CN107590353A - The cluster of the server of air turbulence field stimulation method and use KNL processors - Google Patents
The cluster of the server of air turbulence field stimulation method and use KNL processors Download PDFInfo
- Publication number
- CN107590353A CN107590353A CN201710979267.3A CN201710979267A CN107590353A CN 107590353 A CN107590353 A CN 107590353A CN 201710979267 A CN201710979267 A CN 201710979267A CN 107590353 A CN107590353 A CN 107590353A
- Authority
- CN
- China
- Prior art keywords
- knl
- processors
- server
- cluster
- air turbulence
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of cluster of server the invention discloses air turbulence field stimulation method and using KNL processors.This method includes：Initialize the air turbulence parameter and spatial dimension of different time points；The master server using KNL processors in the cluster using the server of KNL processors obtains air turbulence parameter setting, spatial dimension, the time range required for virtual environment emulation, and each in cluster is distributed them to using KNL processors from server, the thread of respective numbers, each air turbulence field data out of server using KNL processors of parallel generation are each opened from server；By the asynchronous master server passed back using KNL processors of air turbulence field data at each corresponding time point calculated from server using KNL processors.Calculate the air turbulence field data of multiple spatial dimensions and time range respectively using multiple cores, can concurrently carry out algorithm operating, improve computational efficiency, reduce the time of consuming, generate data in real time.
Description
Technical field
The present invention relates to information technology, espespecially a kind of air turbulence field stimulation method and the server using KNL processors
Cluster.
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 tested requires more and more higher.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
Wordization represents, including atmospheric environment, space electromagnetic environment, terrain environment.Current single, Utopian natural environment can not meet
Armament systems are to the demand of environment, and there is an 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 according to scenario and quickly generate ring
Border data.Therefore need according to scenario, integrate atmospheric environment, electromagnetic environment, terrain environment, typical adverse circumstances etc., fast fast-growing
Data supporting is provided into virtual natural environment data, and to dummy experiment system.In a spatial dimension, big compression ring is integrated with
Three kinds of border, electromagnetic environment, terrain environment different types of environment；Typical environment is integrated with background environment, such as in aircraft
Different typical atmospheric environment are integrated on path, typical electromagnetic environment is integrated, typical terrain environment etc. is integrated on ground, so
Verify armament systems under extreme weather by efficiency, and the normal flight that disturbance wind field often influences weapon aircraft is lived
Dynamic, serious also to produce aircraft accident, air turbulence and low are rather common in disturbance wind field.Wherein air turbulence
It is another atmospheric perturbation phenomenon for influenceing flight quality, the aircraft to be flown in field of turbulent flow can produce turbulent flow and jolt, and weigh
Degree turbulent flow can even influence the completion of aerial mission, harm flight.
In hydrodynamics, turbulent flow is a kind of a kind of fluidised form of unordered chaos change, including low energy spreads, high-energy pair
The quick change of fast flow velocity in stream and space-time unique.Turbulence Flow has highly irregular property, and real field of turbulent flow is very multiple
Miscellaneous, and often modeled in engineering using random theory, researcher proposes the longitudinal direction of exponential type by numerous studies
And horizontal correlation function, frequency spectrum correlation function formula is then derived from, as long as generally providing space length, air turbulence chi
Degree, spatial frequency, air turbulence field is then generated by CPU calculating platforms.Prior art generation air turbulence field is to pass through CPU
Calculating platform calculates generation, and calculating is time-consuming longer, therefore can not generate in real time, and the solution method of prior art is pre-configured
The parameter of air turbulence field, generation specified time, place, the air turbulence field data of parameter, and deposit in database, work as progress
During virtual environment test simulation, then inquired about from database, interpolation etc. calculates；The shortcomings that this conventional method is：(1) pass through
CPU calculating platforms calculate, and algorithm is serial operation, take cycle length；(2) data can not be generated in real time, can only previously generate number
According to, thus have certain limitation in the emulation of real-time virtual natural environment.
The content of the invention
In order to solve the above-mentioned technical problem, the invention provides a kind of air turbulence field stimulation method and using KNL processing
The cluster of the server of device, it can carry out parallel work-flow to algorithm, and can generate data in real time.
