CN108494705A - A kind of network message high_speed stamping die and method - Google Patents
A kind of network message high_speed stamping die and method Download PDFInfo
- Publication number
- CN108494705A CN108494705A CN201810205530.8A CN201810205530A CN108494705A CN 108494705 A CN108494705 A CN 108494705A CN 201810205530 A CN201810205530 A CN 201810205530A CN 108494705 A CN108494705 A CN 108494705A
- Authority
- CN
- China
- Prior art keywords
- processor
- network
- network message
- core group
- function core
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/622—Queue service order
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/02—Network architectures or network communication protocols for network security for separating internal from external traffic, e.g. firewalls
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/14—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic
- H04L63/1441—Countermeasures against malicious traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/20—Network architectures or network communication protocols for network security for managing network security; network security policies in general
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention discloses a kind of network message high_speed stamping die and method, is related to network safety filed, and the framework being grouped using the multinuclear of Feiteng processor is run, multinuclear pipeline system in function core group parallel using between function core group, and efficient parallel handles network message;Multinuclear group parallel network data packet is conducive to core group and stretches according to network flow, improve multinuclear utilization ratio, while software development is relatively easy, is conducive to the smooth migration to next-generation processor platform;Multiple cores forward network packet with the formal layout of assembly line in each core group, and each core missions are single clear in assembly line, significantly improve cache hit rate, promote network throughput, improve network message processing speed.Apply the present invention in Network Security Device;The application field for having expanded domestic processor, the production domesticization for improving Network Security Device is horizontal, is of great importance at the autonomous controllable aspect of information security.
Description
Technical field
The present invention relates to network safety filed, specifically a kind of network message high_speed stamping die and method.
Background technology
With Network Traffic amount explosive growth, the demand to Network Security Device performance is also higher and higher.It is each
Manufacturer is constantly improved the hardware and software framework of Network Security Device and perfect, makes the treatability of Network Security Device
It can be continuously improved.At present in network safety filed, Network Security Device hardcore chip is all made of external product or solution party
Case.Network Security Device is the foundation stone of the network information security, and Network Security Device core devices cannot be domestic so that domestic network
Information security critical constraints, it is meant that information security is difficult to administer, and becomes the material risk of domestic network information security field.
In recent years, autonomous controllable, the safe and reliable homemade software and hardware of development is surrounded, some domestic enterprises have carried out positive
It explores, has emerged in large numbers a batch using multi-core processor of soaring as the host processor chip of representative.In hardware aspect, Network Security Device
The handling capacity of equipment generally improves using multi-core processor using the characteristic of multinuclear concurrent processing in manufacturer.
Network packet processing scheme can be divided into parallel (Run-to-Completion) scheme and stream on traditional multi-core processor
Waterline (Pipeline) scheme.
In parallel scheme, for the processing of network packet handler from packet receiving to the entire data packet life cycle given out a contract for a project, program is logical
It crosses core binding technology to be bundled on multiple cores, multiple cores are run parallel, and each core can receive, handle, transmission data packet, multinuclear
Parallel scheme operation is as shown in Figure 1.
Pipeline scheme has used for reference industrial pipelining, is multiple independent by the procedure decomposition of processing data packets
Stage, such as data packet, data packet pretreatment, Message processing, hair data packet are received, each stage is stand-alone program, is operated in
On different core, different core is connected by queue, handles data packet in sequence, the operation of multinuclear pipeline scheme is as schemed
Shown in 2.
Parallel scheme and pipeline scheme are compared as follows shown in table.
Domestic 2000 processor cores of soaring are more, and using the architecture design of multinuclear grouping, if using assembly line completely
Scheme, exploitation design difficulty are bigger;According to parallel scheme, each core program is complicated, and cache hit rate is low, if caching is not ordered
In, it needs data being loaded into caching from memory, this time can not ignore for express network flow chart of data processing.
