CN107483352A - Data fast-forwarding structure and method in super fusion network device - Google Patents
Data fast-forwarding structure and method in super fusion network device Download PDFInfo
- Publication number
- CN107483352A CN107483352A CN201710727244.3A CN201710727244A CN107483352A CN 107483352 A CN107483352 A CN 107483352A CN 201710727244 A CN201710727244 A CN 201710727244A CN 107483352 A CN107483352 A CN 107483352A
- Authority
- CN
- China
- Prior art keywords
- data
- channel
- forwarding
- exchange chip
- network device
- 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.)
- Pending
Links
- 230000004927 fusion Effects 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000012545 processing Methods 0.000 claims description 3
- RTZKZFJDLAIYFH-UHFFFAOYSA-N Diethyl ether Chemical compound CCOCC RTZKZFJDLAIYFH-UHFFFAOYSA-N 0.000 claims 4
- 239000004744 fabric Substances 0.000 claims 1
- 238000011161 development Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000009897 systematic effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000011664 signaling Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/10—Packet switching elements characterised by the switching fabric construction
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Bus Control (AREA)
Abstract
Data fast-forwarding structure and method in a kind of super fusion network device of the present invention, the structure include：Processor, exchange chip, Ethernet interface, the first data forwarding channel and the second forwarding channel, the processor is communicated by the first data forwarding channel with exchange chip, and described exchange chip is communicated by the second data forwarding channel with Ethernet interface；First data forwarding channel includes PCIe buses and ethernet channel；Second data forwarding channel includes PCIe buses and ethernet channel；The present invention improves the forward efficiency of cloud computing center network level, to realize that super fusion network device truly provides basis.
Description
Technical field
The present invention relates to network technique field, and in particular to data fast-forwarding structure and side in super fusion network device
Method.
Background technology
With the rapid development at the fused data center of new generation such as cloud computing, by calculating, network, storage and server virtual
The super UNE framework that facility is integrated based on the resources such as change, as shown in Figure 1, it has also become the focus in industry, and
Current major cloud manufacturer mainly follows the theory of promoter Nutanix " calculating+storage ", and neglects network portion, and network
Data forwarding channel in equipment is by the speed and efficiency of extreme influence data forwarding.
The fusion of " calculating+storage+server virtualization " commercial solution existing in the industry, but to build completely
Super fusion in meaning should also be to consider the overall plan that " network " merges, base of the network portion as fused data center
Plinth, there can be substantial amounts of internal data interaction in supercomputing, distributed storage, virtual server and scheduling process, and it is big
The internal data interaction of amount is then related to forwarding rate and efficiency of the network equipment to data.
In the prior art, legacy network switching equipment forwarding channel, as shown in Fig. 2 control and management plane use low speed
The single channel forwarding structure of rate bus, is scheduled the forwarding for completing datum plane to exchange chip, for example, nx mouths and mx mouths it
Between data forwarding, its interchanger operation inter-process is simple, limited performance, and is easily asked in repeating process by single channel
Topic causes internal congestion, causes data-forwarding rate and effect using this kind of scheme in the UNE of new generation such as cloud service
Rate is substantially insufficient, it is impossible to realizes the data fast dispatch forwarding of network level, causes super UNE to develop into bottleneck.
The content of the invention
For above technical problem, the present invention proposes data fast-forwarding structure and side in a kind of super fusion network device
Method, processor concurrently carry out data surface and data plane signaling by PCIe (PCI-Express) buses and ethernet channel
Forwarding, improve the efficiency of forwarding.
Data fast-forwarding structure in a kind of super fusion network device, including：Processor, exchange chip, Ethernet exchanging
Interface, the first data forwarding channel and the second forwarding channel, the processor pass through the first data forwarding channel and exchange chip
Communication, described exchange chip are communicated by the second data forwarding channel with Ethernet interface；First data forwarding
Passage includes PCIe buses and ethernet channel；Second data forwarding channel includes PCIe buses and ethernet channel.
Preferably, the processor is X 86 processor, and the ethernet channel is 10Gbps*n to 100Gbps*n band
Width, the PCIe buses use 10Gbps to 20Gbps speed standard；N is number of active lanes, for the integer more than or equal to 1.
Preferably, the PCIe buses are communicated by direct memory access DMA read-write operations with exchange chip.
Preferably, it is described to be communicated by direct memory access DMA read-write operations with exchange chip including DMA with packet shape
Formula communicates with exchange chip, it is not necessary to the direct participation of the processor.
Preferably, ethernet channel and the PCIe bus parallel sets and communicated with exchange chip.