In order to reach the object of the invention, the embodiments of the invention provide a kind of air turbulence field stimulation method, this method should
For the cluster of the server using KNL processors, wherein, KNL processors have the core of the first quantity, and each core is supported
The thread of second quantity, this method include：
Initialize the air turbulence parameter and spatial dimension of different time points；
The master server using KNL processors in the cluster of the server using KNL processors obtains virtual
Air turbulence parameter setting, spatial dimension, time range required for environmental simulation, and distribute them to using at KNL
Each in the cluster of the server of device is managed using KNL processors from server, it is each using KNL processors from server
Open the thread of respective numbers, each air turbulence field data out of server using KNL processors of parallel generation；
By the asynchronous biography of air turbulence field data at each corresponding time point calculated from server using KNL processors
Return the master server using KNL processors.
Compared with prior art, the embodiment of the present invention uses KNL processing by the way that air turbulence field stimulation method is applied to
The cluster of the server of device, there is the core of the first quantity using KNL processors and each core supports the line of the second quantity
The characteristics of journey, calculate the air turbulence field data of multiple spatial dimensions and time range respectively using multiple cores, can be parallel
Ground carries out algorithm operating, improves computational efficiency, reduces the time of consuming.Further, since calculated using Multi-core
Method operates, therefore can generate data in real time, the Three-Dimensional Atmospheric Turbulence field being suitable under generation different parameters state in real time.
Further, in an optional embodiment, using KNL processors server also include multichannel dynamic with
Machine memory, multichannel dynamic RAM preserve the multistage intermediate variable of air turbulence field generating algorithm.
Further, in an optional embodiment, order memory access is used using the cluster of the server of KNL processors
Mode each internal memory is circulated in the storage address that accesses it is blocking.
Further, in an optional embodiment, it is total to using the cluster of the server of KNL processors using OpenMP
Enjoy the operation of memory parallel programming mode multi-threaded parallel.Thus, it is possible to maximally utilise the multi-core parallel concurrent meter of KNL processors
Calculation ability.
Further, in an optional embodiment, vectorization meter is used using the cluster of the server of KNL processors
The mode of calculation is run.
In order to reach the object of the invention, the embodiment of the present invention additionally provides a kind of collection of the server using KNL processors
Group, KNL processors have the core of the first quantity, and each core supports the thread of the second quantity, wherein, it is described using at KNL
Manage the air turbulence parameter and spatial dimension of the cluster initialization different time points of the server of device；
The master server using KNL processors in the cluster of the server using KNL processors obtains virtual
Air turbulence parameter setting, spatial dimension, time range required for environmental simulation, and distribute them to using at KNL
Each in the cluster of the server of device is managed using KNL processors from server, it is each using KNL processors from server
Open the thread of respective numbers, each air turbulence field data out of server using KNL processors of parallel generation；
It is each using KNL processors from server by the asynchronous biography of air turbulence field data at the corresponding time point calculated
Return the master server using KNL processors.
Compared with prior art, the cluster of the server using KNL processors in embodiments of the present invention is using at KNL
Reason utensil has the characteristics of thread of the core of the first quantity and each core the second quantity of support, is counted respectively using multiple cores
The air turbulence field data of multiple spatial dimensions and time range is calculated, can concurrently carry out algorithm operating, improves computational efficiency,
Reduce the time expended.Further, since being to carry out algorithm operating using Multi-core, therefore number can be generated in real time
According to the Three-Dimensional Atmospheric Turbulence field being suitable under generation different parameters state in real time.
Further, in an optional embodiment, the server using KNL processors also moves including multichannel
State random access memory, the multichannel dynamic RAM preserve the multistage middle anaplasia of air turbulence field generating algorithm
Amount.
Further, in an optional embodiment, the cluster of the server using KNL processors is using order
The storage address that the mode of memory access accesses during each internal memory is circulated is blocking.
Further, in an optional embodiment, the cluster of the server using KNL processors uses
OpenMP shares the operation of memory parallel programming mode multi-threaded parallel.
Further, in an optional embodiment, the cluster of the server using KNL processors is using vector
Change the mode calculated to run.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by specification, rights
Specifically noted structure is realized and obtained in claim and accompanying drawing.