Invention content
The present invention is directed to the demand and shortcoming of current technology development, provides a kind of based on domestic multi-core processor of soaring
Network message high_speed stamping die and method.
A kind of network message high_speed stamping die of the present invention solves the technical solution of above-mentioned technical problem use such as
Under:A kind of network message high_speed stamping die, the framework being grouped using the multinuclear of Feiteng processor, using between function core group
Parallel to run, multinuclear pipeline system in function core group, efficient parallel handles network message;
Its system architecture includes:
Data input network interface card end, including several network interface cards, are responsible for output network packet to processor;
Processor, using multi-core processor of soaring, the efficient process network message in a manner of adding assembly line parallel;
Data export network interface card end, including several network interface cards, are responsible for receiving the output data of processor.
Specifically, being grouped framework using the multinuclear for multi-core processor of soaring, multiple function core groups are divided into inside processor,
Include multiple cores inside each function core group;Each function core group is run parallel, concurrent processing network message;
The network message process flow of processor is divided into multiple subtasks, and each subtask program is tied to each respectively
In function core group in multiple cores, multiple cores are in the form of assembly line, sequential processes network message.
Specifically, the processor is using 2000 processors of soaring.
Specifically, described 2000 processor, 64 cores of soaring are divided into 16 function core groups;Each function core group includes 4
A core;Each function core group is run parallel, concurrent processing network packet.
Specifically, the data packet detection forwarding process of the processor is divided into 4 stages, respectively received data packet,
Data packet pretreatment, Packet Filtering, data packet are sent;The program in each stage is tied to each work(using binding technology respectively
In 4 cores of energy core group;Multiple cores are connected using queue in each function core group, are run in a pipeline fashion, suitable
Sequence handles data packet.
The invention also provides a kind of network message high speed processing method, the frame being grouped using the multinuclear of Feiteng processor
Structure is run parallel using between function core group, multinuclear pipeline system in function core group, and efficient parallel handles network message;
Specific implementation flow includes:
Step 1, data input network interface card end exports network packet to processor;
Step 2, soar multi-core processor by parallel plus assembly line in a manner of efficient process network message;
Step 3, processor gives data output network interface card end to output after network message high speed processing.
Specifically, the step 2,
It is grouped framework using the multinuclear for multi-core processor of soaring, multiple function core groups, Mei Gegong are divided into inside processor
Can include multiple cores inside core group;Each function core group is run parallel, concurrent processing network message;
The network message process flow of processor is divided into multiple subtasks, and each subtask program is tied to each respectively
In function core group in multiple cores, multiple cores are in the form of assembly line, sequential processes network message.
Specifically, the processor is using 2000 processors of soaring.
Specifically, described 2000 processor, 64 cores of soaring are divided into 16 function core groups;Each function core group includes 4
A core;Each function core group is run parallel, concurrent processing network packet.
Specifically, the data packet detection forwarding process of the processor is divided into 4 stages, respectively received data packet,
Data packet pretreatment, Packet Filtering, data packet are sent;The program in each stage is tied to each work(using binding technology respectively
In 4 cores of energy core group;Multiple cores are connected using queue in each function core group, are run in a pipeline fashion, suitable
Sequence handles data packet.
A kind of network message high_speed stamping die of the present invention and method, have the advantage that compared with prior art
It is:The present invention utilizes the multinuclear processing capacity of domestic processor, comprehensive traditional parallel scheme and pipeline scheme to realize parallel
Assembly line is added to carry out the purpose of high speed processing network message;The parallel design of multinuclear group, by core group resource pool, using core group as work(
Energy unit modularized design is conducive to core group and stretches according to network flow, improves multinuclear utilization ratio, while software development is relatively simple
It is single, it is conducive to the smooth migration to next-generation processor platform;
Multiple cores forward network packet with the formal layout of assembly line in each core group;Each core missions list in assembly line
One is clear, significantly improves cache hit rate, promotes network throughput;Secondly, in core group each core share the caching of this group with
High-speed bus improves intercore communication rate, reduces assembly line and submits time delay;
Therefore the present invention using the grouping of Feiteng processor multinuclear framework, using multinuclear flowing water in parallel between core group, core group
Design, the operation of multinuclear high-speed parallel, concurrent processing network packet improves network message processing speed, improves network and set
Standby handling capacity;Apply the present invention to Network Security Device, such as fire wall, intruding detection system, Unified Threat Management are set
In standby;The application field for having expanded domestic processor, the production domesticization for improving Network Security Device is horizontal, autonomous in information security
Controllable aspect is of great importance.