Data fast forwarding method in a kind of super fusion network device, including：
Processor selects the first data forwarding channel and control instruction is sent into exchange chip；
Exchange chip is sent data to corresponding Ethernet exchanging by the second data forwarding channel according to control instruction
Interface；
First data forwarding channel includes PCIe buses and ethernet channel；
Second data forwarding channel includes PCIe buses and ethernet channel.
Preferably, the processor is X 86 processor, and the ethernet channel is 10Gbps*n to 100Gbps*n band
Width, the PCIe buses use 10Gbps to 20Gbps speed standard；N is number of active lanes, for the integer more than or equal to 1.
Preferably, the PCIe buses are communicated by direct memory access DMA read-write operations with exchange chip.
Preferably, it is described to be communicated by direct memory access DMA read-write operations with exchange chip including DMA with packet shape
Formula communicates with exchange chip, it is not necessary to the direct participation of the processor.
Preferably, the ethernet channel of the first data forwarding channel and PCIe bus parallels are set, and the second data forwarding is led to
The ethernet channel and PCIe bus parallels in road are set.Compared with prior art, the present invention overcomes list using binary channels design
The congestion problems that passage is brought, while can be control signaling and data selector channel, the payload amount of passage is improved, is made described
The systematic function of the network equipment is significantly lifted.
Brief description of the drawings
Fig. 1 is super fusion basic network architectures schematic diagram；
Fig. 2 is conventional switch forwarding channel structural representation；
Fig. 3 is data fast-forwarding structure preferred embodiment structural representation in the super fusion network device of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that institute
The embodiment of description is only part of the embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention,
The every other embodiment that those of ordinary skill in the art are obtained under the premise of creative work is not made, belongs to this hair
The scope of bright protection.
Preferably, data forwarding is carried out between processor and exchange chip of the invention, for improving cloud computing center network
The forward efficiency of aspect, to realize that a kind of super fusion network device truly provides basis.It is as follows by illustrating：
Data fast-forwarding structure in a kind of super fusion network device, as shown in figure 3, including：Processor, exchange chip,
Ethernet interface, the first data forwarding channel and the second data forwarding channel, the processor pass through the first data forwarding
Passage is communicated with exchange chip, and described exchange chip is communicated by the second data forwarding channel with Ethernet interface.
Processor of the present invention can use 16,32,64 bit processors.It is preferred that the processor is X86 processing
Device.
Preferably, described data forwarding channel includes PCIe buses and ethernet channel；Ethernet channel, generally
(n is number of active lanes to 10Gbps*n to 100Gbps*n bandwidth, for the integer more than or equal to 1, according to technology development
It is fixed), directly communicated with X 86 processor, dispatched by X 86 processor, so as to realize the number of data plane and control plane
According to forwarding；PCIe buses, 10Gbps to 20Gbps speed standard (depending on PCIe technical standards) can be used, with X86
Processor is communicated.
It is preferred that in structures and methods of the present invention, the processor is X 86 processor, and the ethernet channel is
10Gbps*n to 100Gbps*n bandwidth, the PCIe buses use 10Gbps to 20Gbps speed standard.Through testing repeatedly
Know, the configuration causes processor to be matched in maximum efficiency in transmission rate with ethernet channel and PCIe buses.
Preferably, the PCIe buses are communicated by direct memory access DMA read-write operations with exchange chip.
Further, it is described to be communicated by direct memory access DMA read-write operations with exchange chip including DMA with packet
Form communicates with exchange chip, it is not necessary to the direct participation of the processor, can effectively reduce CPU occupancy, greatly improve
The throughput of data, makes the systematic function of the network equipment significantly be lifted.
Preferably, communicated with exchange chip ethernet channel and the PCIe bus parallel.
Data fast forwarding method in a kind of super fusion network device, including：
Processor selects the first data forwarding channel and control instruction is sent into exchange chip；
Exchange chip is sent data to corresponding Ethernet exchanging by the second data forwarding channel according to control instruction
Interface；
First data forwarding channel includes PCIe buses and ethernet channel；
Second data forwarding channel includes PCIe buses and ethernet channel.
Preferably, described data forwarding channel includes PCIe buses and ethernet channel；Ethernet channel, generally
(n is number of active lanes to 10Gbps*n or 100Gbps*n bandwidth, for the integer more than or equal to 1, according to technology development
It is fixed), directly communicated with X 86 processor, dispatched by X 86 processor, so as to realize the number of data plane and control plane
According to forwarding；PCIe buses, the general speed standard (depending on PCIe technical standards) for using 10Gbps or 20Gbps, with X86
Processor is communicated.
Preferably, the PCIe buses are communicated by direct memory access DMA read-write operations with exchange chip.