Brief description of the drawings
Accompanying drawing is used for providing further understanding technical solution of the present invention, and a part for constitution instruction, with this
The embodiment of application is used to explain technical scheme together, does not form the limitation to technical solution of the present invention.
Fig. 1 is the flow chart according to the air turbulence field stimulation method of embodiments of the invention；
Fig. 2 is the structural representation according to the cluster of the server using KNL processors of embodiments of the invention.
Embodiment
For the object, technical solutions and advantages of the present invention are more clearly understood, below in conjunction with accompanying drawing to the present invention
Embodiment be described in detail.It should be noted that in the case where not conflicting, in the embodiment and embodiment in the application
Feature can mutually be combined.
Can be in the computer system of such as one group computer executable instructions the flow of accompanying drawing illustrates the step of
Perform.Also, although logical order is shown in flow charts, in some cases, can be with suitable different from herein
Sequence performs shown or described step.
On the one hand, the embodiment provides a kind of air turbulence field stimulation method, this method is applied to use KNL
The cluster of the server of processor, wherein, KNL processors have the core of the first quantity, and each core supports the second quantity
Thread, as shown in figure 1, the method comprising the steps of S101- steps S109.
Step S101, initialize the air turbulence parameter and spatial dimension of different time points.
Step S105, the master server using KNL processors in the cluster using the server of KNL processors obtain
Air turbulence parameter setting, spatial dimension, time range required for virtual environment emulation, and distribute them to and using
Each in the cluster of the server of KNL processors using KNL processors from server, it is each using KNL processors from
Server opens the thread of respective numbers, each air turbulence number of fields out of server using KNL processors of parallel generation
According to.
KNL is Knights Landing abbreviation, is second generation Intel to melting core processor by force, before it is combined
To (host CPU can be done, be many-core again -- 72 cores) the advantages of strong and coprocessor.
Step S109, by the air turbulence field at each corresponding time point calculated from server using KNL processors
The asynchronous master server passed back using KNL processors of data.
Here, because the different nodes (server for using KNL processors) in server cluster calculate different time points
Air turbulence field speed, therefore wait need not be synchronized, therefore using the number of asynchronous system transmission air turbulence field
According to.
The beneficial effect of embodiments of the invention is：By the way that air turbulence field stimulation method is applied to using at KNL
The cluster of the server of device is managed, there is the core of the first quantity using KNL processors and each core supports the second quantity
The characteristics of thread, calculate the air turbulence field data of multiple spatial dimensions and time range respectively using multiple cores, can be simultaneously
Algorithm operating is carried out capablely, improves computational efficiency, reduces the time of consuming.Further, since carried out using Multi-core
Algorithm operating, therefore data can be generated in real time, the Three-Dimensional Atmospheric Turbulence field being suitable under generation different parameters state in real time.
Further, in an optional embodiment, using KNL processors server also include multichannel dynamic with
Machine memory, multichannel dynamic RAM preserve the multistage intermediate variable of air turbulence field generating algorithm.
Each server using KNL processors includes 16G multichannel dynamic RAM (being abbreviated as MCDRAM),
MCDRAM transmission speed can reach more than 400G/s, and the 90G/s compared to DDR4 has very big lifting.Thus, it is possible to will
MCDRAM caches (L3 cachings) as the three-level of KNL processors, preserves the multistage middle anaplasia of air turbulence field generating algorithm
Amount.Because MCDRAM transmission speeds are high and memory space is huge, thus by air turbulence field generating algorithm it is multistage in
Between variable be stored in MCDRAM, can be when KNL processors extract the characteristic value of air turbulence field, miss
In the case of the level cache and L2 cache of (cache miss) KNL processors, (cache hit) MCDRAM is hit, from
And reduce to need to access because of miss caching (cache miss) and the increasing of memory time is accessed caused by server memory
The probability added, improve the speed that air turbulence field data calculating is carried out using the server of KNL processors.
Further, in an optional embodiment, order memory access is used using the cluster of the server of KNL processors
Mode each internal memory is circulated in the storage address that accesses it is blocking.
