by a link layer is the same as that of a transmitting end, so the disorder recovery mechanism is designed on the link layer; and for some time-sensitive applications, configurable out-of-order detection recovery time is designed, and out-of-order rearrangement and time-sensitive transmission applications are considered.

The above description is only a preferred embodiment of the present invention, and should not be taken as limiting the invention in any way, and any simple modification, equivalent change and modification made by those skilled in the art according to the technical spirit of the present invention are still within the technical scope of the present invention without departing from the technical scope of the present invention.

10页详细技术资料下载
上一篇:一种医用注射器针头装配设备
下一篇:一种基于千兆PON/光模块的硬件反馈信息适应调节系统

网友询问留言

已有0条留言

还没有人留言评论。精彩留言会获得点赞!

精彩留言,会给你点赞!