Preferably, it is described to be communicated by direct memory access DMA read-write operations with exchange chip including DMA with packet shape
Formula communicates with exchange chip, it is not necessary to the direct participation of the processor, can effectively reduce CPU occupancy, substantially increase
The throughput of data, the systematic function of the network equipment is set significantly to be lifted.
Preferably, communicated with exchange chip ethernet channel and the PCIe bus parallel；Wherein, ethernet channel mould
In block, generally 10Gbps*n or 100Gbps*n bandwidth (depending on technology development), directly enter with X 86 processor
Row communication, is dispatched by X 86 processor, so as to realize the data forwarding of data plane and control plane；In PCIe bus modules,
The general speed standard (depending on PCIe technical standards) for using 10Gbps or 20Gbps, is communicated with X 86 processor.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
To instruct the hardware of correlation to complete by program, the program can be stored in a computer-readable recording medium, storage
Medium can include：ROM, RAM, disk or CD etc..
The object, technical solutions and advantages of the present invention are further described above example of the present invention, institute
It should be understood that embodiment provided above or embodiment are only the preferred embodiment of the present invention, not limiting
System is of the invention, within the spirit and principles of the invention any modification, equivalent substitution and improvements made for the present invention etc.,
It should be included within protection scope of the present invention.
Claims (10)
- A kind of 1. data fast-forwarding structure in super fusion network device, it is characterised in that including：Processor, exchange chip, with Too net Fabric Interface, the first data forwarding channel and the second data forwarding channel, the processor are led to by the first data forwarding Road is communicated with exchange chip, and the exchange chip is communicated by the second data forwarding channel with Ethernet interface；Described One data forwarding channel includes PCIe buses and ethernet channel；Second data forwarding channel includes PCIe buses and ether Netcom road.
- 2. data fast-forwarding structure in super fusion network device according to claim 1, it is characterised in that the processing Device is X 86 processor, and the ethernet channel is 10Gbps*n to 100Gbps*n bandwidth, and the PCIe buses use 10Gbps to 20Gbps speed standard；N is number of active lanes, for the integer more than or equal to 1.
- 3. data fast-forwarding structure in super fusion network device according to claim 1, it is characterised in that the PCIe Bus is communicated by direct memory access DMA read-write operations with exchange chip.
- 4. data fast-forwarding structure in super fusion network device according to claim 3, it is characterised in that described to pass through Direct memory access DMA read-write operations communicate with exchange chip to be communicated including DMA with data packet form with exchange chip, it is not necessary to The direct participation of the processor.
- 5. data fast-forwarding structure in super fusion network device according to claim 1, it is characterised in that the ether Netcom road and PCIe bus parallels are set to communicate with exchange chip.
- A kind of 6. data fast forwarding method in super fusion network device, it is characterised in that including：Processor selects the first data forwarding channel and control instruction is sent into exchange chip；Exchange chip is sent data to corresponding Ethernet interface by the second data forwarding channel according to control instruction；First data forwarding channel includes PCIe buses and ethernet channel；Second data forwarding channel includes PCIe buses and ethernet channel.
- 7. data fast forwarding method in super fusion network device according to claim 6, it is characterised in that the processing Device is X 86 processor, and the ethernet channel is 10Gbps*n to 100Gbps*n bandwidth, and the PCIe buses use 10Gbps to 20Gbps speed standard；N is number of active lanes, for the integer more than or equal to 1.
- 8. data fast forwarding method in super fusion network device according to claim 6, it is characterised in that the PCIe Bus is communicated by direct memory access DMA read-write operations with exchange chip.
- 9. data fast forwarding method in super fusion network device according to claim 8, it is characterised in that described to pass through Direct memory access DMA read-write operations communicate with exchange chip to be communicated including DMA with data packet form with exchange chip, it is not necessary to The direct participation of the processor.
