CN106897148A - A kind of system and method for generating micro-downburst - Google Patents
A kind of system and method for generating micro-downburst Download PDFInfo
- Publication number
- CN106897148A CN106897148A CN201710112528.1A CN201710112528A CN106897148A CN 106897148 A CN106897148 A CN 106897148A CN 201710112528 A CN201710112528 A CN 201710112528A CN 106897148 A CN106897148 A CN 106897148A
- Authority
- CN
- China
- Prior art keywords
- downburst
- knl
- micro
- node
- information
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of system and method for generating micro-downburst, for generating the static micro-downburst field data in big regional extent or the micro- down-wass field data in dynamic generation different range, and between making the node of KNL group systems, calculating in node reaches the load balancing of calculating, so as to greatest extent using the performance of KNL, the efficiency of the overall operation of raising system, greatly shorten the run time of program, accelerate the deployment of scenario environment, solve that the existing emulation mode for generating micro-downburst by CPU can only be generated to small range and formation efficiency is slow, the technical problem of long operational time.Present invention method includes：Storage system, express network, KNL group systems, control system；The express network is used to connect the storage nodes, the KNL cluster nodes and the control system node.
Description
Technical field
The present invention relates to high-performance computing sector, more particularly to a kind of system and method for generating micro-downburst.
Background technology
The use of various weapons is in real natural environment, with the development of virtual experiment technology, to virtual examination
The confidence level requirement more and more higher tested.Virtual natural environment is indispensable important set in nowadays popular virtual experiment technology
Into part.Virtual natural environment is namely based on virtual experiment technology, to the number of natural environmental elements in modeling and simulation system
Word is represented, including atmospheric environment, space electromagnetic environment, terrain environment.Current single, Utopian natural environment can not meet
Armament systems to the demand of environment, in the urgent need to the natural environment of integrated complex.And synthetic natural environment is generally required according to not
Individually developed with dummy experiment system, expend resource, and user can not add typical environment key element and quickly generate ring according to scenario
Border data.Therefore need according to scenario, integrated atmospheric environment, electromagnetic environment, terrain environment, typical adverse circumstances etc., fast fast-growing
Into virtual natural environment data, and data supporting is provided to dummy experiment system.As shown in figure 1, in a spatial dimension, collection
Into atmospheric environment, electromagnetic environment, three kinds of different types of environment of terrain environment；Typical environment is integrated with background environment,
The integrated different typical atmospheric environment such as on aerial vehicle trajectory route, integrated typical electromagnetic environment, in the integrated typical landform in ground
Environment etc., so checking armament systems under extreme weather by efficiency, and disturbance wind field often influences weapon aircraft
Normal flight activity, serious can also produce aircraft accident, and air turbulence and low are rather common in disturbance wind field.
The harm for causing that wherein low flies to aircraft is it is well known that there is document to list 20th century the eighties to nine
The aircraft accident that the ten's were caused for 28 times by low, and have in the total death toll for having counted and die from low latitude on about 40%
Wind shear.Wind speed vertical direction or horizontal direction cause with respect to change when low refers to short distance, and micro- undershoot
Air-flow is a kind of serious low.
The emulation mode of existing generation micro-downburst can only be generated to small range, and formation efficiency is slow, operation
Time is long.
The content of the invention
A kind of system and method for generating micro-downburst are the embodiment of the invention provides, for generating big regional extent
Micro- down-wass field data in interior static micro-downburst field data or dynamic generation different range, and make KNL group systems
Node between, the calculating in node reach the load balancing of calculating, so as to greatest extent using the performance of KNL, improve system
Overall operation efficiency, greatly shorten the run time of program, accelerate the deployment of scenario environment, solve existing passing through
The emulation mode of CPU generation micro-downbursts can only small range ground generation and formation efficiency is slow, the technology of long operational time is asked
Topic.
A kind of system for generating micro-downburst is the embodiment of the invention provides, including：Storage system, express network,
KNL group systems, control system；
The express network is used to connect the storage nodes, the KNL cluster nodes and the control
System node.