Description of the drawings
Illustrate the embodiment of the present invention or technology contents in the prior art in order to clearer, below to the embodiment of the present invention
Or required attached drawing does simple introduction in the prior art.It will be apparent that attached drawing disclosed below is only the one of the present invention
Section Example to those skilled in the art without creative efforts, can also be attached according to these
Figure obtains other attached drawings, but within protection scope of the present invention.
Attached drawing 1 is the schematic diagram of network packet parallel processing plan on traditional multi-core processor;
Attached drawing 2 is the schematic diagram of network packet stream waterline processing scheme on traditional multi-core processor;
Attached drawing 3 is the schematic diagram of 1 network message high_speed stamping die of embodiment.
Specific implementation mode
The technical issues of to make technical scheme of the present invention, solving and technique effect are more clearly understood, below in conjunction with tool
Body embodiment is checked technical scheme of the present invention, is completely described, it is clear that described embodiment is only this hair
Bright a part of the embodiment, instead of all the embodiments.Based on the embodiment of the present invention, those skilled in the art are not doing
All embodiments obtained under the premise of going out creative work, all within protection scope of the present invention.
Embodiment 1:
In order to realize the high speed of network processes on domestic multi-core processor, and development difficulty and scalability are taken into account,
The present embodiment proposes a kind of network message high_speed stamping die, and using the multinuclear processing capacity of Feiteng processor, synthesis is traditional
Parallel scheme and pipeline scheme realize purpose of the parallel plus assembly line to network message high speed processing, multinuclear high-speed parallel fortune
Row promotes network message processing speed, improves the handling capacity of the network equipment.
The present embodiment network message high_speed stamping die, as shown in Fig. 3, system architecture includes:
Data input network interface card end, including several network interface cards, are responsible for output network packet to processor;
Processor, using multi-core processor of soaring, the efficient process network message in a manner of adding assembly line parallel;
Specifically, being grouped framework using the multinuclear for multi-core processor of soaring, multiple function core groups are divided into inside processor,
Include multiple cores inside each function core group;Each function core group is run parallel, concurrent processing network message;
The network message process flow of processor is divided into multiple subtasks, and each subtask program is tied to each respectively
In function core group in multiple cores, multiple cores are in the form of assembly line, sequential processes network message.
Data export network interface card end, including several network interface cards, are responsible for receiving the output data of processor.
The present embodiment also proposed a kind of network message high speed processing method, the frame being grouped using the multinuclear of Feiteng processor
Structure, using parallel between function core group, multinuclear flowing water design in function core group, multiple function core group high-speed parallels operations promote net
Network message processing speed improves the handling capacity of the network equipment.
The present embodiment also proposes that a kind of network message high speed processing method, technical solution are high with 1 network message of embodiment
Fast processing system can be cross-referenced, and specific implementation flow includes:
Step 1, data input network interface card end exports network packet to processor;
Step 2, soar multi-core processor by parallel plus assembly line in a manner of efficient process network message;
Specifically, being grouped framework using the multinuclear for multi-core processor of soaring, multiple function core groups are divided into inside processor,
Include multiple cores inside each function core group;Each function core group is run parallel, concurrent processing network message;
The network message process flow of processor is divided into multiple subtasks, and each subtask program is tied to each respectively
In function core group in multiple cores, multiple cores are in the form of assembly line, sequential processes network message.
Step 3, processor gives data output network interface card end to output after network message high speed processing.