In order to make full use of the computing capability in each internal memory circulation, therefore, in the embodiment of this method embodiment, adopt
The storage address accessed during each internal memory is circulated with the cluster of the server of KNL processors is blocking, so as to avoid jumping
Memory is accessed likes, makes full use of L2 cache, reduces the time accessed spent by memory, raising uses KNL processors
Server efficiency.
Further, in an optional embodiment, it is total to using the cluster of the server of KNL processors using OpenMP
Enjoy the operation of memory parallel programming mode multi-threaded parallel.Thus, it is possible to maximally utilise the multi-core parallel concurrent meter of KNL processors
Calculation ability.
Further, in an optional embodiment, vectorization meter is used using the cluster of the server of KNL processors
The mode of calculation is run.
It is a kind of mode of special parallel computation that vectorization, which calculates, and one is only carried out in the same time compared to general procedure
The mode of individual operation, it can perform multi-pass operation in the same time, typically different data are performed same one or
A collection of instruction, in other words application of instruction in an array/vector.The gain effect for making full use of vectorization to bring can be very big
Raising program operation speed.For KNL processors, each thread that handles can handle 16 single precision floating datums every time
Operation, once-through operation can only be carried out every time within the identical time and for the processor of prior art, during serial process.
Vectorization technology is considered in program implement, be capable of vectorization position compile vectorization, it is determined that no data according to
The manual vectorization of bad program segment, is finally reached the vectorization of processor rank.
On the other hand, the embodiment provides a kind of cluster 200 of the server using KNL processors, at KNL
Reason utensil has the core of the first quantity, and each core supports the thread of the second quantity, as shown in Fig. 2 wherein,
The air turbulence parameter and space model of different time points are initialized using the cluster 200 of the server of KNL processors
Enclose；
Void is obtained using the master server 210 of KNL processors in the cluster 200 using the server of KNL processors
Intend air turbulence parameter setting, spatial dimension, the time range required for environmental simulation, and required for virtual environment is emulated
Air turbulence parameter setting, spatial dimension, time range be distributed in the cluster 200 using the server of KNL processors
It is each using KNL processors from server 220, it is each that respective numbers are opened from server 220 using KNL processors
Thread, each air turbulence field data out of server using KNL processors of parallel generation；
It is each from server 220 that the air turbulence field data at the corresponding time point calculated is different using KNL processors
Step passes the master server 210 using KNL processors back.
The beneficial effect of embodiments of the invention is：Using core of the KNL processors with the first quantity and each
Core supports the characteristics of thread of the second quantity, calculates the air of multiple spatial dimensions and time range respectively using multiple cores
Turbulent field data, algorithm operating can be concurrently carried out, improve computational efficiency, reduce the time of consuming.Further, since it is to use
Algorithm operating is carried out Multi-core, therefore can generate data in real time, is suitable for generating under different parameters state in real time
Three-Dimensional Atmospheric Turbulence field.
Further, in an optional embodiment, using KNL processors server (including use KNL processors
Master server and using KNL processors from server) also include multichannel dynamic RAM, this multichannel is dynamically
Random access memory preserves the multistage intermediate variable of air turbulence field generating algorithm.
Further, in an optional embodiment, order memory access is used using the cluster of the server of KNL processors
Mode each internal memory is circulated in the storage address that accesses it is blocking.
Further, in an optional embodiment, it is total to using the cluster of the server of KNL processors using OpenMP
Enjoy the operation of memory parallel programming mode multi-threaded parallel.
Further, in an optional embodiment, vectorization meter is used using the cluster of the server of KNL processors
The mode of calculation is run.
Although disclosed herein embodiment as above, above-mentioned content be only readily appreciate the present invention and use
Embodiment, it is not limited to the present invention.Technical staff in any art of the present invention, taken off not departing from the present invention
On the premise of the spirit and scope of dew, any modification and change, but the present invention can be carried out in the form and details of implementation
Scope of patent protection, still should be subject to the scope of the claims as defined in the appended claims.