Preferably,
The storage system specifically includes shared storage subsystem and subsystem is locally stored；
The shared storage subsystem is used to store large-scale calculations data；
The subsystem that is locally stored is for storing the supplemental characteristic and calculation result data of the control system.
A kind of method for generating micro-downburst is the embodiment of the invention provides, including：
Each node in KNL group systems receives the first information that the control system transmission comes, the first information
Including collar vortex parameter setting, spatial dimension, resolution ratio, time range；
Each node in KNL group systems generates micro- lower qi of chong channel ascending adversely according to the first information and by collar vortex guidance model
Flow field data.
Preferably,
Each node in KNL group systems also includes before receiving the first information that the control system transmission comes：
The control system obtains the first information of generation micro-downburst place need and is distributed to the KNL clusters
Each node in system.
Preferably,
Each node in KNL group systems is received after the first information that the control system transmission comes, in KNL collection
Before each node in group's system generates micro-downburst field data according to the first information and by collar vortex guidance model,
Also include：
The spatial dimension in the first information carries out piecemeal treatment.
Preferably,
Each node in KNL group systems generates micro- undershoot according to the first information and by collar vortex guidance model
After air-flow field data, also include：
The micro-downburst field data asynchronous communication that each node in KNL group systems will be generated controls system to described
System.
Preferably,
Each node in KNL group systems generates micro- lower qi of chong channel ascending adversely according to the first information and by collar vortex guidance model
Flow field data are specially：
Each node in KNL group systems is according to the first information, the side by collar vortex guidance model and vectorization
Formula generates micro-downburst field data.
Preferably,
Each node in KNL group systems generates micro- lower qi of chong channel ascending adversely according to the first information and by collar vortex guidance model
Flow field data are specially：
If each node in KNL group systems opens main line corresponding with the spatial dimension processed by piecemeal
Number of passes, while generating micro-downburst field data by collar vortex guidance model according to the first information.
As can be seen from the above technical solutions, the embodiment of the present invention has advantages below：
A kind of system and method for generating micro-downburst are the embodiment of the invention provides, for generating big regional extent
Micro- down-wass field data in interior static micro-downburst field data or dynamic generation different range, and make KNL group systems
Node between, the calculating in node reach the load balancing of calculating, so as to greatest extent using the performance of KNL, improve system
Overall operation efficiency, greatly shorten the run time of program, accelerate the deployment of scenario environment, solve existing passing through
The emulation mode of CPU generation micro-downbursts can only small range ground generation and formation efficiency is slow, the technology of long operational time is asked
Topic.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
Other accompanying drawings are obtained with according to these accompanying drawings.
Fig. 1 is virtual natural environment schematic diagram；
Fig. 2 is a kind of topological diagram of system for generating micro-downburst provided in an embodiment of the present invention；
Fig. 3 is that a kind of flow of the first embodiment of method for generating micro-downburst provided in an embodiment of the present invention is shown
It is intended to；
Fig. 4 is that a kind of flow of the second embodiment of method for generating micro-downburst provided in an embodiment of the present invention is shown
It is intended to；
Fig. 5 is collar vortex guidance model signal in a kind of method for generating micro-downburst provided in an embodiment of the present invention
Figure.
Specific embodiment
A kind of system and method for generating micro-downburst are the embodiment of the invention provides, for generating big regional extent
Micro- down-wass field data in interior static micro-downburst field data or dynamic generation different range, and make KNL group systems
Node between, the calculating in node reach the load balancing of calculating, so as to greatest extent using the performance of KNL, improve system
Overall operation efficiency, greatly shorten the run time of program, accelerate the deployment of scenario environment, solve existing passing through
The emulation mode of CPU generation micro-downbursts can only small range ground generation and formation efficiency is slow, the technology of long operational time is asked
Topic.
To enable that goal of the invention of the invention, feature, advantage are more obvious and understandable, below in conjunction with the present invention
Accompanying drawing in embodiment, is clearly and completely described, it is clear that disclosed below to the technical scheme in the embodiment of the present invention
Embodiment be only a part of embodiment of the invention, and not all embodiment.Based on the embodiment in the present invention, this area
All other embodiment that those of ordinary skill is obtained under the premise of creative work is not made, belongs to protection of the present invention
Scope.