Embodiment 2:
A kind of network message high_speed stamping die that the present embodiment proposes is the another embodiment of the present invention,
On the basis of 1 network message high_speed stamping die of embodiment, 2000 processors of soaring may be used in processor, soar at 2000
8 cores are divided into 2 function core groups in reason device core group, and each function core group includes 4 cores;It soars 2000 processor 64
Core is divided into 16 function core groups.
The data packet detection forwarding process of the processor is divided into 4 stages, and respectively received data packet, data packet is pre-
Processing, Packet Filtering, data packet are sent;The program in each stage is tied to 4 of function core group using binding technology respectively
In core.
Multiple cores are connected using queue in each function core group, are run in a pipeline fashion, sequence processing data
Packet.Entire 2000 processor, 16 function core groups of soaring are run parallel, concurrent processing network packet.
The present embodiment also proposes that a kind of network message high speed processing method, technical solution are high with 2 network message of embodiment
Fast processing system can be cross-referenced, carrys out the technical solution and technical characterstic that the present invention will be described in detail with 2000 processors of soaring.
The step 2, multi-core processor of soaring is using 2000 processors of soaring, 8 cores in 2000 processor core groups of soaring
The heart is divided into 2 function core groups, and each function core group includes 4 cores;2000 processor, 64 cores of soaring are divided into 16
Function core group;Each function core group is run parallel, concurrent processing network packet.
The network message process flow of the processor is divided into 4 stages, and respectively received data packet, data packet is located in advance
Reason, Packet Filtering, data packet are sent;The program in each stage is tied to the 4 of each function core group using binding technology respectively
In a core;Multiple cores are connected using queue in each function core group, are run in a pipeline fashion, sequence processing data
Packet.
The present embodiment network message high speed processing method, using the framework for 2000 processor multinuclears grouping of soaring, using core
Parallel between group, the flow of data packet detection forwarding is divided into multiple subtasks, Ge Gehe by the design of multinuclear flowing water in core group
Multiple cores in group bind these subtask processes respectively, and network packet, Duo Gehe are forwarded with the formal layout of assembly line
Group is run in parallel mode, concurrent processing network packet;Multinuclear group is parallel, is designed by functional unitization of core group, is conducive to core
Group is flexible according to network flow, improves multinuclear utilization ratio;Each core missions are single clear in single core group assembly line, significantly
Cache hit rate is improved, network throughput is promoted.
Use above specific case elaborates the principle of the present invention and embodiment, these embodiments are
It is used to help understand core of the invention technology contents, the protection domain being not intended to restrict the invention, technical side of the invention
Case is not limited in above-mentioned specific implementation mode.Based on the above-mentioned specific embodiment of the present invention, those skilled in the art
Without departing from the principle of the present invention, any improvement and modification to made by the present invention should all fall into the special of the present invention
Sharp protection domain.
Claims (10)
1. a kind of network message high_speed stamping die, which is characterized in that the framework being grouped using the multinuclear of Feiteng processor is used
It is run parallel between function core group, multinuclear pipeline system in function core group, efficient parallel handles network message;
Its system architecture includes:
Data input network interface card end, including several network interface cards, are responsible for output network packet to processor;
Processor, using multi-core processor of soaring, the efficient process network message in a manner of adding assembly line parallel;
Data export network interface card end, including several network interface cards, are responsible for receiving the output data of processor.
2. a kind of network message high_speed stamping die according to claim 1, which is characterized in that divided inside the processor
For multiple function core groups, each function core group inside includes multiple cores;Each function core group is run parallel, concurrent processing network
Message;
The network message process flow of processor is divided into multiple subtasks, and each subtask program is tied to each function respectively
In core group in multiple cores, multiple cores are in the form of assembly line, sequential processes network message.
3. a kind of network message high_speed stamping die according to claim 2, which is characterized in that the processor is used and soared
2000 processors.
