当前位置:网站首页>Hongke share | FPGA implementation of pass through and store and forward switching delay
Hongke share | FPGA implementation of pass through and store and forward switching delay
2022-07-29 06:40:00 【Hongke Industrial Communication Division】
In this article , We will show two kinds of available in FPGA Implemented on the COTS IEC 62439-3 Switch IP Comparison of nuclear delay . The first is to mix pass through switching and storage - Forwarding switching architecture , The second is based only on storage - Forwarding switching technology .
Reliable technology ——HSR&PRP
Now ,HSR/PRP And other reliable Ethernet networks are being recognized by many industrial automation applications . A strong evidence of this evolution is that the International Electrotechnical Commission has adopted seamless redundancy based on high reliability (HSR) Ethernet protocol and parallel redundancy protocol for Substation Automation (PRP) (IEC 62439-3 The first 5 and 4 strip ). Both protocols provide zero handover delay , Do not lose frames in case of failure , And be regarded as in the 2 Layer is a powerful means of network supervision .
chart :HSR Unicast traffic ring configuration example
HSR Frames are incompatible with traditional Ethernet infrastructure , and PRP Then it is allowed to send repeated frames through two traditional Ethernet Networks . therefore ,PRP The application field of , Although it is not specifically for “ real time ” Ethernet environment design .
“ real time ” It means responding to the signal within a predictable time after it occurs . for example , Modern digital control loops need to be lower than 10µs Reaction time . The latest Ethernet based control protocols such as EtherCAT or Sercos III It is often based on hardware to achieve predictable synchronization behavior and extremely low delay time .
- HSR Designed to meet Process Bus Set strict communication requirements .HSR Connect the intelligent electronic devices in each compartment (IED) interconnection .
- PRP Apply to Station and Inter-Bay Buses. Due to the flexibility of the agreement , It can connect many heterogeneous devices .
In order to maintain redundancy in communication ,PRP and HSR Interconnection between networks is performed using redundant gateways . Every HSR The link uses two gateway devices to connect to each PRP LAN. therefore , Avoid potential “ A single point of failure ” problem .

chart : adopt HSR and PRP Substation network communication
Pass through and store and forward
Pass through and store and forward L2 Switches are based on the destination of packets MAC Address makes forwarding decisions . The main difference between them is :
- The store and forward switch makes a decision after receiving the entire packet .
- The pass through switch is analyzing the target MAC Make a forwarding decision after the address , The address is in the first part of the frame .
In the store and forward switch , The delay time includes the time required to receive the whole frame . therefore , Compared with direct exchange , Longer delay .
Forwarding delay time
stay Xilinx FPGA Hongke on HSR-PRP IP These two methods are implemented in the kernel ( A hybrid pass through and store and forward , The other is pure store and forward ), give the result as follows :
therefore , It can be said that Hongke HSR-PRP IP The nuclear implementation is designed for PRP and HSR Exchange architecture of protocol design .
The theoretical minimum delay time is calculated by considering the forced field of Ethernet frame , This means that these fields must be analyzed to make exchange decisions . under these circumstances , In pass through , Time is independent of frame length , So it will be constant .
In the case of pure store and forward method , You need to store the whole frame before starting retransmission , So the delay depends on the frame length . It can be seen that , It is one order of magnitude larger than the optimized direct switching architecture .
Conclusion
- The analysis shows that , The custom architecture that combines pass through and store and forward methods can provide the best delay time in any case .
- FPGA Its role in these new agreements is crucial . One side , They allow low latency 、 Flexible and scalable solutions to meet the stringent requirements set in these Standards . On the other hand , When industrial manufacturers combine new agreements and specific agreements to provide equipment for the market ,FPGA It can reduce the time to market and risk .
If you want more information , Welcome private message 、 Comment or contact [email protected]
边栏推荐
- day09_ Static & Final & code block & abstract class & Interface & internal class
- Multiple IO usage
- Raw advanced socket experiment
- Right value reference and mobile construction
- Several misunderstandings about DDoS
- day14_ Unit test & Date common class & String common class
- Webshell管理工具的流量特征
- DDoS详解
- FIR filter design (1) -- using the FDATool toolbox of MATLAB to design FIR filter parameters
- Complex floating point multiplication of vivado IP core floating point
猜你喜欢
随机推荐
网络安全学习(二)
APP受到的那些漏洞攻击
Plugin location in mavan
如何判断业务被DDoS攻击?又会造成哪些危害?
Traffic characteristics of webshell management tools
day09_ Static & Final & code block & abstract class & Interface & internal class
What is the lifecycle of automated testing?
day14_ Unit test & Date common class & String common class
Circular linked list and bidirectional linked list
Sequence list and linked list
Advanced socket programming (options and control information)
Those vulnerability attacks on app
Is it OK to directly compare the size of two numbers in FPGA?
Complex floating point division of vivado IP core floating point
Idea practical shortcut key novice must see
day17_ Under collection
On defect description style
3、 Wide area communication network
什么是DNS放大攻击
What is DNS amplification attack


![[leetcode skimming] array 2 - binary search](/img/50/c006cbe5a91774c99eb782d9203fa0.png)