Fig. 2 is referred to, a kind of system for generating micro-downburst is the embodiment of the invention provides, including：Storage system,
Express network, KNL group systems 2, control system 1；
Express network is used to connect storage nodes, the node of KNL group systems 2 and the node of control system 1, realizes each
High-speed communication between individual node, so as to constitute the group system of centralization, the node control KNL group systems 2 of control system 1 are saved
Point, and United Dispatching is carried out to the node of KNL group systems 2, the unified distribution of data is realized, while the communications in the system
With the asynchronous execution of parallel computation, the overall operational efficiency of system is improve, meet the requirement of performance application；
Storage system specifically includes shared storage subsystem and subsystem is locally stored；
Shared storage subsystem is used to store large-scale calculations data；
Subsystem is locally stored for the supplemental characteristic and calculation result data of storage control system；
It should be noted that some KNL composition KNL group systems 2, storage system is included in KNL group systems 2 and control
In system processed 1, so storage system and node do not have being marked in figure；In KNL group systems 2, each KNL includes 72
Individual core, 16GMCDRAM memory headrooms, maximum supports the DDR4 internal memories of 384G, and transmission speed can reach more than 400G/s, often
Individual core supports four threads, and KNL processors realize the independent function as primary processor, and are realized on the basis of KNC
Great lifting, the embodiment of the present invention makes full use of high speed vector memory bandwidth technology and high speed L2 the caching skill of KNL platforms
Art, allows that memory access is controlled within L2 cache sizes in loop as far as possible, can quickly finish circulation, accelerates a section
The formation speed of domain micro-downburst, for KNL, two L2 cachings of the shared 1M of the core of KNL processors are realized
Generate the acceleration of the KNL many-core cluster platforms of the three-dimensional micro-downburst of multizone different parameters；The embodiment of the present invention also has
There is certain autgmentability, the generation of the micro-downburst of each KNL core/small region of thread independent process one, so this
The degree of parallelism of inventive embodiments is very high, and autgmentability is also substantially in the level of linear expansion.
Fig. 3 and Fig. 5 is referred to, a kind of the first implementation of method for generating micro-downburst is the embodiment of the invention provides
Example, including：
Each node in 101, KNL group systems receives the first information that control system transmission comes, and the first information includes
Collar vortex parameter setting, spatial dimension, resolution ratio, time range；
In embodiments of the present invention, it is necessary first to which each node in KNL group systems receives what control system transmission came
The first information, the first information includes but is not limited to collar vortex parameter setting, spatial dimension, resolution ratio, time range.
Each node in 102, KNL group systems generates micro- lower qi of chong channel ascending adversely according to the first information and by collar vortex guidance model
Flow field data；
In embodiments of the present invention, each node in KNL group systems receives the first letter that control system transmission comes
After breath, also micro- undershoot is generated according to the first information and by collar vortex guidance model including each node in KNL group systems
Air-flow field data is, it is necessary to explanation, can generate the micro-downburst field data of static state, it is also possible to real-time according to the first information
Generate dynamic micro-downburst field data.
Fig. 4 and Fig. 5 is referred to, a kind of the second implementation of method for generating micro-downburst is the embodiment of the invention provides
Example, including：
201, control system obtains the first information of generation micro-downburst place need and is distributed in KNL group systems
Each node；
In embodiments of the present invention, it is necessary first to which the first information that control system acquisition generation micro-downburst place needs is simultaneously
Be distributed to each node in KNL group systems, be sent to each node the first information information content can with it is identical can also
Difference, control system realizes load balancing between controlling each calculate node, and system equipment utilization rate is high, different calculate nodes, can be with
Realize calculating balanced, be not to wait for each other, computing device occurs without idle condition in system, whole system will realize efficiently fortune
Turn.