4. a kind of network message high_speed stamping die according to claim 3, which is characterized in that 2000 processors of soaring
64 cores are divided into 16 function core groups;Each function core group includes 4 cores;Each function core group is run parallel, concurrently
Network data packet.
5. a kind of network message high_speed stamping die according to claim 4, which is characterized in that the data packet of the processor
Detection forwarding process is divided into 4 stages, respectively received data packet, data packet pretreatment, Packet Filtering, data packet hair
It send;The program in each stage is tied to using binding technology respectively in 4 cores of each function core group;In each function core group
Multiple cores are connected using queue, are run in a pipeline fashion, sequence processing data packet.
6. a kind of network message high speed processing method, which is characterized in that the framework being grouped using the multinuclear of Feiteng processor is used
It is run parallel between function core group, multinuclear pipeline system in function core group, efficient parallel handles network message;
Specific implementation flow includes:
Step 1, data input network interface card end exports network packet to processor;
Step 2, soar multi-core processor by parallel plus assembly line in a manner of efficient process network message;
Step 3, processor gives data output network interface card end to output after network message high speed processing.
7. a kind of network message high speed processing method according to claim 6, which is characterized in that the step 2,
It is grouped framework using the multinuclear for multi-core processor of soaring, multiple function core groups, each function core are divided into inside processor
Internal group includes multiple cores;Each function core group is run parallel, concurrent processing network message;
The network message process flow of processor is divided into multiple subtasks, and each subtask program is tied to each function respectively
In core group in multiple cores, multiple cores are in the form of assembly line, sequential processes network message.
8. a kind of network message high speed processing method according to claim 7, which is characterized in that the processor is used and soared
2000 processors.
9. a kind of network message high speed processing method according to claim 8, which is characterized in that 2000 processors of soaring
64 cores are divided into 16 function core groups;Each function core group includes 4 cores;Each function core group is run parallel, concurrently
Network data packet.
10. a kind of network message high speed processing method according to claim 9, which is characterized in that the data of the processor
Packet detection forwarding process is divided into 4 stages, respectively received data packet, data packet pretreatment, Packet Filtering, data packet hair
It send;The program in each stage is tied to using binding technology respectively in 4 cores of each function core group;In each function core group
Multiple cores are connected using queue, are run in a pipeline fashion, sequence processing data packet.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810205530.8A CN108494705A (en) | 2018-03-13 | 2018-03-13 | A kind of network message high_speed stamping die and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810205530.8A CN108494705A (en) | 2018-03-13 | 2018-03-13 | A kind of network message high_speed stamping die and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108494705A true CN108494705A (en) | 2018-09-04 |
Family
ID=63339032
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810205530.8A Pending CN108494705A (en) | 2018-03-13 | 2018-03-13 | A kind of network message high_speed stamping die and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108494705A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110097268A (en) * | 2019-04-19 | 2019-08-06 | 北京金山安全软件有限公司 | Task allocation method and device, electronic equipment and storage medium |
CN115185878A (en) * | 2022-05-24 | 2022-10-14 | 中科驭数(北京)科技有限公司 | Multi-core packet network processor architecture and task scheduling method |
CN111949740B (en) * | 2019-05-15 | 2024-03-26 | 中国科学院声学研究所 | Parallel network packet classification method, system and network equipment based on multi-core processor |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102331923A (en) * | 2011-10-13 | 2012-01-25 | 西安电子科技大学 | Multi-core and multi-threading processor-based functional macropipeline implementing method |