Claims (10)
- A kind of 1. air turbulence field stimulation method, it is characterised in that methods described is applied to the server for using KNL processors Cluster, wherein, KNL processors have the core of the first quantity, and each core supports the thread of the second quantity, methods described bag Include：Initialize the air turbulence parameter and spatial dimension of different time points；The master server using KNL processors in the cluster of the server using KNL processors obtains virtual environment Air turbulence parameter setting, spatial dimension, time range required for emulation, and distribute them to and using KNL processors Server cluster in it is each using KNL processors from server, it is each to be opened using KNL processors from server The thread of respective numbers, each air turbulence field data out of server using KNL processors of parallel generation；Institute is passed back by the air turbulence field data at each corresponding time point calculated from server using KNL processors is asynchronous State the master server using KNL processors.
- 2. according to the method for claim 1, it is characterised in that the server using KNL processors also includes more logical Road dynamic RAM, the multichannel dynamic RAM preserve the multistage centre of air turbulence field generating algorithm Variable.
- 3. according to the method for claim 1, it is characterised in that the cluster of the server using KNL processors uses The storage address that the mode of order memory access accesses during each internal memory is circulated is blocking.
- 4. according to the method for claim 1, it is characterised in that the cluster of the server using KNL processors uses OpenMP shares the operation of memory parallel programming mode multi-threaded parallel.
- 5. according to the method for claim 1, it is characterised in that the cluster of the server using KNL processors uses The mode that vectorization calculates is run.
- A kind of 6. cluster of server using KNL processors, it is characterised in that KNL processors have the core of the first quantity, Each core supports the thread of the second quantity, wherein,The air turbulence parameter and spatial dimension of the cluster initialization different time points of the server using KNL processors；The master server using KNL processors in the cluster of the server using KNL processors obtains virtual environment Air turbulence parameter setting, spatial dimension, time range required for emulation, and distribute them to and using KNL processors Server cluster in it is each using KNL processors from server, it is each to be opened using KNL processors from server The thread of respective numbers, each air turbulence field data out of server using KNL processors of parallel generation；It is each that institute is passed back by the air turbulence field data at the corresponding time point calculated is asynchronous from server using KNL processors State the master server using KNL processors.
- 7. cluster according to claim 6, it is characterised in that the server using KNL processors also includes more logical Road dynamic RAM, the multichannel dynamic RAM preserve the multistage centre of air turbulence field generating algorithm Variable.
- 8. cluster according to claim 6, it is characterised in that the cluster of the server using KNL processors uses The storage address that the mode of order memory access accesses during each internal memory is circulated is blocking.
- 9. cluster according to claim 6, it is characterised in that the cluster of the server using KNL processors uses OpenMP shares the operation of memory parallel programming mode multi-threaded parallel.
- 10. cluster according to claim 6, it is characterised in that the cluster of the server using KNL processors uses The mode that vectorization calculates is run.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710979267.3A CN107590353A (en) | 2017-10-19 | 2017-10-19 | The cluster of the server of air turbulence field stimulation method and use KNL processors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710979267.3A CN107590353A (en) | 2017-10-19 | 2017-10-19 | The cluster of the server of air turbulence field stimulation method and use KNL processors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107590353A true CN107590353A (en) | 2018-01-16 |
Family
ID=61052937
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710979267.3A Pending CN107590353A (en) | 2017-10-19 | 2017-10-19 | The cluster of the server of air turbulence field stimulation method and use KNL processors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107590353A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113239462A (en) * | 2021-05-25 | 2021-08-10 | 江苏普旭科技股份有限公司 | Simulation method for aircraft turbulent environment simulation |