Each node in 202, KNL group systems receives the first information that control system transmission comes, and the first information includes
Collar vortex parameter setting, spatial dimension, resolution ratio, time range；
In embodiments of the present invention, obtain the first information of generation micro-downburst place need in control system and be distributed to
After each node in KNL group systems, in addition it is also necessary to which each node in KNL group systems receives what control system transmission came
The first information, the first information include collar vortex parameter setting, spatial dimension, resolution ratio, time range, it is necessary to explanation, each
The information content of the first information that node is received can be the same or different, and specifically be determined by control system.
203, piecemeal treatment is carried out to the spatial dimension in the first information；
In embodiments of the present invention, each node in KNL group systems receives the first letter that control system transmission comes
After breath, in addition it is also necessary to piecemeal treatment is carried out to the spatial dimension in the first information, such as an area dividing in spatial dimension
Into multiple regions, the MCDRAM internal memories and L2 spatial caches of high-transmission bandwidth are made full use of, solve the interior of existing hardware presence
The not enough technical problem of capacity, the network bandwidth is deposited, is reduced due to the delay that memory access brings.
204, if each node in KNL group systems opens main line corresponding with the spatial dimension processed by piecemeal
Number of passes, while generating micro-downburst field data by way of collar vortex guidance model and vectorization according to the first information；
In embodiments of the present invention, after the spatial dimension in the first information carries out piecemeal treatment, in addition it is also necessary to KNL collection
Each node in group's system opens some Thread Counts corresponding with the spatial dimension processed by piecemeal, while according to first
Information generates micro-downburst field data by way of collar vortex guidance model and vectorization；
It should be noted that in embodiments of the present invention, making full use of the multithreading of KNL, each node opens OMP
Multi-threaded parallel, so as to utilize KNL computing capabilitys, the micro-downburst at multiple nodes generation multiple time points on to greatest extent
Field data, does not interfere with each other, and without mutually wait between node, using dynamic load patterns, the upper multi-cores of KNL are used to generate micro- lower qi of chong channel ascending adversely
The vector field of the three dimensions different directions in flow field, makes full use of many-core technology and multi-node cluster technology, a KNL server
A panel region or a micro-downburst for time range are realized, many KNL clusters complete many continuous big regions or multiple not
With the generation of the micro-downburst of time range, realize that high-performance is quickly calculated, so as to quickly generate the three-dimensional micro- of big region
Down wash；During the embodiment of the present invention generates micro-downburst field data by collar vortex guidance model, also by collar vortex
A large amount of intermediate variables under induction algorithm multi-step are stored on MCDRAM internal memories, by the memory access address block in each loop
Change, improve the collar vortex induction algorithm of traditional static, realize the aspect optimization of algorithm, reduce memory access number of times, it is to avoid great-jump-forward
Memory access, substantially increases the time that collar vortex induces all directions different phase formation speed vector field；And, implement in the present invention
In example, the mode of vectorization is additionally used, such as in the wind speed in three directions in generating micro-downburst field data, fill as far as possible
Dividing the gain effect brought using vectorization can greatly improve program operation speed.
The micro-downburst field data asynchronous communication of generation is by each node in 205, KNL group systems to controlling
System；
In embodiments of the present invention, each node in KNL group systems is opened and the space model processed by piecemeal
Corresponding some Thread Counts are enclosed, while generating micro- undershoot by way of collar vortex guidance model and vectorization according to the first information
After air-flow field data, in addition it is also necessary to the micro-downburst field data asynchronous communication that each node in KNL group systems will be generated
To control system, because different nodes calculate the speed of the micro-downburst of different time points, the calculating time is different, must not
Wait is synchronized, therefore uses asynchronous system.
It is apparent to those skilled in the art that, for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, may be referred to the corresponding process in preceding method embodiment, will not be repeated here.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method can be with
Realize by another way.For example, device embodiment described above is only schematical, for example, the unit
Divide, only a kind of division of logic function there can be other dividing mode when actually realizing, for example multiple units or component
Can combine or be desirably integrated into another system, or some features can be ignored, or do not perform.It is another, it is shown or
The coupling each other for discussing or direct-coupling or communication connection can be the indirect couplings of device or unit by some interfaces
Close or communicate to connect, can be electrical, mechanical or other forms.