CN103838552A (en) * | 2014-03-18 | 2014-06-04 | 北京邮电大学 | System and method for processing multi-core parallel assembly line signals of 4G broadband communication system |
CN104639460A (en) * | 2015-02-03 | 2015-05-20 | 重庆邮电大学 | High-speed network data packet parallel receiving method based on many-core processor |
CN105022671A (en) * | 2015-07-20 | 2015-11-04 | 中国科学院计算技术研究所 | Load balancing method for parallel processing of stream data |
CN105227547A (en) * | 2015-09-09 | 2016-01-06 | 重庆邮电大学 | A kind of flow media flux generation systems based on many core platforms |
-
2018
- 2018-03-13 CN CN201810205530.8A patent/CN108494705A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102331923A (en) * | 2011-10-13 | 2012-01-25 | 西安电子科技大学 | Multi-core and multi-threading processor-based functional macropipeline implementing method |
CN103838552A (en) * | 2014-03-18 | 2014-06-04 | 北京邮电大学 | System and method for processing multi-core parallel assembly line signals of 4G broadband communication system |
CN104639460A (en) * | 2015-02-03 | 2015-05-20 | 重庆邮电大学 | High-speed network data packet parallel receiving method based on many-core processor |
CN105022671A (en) * | 2015-07-20 | 2015-11-04 | 中国科学院计算技术研究所 | Load balancing method for parallel processing of stream data |
CN105227547A (en) * | 2015-09-09 | 2016-01-06 | 重庆邮电大学 | A kind of flow media flux generation systems based on many core platforms |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110097268A (en) * | 2019-04-19 | 2019-08-06 | 北京金山安全软件有限公司 | Task allocation method and device, electronic equipment and storage medium |
CN110097268B (en) * | 2019-04-19 | 2022-08-19 | 北京金山安全软件有限公司 | Task allocation method and device, electronic equipment and storage medium |
CN111949740B (en) * | 2019-05-15 | 2024-03-26 | 中国科学院声学研究所 | Parallel network packet classification method, system and network equipment based on multi-core processor |
CN115185878A (en) * | 2022-05-24 | 2022-10-14 | 中科驭数(北京)科技有限公司 | Multi-core packet network processor architecture and task scheduling method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3612942B1 (en) | Queue management for direct memory access | |
CN102710477B (en) | Data processing system based on VPX bus structure | |
JP6974270B2 (en) | Intelligent high bandwidth memory system and logic die for it | |
CN100449522C (en) | Matrix multiplication parallel computing system based on multi-FPGA | |
CN107392308A (en) | A kind of convolutional neural networks accelerated method and system based on programming device | |
DE102018126150A1 (en) | DEVICE, METHOD AND SYSTEMS FOR MULTICAST IN A CONFIGURABLE ROOM ACCELERATOR | |
CN108494705A (en) | A kind of network message high_speed stamping die and method | |
CN104820657A (en) | Inter-core communication method and parallel programming model based on embedded heterogeneous multi-core processor | |
CN106951926A (en) | The deep learning systems approach and device of a kind of mixed architecture | |
CN104699654B (en) | One kind is based on interconnection interconnection adaption system and method between CHI on-chip interconnection buss and QPI pieces | |
CN107102964A (en) | A kind of method that GPU cluster expansion is carried out using high-speed connector | |
EP1835414A3 (en) | Reduction processing method for parallel computer, and parallel computer | |
CN113312283B (en) | Heterogeneous diagram learning system based on FPGA acceleration | |
CN108595379A (en) | A kind of parallelization convolution algorithm method and system based on multi-level buffer | |
CN209149287U (en) | Big data operation acceleration system | |
CN103312627B (en) | Regular expression matching method based on two-level storage | |
CN110059797B (en) | Computing device and related product | |
WO2023040197A1 (en) | Cross-node communication method and apparatus, device, and readable storage medium | |
CN109472734A (en) | A kind of target detection network and its implementation based on FPGA | |
CN104850516B (en) | A kind of DDR Frequency Conversion Designs method and apparatus | |
CN111653317B (en) | Gene comparison acceleration device, method and system | |
CN107277062A (en) | The method for parallel processing and device of packet | |
CN107301143A (en) | A kind of asynchronous arbiter based on the phase Handshake Protocol of coding and double track four | |
CN107179895A (en) | A kind of method that application compound instruction accelerates instruction execution speed in data flow architecture | |
CN111078286A (en) | Data communication method, computing system and storage medium |
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: 20180904 |