- 10. data fast forwarding method in super fusion network device according to claim 6, it is characterised in that the first number Set according to the ethernet channel and PCIe bus parallels of forwarding channel, the ethernet channel and PCIe of the second data forwarding channel are total Line is set parallel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710727244.3A CN107483352A (en) | 2017-08-23 | 2017-08-23 | Data fast-forwarding structure and method in super fusion network device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710727244.3A CN107483352A (en) | 2017-08-23 | 2017-08-23 | Data fast-forwarding structure and method in super fusion network device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107483352A true CN107483352A (en) | 2017-12-15 |
Family
ID=60602105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710727244.3A Pending CN107483352A (en) | 2017-08-23 | 2017-08-23 | Data fast-forwarding structure and method in super fusion network device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107483352A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131093A (en) * | 2019-12-16 | 2020-05-08 | 北京计算机技术及应用研究所 | Seven-layer switching system for three-wire interconnection of computing chip and switching chip |
| CN113966012A (en) * | 2021-11-05 | 2022-01-21 | 杭州安恒信息技术股份有限公司 | Access method and system of super-fusion storage system and computer storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101252537A (en) * | 2008-03-31 | 2008-08-27 | 杭州华三通信技术有限公司 | A switching network communication system, method and main control board |
| CN101277195A (en) * | 2007-03-30 | 2008-10-01 | 杭州华三通信技术有限公司 | A switching network communication system, implementation method and switching device |
| CN102739490A (en) * | 2012-06-26 | 2012-10-17 | 国电南瑞科技股份有限公司 | Multi-channel synchronous Ethernet expansion system based on peripheral component interface express (PCIe) bus |
| CN102769574A (en) * | 2012-08-06 | 2012-11-07 | 华为技术有限公司 | A device capable of hardware acceleration of business and its method |
| US20130268694A1 (en) * | 2012-04-06 | 2013-10-10 | International Business Machines Corporation | Pass-through converged network adaptor (cna) using existing ethernet switching device |
-
2017
- 2017-08-23 CN CN201710727244.3A patent/CN107483352A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101277195A (en) * | 2007-03-30 | 2008-10-01 | 杭州华三通信技术有限公司 | A switching network communication system, implementation method and switching device |
| CN101252537A (en) * | 2008-03-31 | 2008-08-27 | 杭州华三通信技术有限公司 | A switching network communication system, method and main control board |
| US20130268694A1 (en) * | 2012-04-06 | 2013-10-10 | International Business Machines Corporation | Pass-through converged network adaptor (cna) using existing ethernet switching device |
| CN102739490A (en) * | 2012-06-26 | 2012-10-17 | 国电南瑞科技股份有限公司 | Multi-channel synchronous Ethernet expansion system based on peripheral component interface express (PCIe) bus |
| CN102769574A (en) * | 2012-08-06 | 2012-11-07 | 华为技术有限公司 | A device capable of hardware acceleration of business and its method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131093A (en) * | 2019-12-16 | 2020-05-08 | 北京计算机技术及应用研究所 | Seven-layer switching system for three-wire interconnection of computing chip and switching chip |
| CN113966012A (en) * | 2021-11-05 | 2022-01-21 | 杭州安恒信息技术股份有限公司 | Access method and system of super-fusion storage system and computer storage medium |
| CN113966012B (en) * | 2021-11-05 | 2024-05-28 | 杭州安恒信息技术股份有限公司 | Access method and system of super-fusion storage system and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Software defined green data center network with exclusive routing | |
| US20220247696A1 (en) | Reliable transport offloaded to network devices | |
| KR101502610B1 (en) | 50 Gb/s ETHERNET USING SERIALIZER/DESERIALIZER LANES | |
| CN101547150B (en) | method and device for scheduling data communication input port | |
| CN103699447B (en) | Cloud computing-based transcoding and distribution system for video conference | |
| CN104780122B (en) | Control method based on the stratification network-on-chip router that caching is reallocated | |
| CN102185833B (en) | Fiber channel (FC) input/output (I/O) parallel processing method based on field programmable gate array (FPGA) | |
| CN103916909B (en) | Baseband Pool System | |
| CN104540234B (en) | A kind of associated task scheduling mechanism synchronously constrained based on CoMP under C RAN frameworks | |
| CN105871745A (en) | Mixed-service-based dispatching method in SDN network | |
| CN102307141B (en) | Message forwarding method and device | |
| CN108023840A (en) | OVS network traffics accelerate optimization method and OVS network traffics to accelerate optimization system | |
| CN108173782A (en) | The method, apparatus and storage medium of transmitting data stream in virtual private cloud | |
| CN103888374B (en) | Comprehensive sensor network service middle piece and service transmission achieving method thereof | |
| CN109729025A (en) | Method for processing data of flexible Ethernet and related equipment | |
| CN111683129A (en) | A dynamic ad hoc network mobile cloud computing system and method | |
| CN107483352A (en) | Data fast-forwarding structure and method in super fusion network device | |
| CN112148663A (en) | A data exchange chip and server | |
| CN101488923B (en) | Implementing method for network-on-chip data packet encoding optimization | |
| CN115622953A (en) | Non-real-time data transmission system and method for idle reserved bandwidth of software defined network | |
| CN104822175B (en) | A kind of code migration method and system suitable for cellular network | |
| CN107181702A (en) | It is a kind of to realize the device that RapidIO and Ethernet fusion are exchanged | |
| CN118381763A (en) | Congestion control method and device | |
| CN105517176B (en) | The method of dynamic dispatching virtualization base station resource | |
| CN105959161A (en) | High-speed data packet construction and distribution control method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171215 |