The unit that is illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit
The part for showing can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be according to the actual needs selected to realize the mesh of this embodiment scheme
's.
In addition, during each functional unit in each embodiment of the invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to which two or more units are integrated in a unit.Above-mentioned integrated list
Unit can both be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If the integrated unit is to realize in the form of SFU software functional unit and as independent production marketing or use
When, can store in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part for being contributed to prior art in other words or all or part of the technical scheme can be in the form of software products
Embody, the computer software product is stored in a storage medium, including some instructions are used to so that a computer
Equipment (can be personal computer, server, or network equipment etc.) performs the complete of each embodiment methods described of the invention
Portion or part steps.And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store journey
The medium of sequence code.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to preceding
Embodiment is stated to be described in detail the present invention, it will be understood by those within the art that：It still can be to preceding
State the technical scheme described in each embodiment to modify, or equivalent is carried out to which part technical characteristic；And these
Modification is replaced, and does not make the spirit and scope of the essence disengaging various embodiments of the present invention technical scheme of appropriate technical solution.
Claims (8)
1. a kind of system for generating micro-downburst, it is characterised in that including：Storage system, express network, KNL clusters system
System, control system；
The express network is used to connect the storage nodes, the KNL cluster nodes and the control system
Node.
2. the system of generation micro-downburst according to claim 1, it is characterised in that the storage system is specifically wrapped
Include shared storage subsystem and subsystem is locally stored；
The shared storage subsystem is used to store large-scale calculations data；
The subsystem that is locally stored is for storing the supplemental characteristic and calculation result data of the control system.
3. a kind of method for generating micro-downburst, it is characterised in that including：
Each node in KNL group systems receives the first information that the control system transmission comes, and the first information includes
Collar vortex parameter setting, spatial dimension, resolution ratio, time range；
Each node in KNL group systems generates micro-downburst according to the first information and by collar vortex guidance model
Data.
4. the method for generation micro-downburst according to claim 3, it is characterised in that in KNL group systems
Each node also includes before receiving the first information that the control system transmission comes：
The control system obtains the first information of generation micro-downburst place need and is distributed to the KNL group systems
In each node.
5. the method for generation micro-downburst according to claim 3, it is characterised in that in KNL group systems
Each node is received after the first information that the control system transmission comes, and each node in KNL group systems is according to institute
State the first information and by collar vortex guidance model generate micro-downburst field data before, also include：
The spatial dimension in the first information carries out piecemeal treatment.
6. the method for generation micro-downburst according to claim 3, it is characterised in that in KNL group systems
Each node according to the first information and by collar vortex guidance model generate micro-downburst field data after, also include：
The micro-downburst field data asynchronous communication that each node in KNL group systems will be generated is to the control system.
7. the method for generation micro-downburst according to claim 3, it is characterised in that every in KNL group systems
Individual node is specially according to the first information and by collar vortex guidance model generation micro-downburst field data：
Each node in KNL group systems is given birth to according to the first information, by way of collar vortex guidance model and vectorization
Into micro-downburst field data.
8. the method for generation micro-downburst according to claim 5, it is characterised in that every in KNL group systems
Individual node is specially according to the first information and by collar vortex guidance model generation micro-downburst field data：
Each node in KNL group systems opens some threads corresponding with the spatial dimension processed by piecemeal
Number, while generating micro-downburst field data by collar vortex guidance model according to the first information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710112528.1A CN106897148A (en) | 2017-02-28 | 2017-02-28 | A kind of system and method for generating micro-downburst |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710112528.1A CN106897148A (en) | 2017-02-28 | 2017-02-28 | A kind of system and method for generating micro-downburst |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106897148A true CN106897148A (en) | 2017-06-27 |
Family
ID=59184912
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710112528.1A Pending CN106897148A (en) | 2017-02-28 | 2017-02-28 | A kind of system and method for generating micro-downburst |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106897148A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590353A (en) * | 2017-10-19 | 2018-01-16 | 郑州云海信息技术有限公司 | The cluster of the server of air turbulence field stimulation method and use KNL processors |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104580503A (en) * | 2015-01-26 | 2015-04-29 | 浪潮电子信息产业股份有限公司 | Efficient dynamic load balancing system and method for processing large-scale data |
| CN104881559A (en) * | 2015-06-23 | 2015-09-02 | 陕西宝成航空仪表有限责任公司 | Micro undershoot airflow engineering simulation modeling method |
| CN105182989A (en) * | 2015-09-25 | 2015-12-23 | 北京航空航天大学 | Airplane attitude control method under influence of wind field |
-
2017
- 2017-02-28 CN CN201710112528.1A patent/CN106897148A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104580503A (en) * | 2015-01-26 | 2015-04-29 | 浪潮电子信息产业股份有限公司 | Efficient dynamic load balancing system and method for processing large-scale data |
| CN104881559A (en) * | 2015-06-23 | 2015-09-02 | 陕西宝成航空仪表有限责任公司 | Micro undershoot airflow engineering simulation modeling method |
| CN105182989A (en) * | 2015-09-25 | 2015-12-23 | 北京航空航天大学 | Airplane attitude control method under influence of wind field |
Non-Patent Citations (1)
| Title |
|---|
| 丁丹迪 等: "LPL-S在Knights Landing上的优化与性能评估", 《计算机科学与探索》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590353A (en) * | 2017-10-19 | 2018-01-16 | 郑州云海信息技术有限公司 | The cluster of the server of air turbulence field stimulation method and use KNL processors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106446351A (en) | Real-time drawing-oriented large-scale scene organization and scheduling technology and simulation system | |
| CN108563808A (en) | The design method of heterogeneous reconfigurable figure computation accelerator system based on FPGA | |
| CN102393826B (en) | Multi-core parallel processing based flexible scene continuous collision detection method | |
| CN102736940B (en) | Resource loading method | |
| CN103970602B (en) | Data flow program scheduling method oriented to multi-core processor X86 | |
| CN103617150A (en) | GPU (graphic processing unit) based parallel power flow calculation system and method for large-scale power system | |
| CN104077138A (en) | Multiple core processor system for integrating network router, and integrated method and implement method thereof | |
| CN110362380A (en) | A kind of multiple-objection optimization virtual machine deployment method in network-oriented target range | |
| CN104572501B (en) | Access trace locality analysis-based shared buffer optimization method in multi-core environment | |
| CN111371616B (en) | Virtual network function chain deployment method and system for NUMA (non Uniform memory Access) architecture server | |
| CN105468439A (en) | Adaptive parallel algorithm for traversing neighbors in fixed radius under CPU-GPU (Central Processing Unit-Graphic Processing Unit) heterogeneous framework | |
| CN114490011B (en) | Parallel acceleration realization method of N-body simulation in heterogeneous architecture | |
| CN113343489B (en) | Satellite communication simulation method and system based on container technology and digital twin technology | |
| CN112732436B (en) | Deep reinforcement learning acceleration method of multi-core processor-single graphics processor | |
| CN106776466A (en) | A kind of FPGA isomeries speed-up computation apparatus and system | |
| CN113342034A (en) | Combined strategy algorithm for unmanned aerial vehicle channel inspection and refined inspection | |
| CN107463448A (en) | A kind of deep learning weight renewing method and system | |
| CN106897148A (en) | A kind of system and method for generating micro-downburst | |
| CN101800768A (en) | Gridding data transcription generation method based on storage alliance subset partition | |
| CN104050189B (en) | The page shares processing method and processing device | |
| CN106407005A (en) | Concurrent process merging method and system based on multi-scale coupling | |
| CN107590353A (en) | The cluster of the server of air turbulence field stimulation method and use KNL processors | |
| CN104331336B (en) | Be matched with the multilayer nest balancing method of loads of high-performance computer structure | |
| CN107392935A (en) | A kind of particle computational methods and particIe system based on integral formula | |
| CN106846238A (en) | A kind of cross-platform automotive engine system of Elf3D |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| 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: 20170627 |