| CN113688578A (en) * | 2021-07-13 | 2021-11-23 | 中国空气动力研究与发展中心计算空气动力研究所 | Flow field key time step extraction and reconstruction method based on multivariate fusion |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000322402A (en) * | 1999-05-11 | 2000-11-24 | Hitachi Ltd | Stream of people analyzing method |
| CN101650883A (en) * | 2009-02-13 | 2010-02-17 | 中国人民解放军空军航空大学 | Simulation method of atmospheric turbulence on flight simulator |
| CN105387991A (en) * | 2015-12-02 | 2016-03-09 | 同济大学 | Wind-tunnel turbulent flow field simulation method and device |
| CN106096091A (en) * | 2016-05-31 | 2016-11-09 | 中国航空工业集团公司西安飞机设计研究所 | A kind of airplane motion analogy method |
| CN106844037A (en) * | 2017-02-22 | 2017-06-13 | 郑州云海信息技术有限公司 | A kind of method of testing and system based on KNL |
| CN106897148A (en) * | 2017-02-28 | 2017-06-27 | 郑州云海信息技术有限公司 | A kind of system and method for generating micro-downburst |
-
2017
- 2017-10-19 CN CN201710979267.3A patent/CN107590353A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000322402A (en) * | 1999-05-11 | 2000-11-24 | Hitachi Ltd | Stream of people analyzing method |
| CN101650883A (en) * | 2009-02-13 | 2010-02-17 | 中国人民解放军空军航空大学 | Simulation method of atmospheric turbulence on flight simulator |
| CN105387991A (en) * | 2015-12-02 | 2016-03-09 | 同济大学 | Wind-tunnel turbulent flow field simulation method and device |
| CN106096091A (en) * | 2016-05-31 | 2016-11-09 | 中国航空工业集团公司西安飞机设计研究所 | A kind of airplane motion analogy method |
| CN106844037A (en) * | 2017-02-22 | 2017-06-13 | 郑州云海信息技术有限公司 | A kind of method of testing and system based on KNL |
| CN106897148A (en) * | 2017-02-28 | 2017-06-27 | 郑州云海信息技术有限公司 | A kind of system and method for generating micro-downburst |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113239462A (en) * | 2021-05-25 | 2021-08-10 | 江苏普旭科技股份有限公司 | Simulation method for aircraft turbulent environment simulation |
| CN113688578A (en) * | 2021-07-13 | 2021-11-23 | 中国空气动力研究与发展中心计算空气动力研究所 | Flow field key time step extraction and reconstruction method based on multivariate fusion |
| CN113688578B (en) * | 2021-07-13 | 2023-05-23 | 中国空气动力研究与发展中心计算空气动力研究所 | Flow field key time step extraction and reconstruction method based on multivariate fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109002659B (en) | Fluid machinery simulation program optimization method based on super computer | |
| CN103617150A (en) | GPU (graphic processing unit) based parallel power flow calculation system and method for large-scale power system | |
| CN102393826B (en) | Multi-core parallel processing based flexible scene continuous collision detection method | |
| CN103064819A (en) | Method for utilizing microwave integrated circuit (MIC) to rapidly achieve lattice Boltzmann parallel acceleration | |
| CN112632874A (en) | Optimization method and system for numerical simulation of helicopter flow field | |
| CN106776466A (en) | A kind of FPGA isomeries speed-up computation apparatus and system | |
| CN107590353A (en) | The cluster of the server of air turbulence field stimulation method and use KNL processors | |
| CN107064930A (en) | Radar foresight imaging method based on GPU | |
| Chen et al. | Tetrahedral mesh improvement by shell transformation | |
| Gao et al. | Optimization for viewshed analysis on GPU | |
| CN108197075B (en) | Multi-core implementation method of Inceptation structure | |
| CN103914428A (en) | Efficient communication method of structural analysis under multi-core distributed computing environment | |
| Mostafazadeh Davani et al. | Unsteady Navier-Stokes computations on GPU architectures | |
| Jameson et al. | Automatic aerodynamic optimization on distributed memory architectures | |
| CN106484532B (en) | GPGPU parallel calculating method towards SPH fluid simulation | |
| Tang et al. | Optimizing and auto-tuning iterative stencil loops for GPUs with the in-plane method | |
| Wei et al. | GPU acceleration of a 2D compressible Euler solver on CUDA-based block-structured Cartesian meshes | |
| CN111177979A (en) | Fluid dynamics software GASFLOW optimization method based on OpenMP | |
| Schröder et al. | Field strength prediction for environment aware MIMO channel models | |
| CN109522127A (en) | A kind of fluid machinery simulated program isomery accelerated method based on GPU | |
| CN105572730B (en) | 3 D complex structure sound wave forward modeling method | |
| CN112182781B (en) | Aircraft surface structure setting method, device, equipment and storage medium | |
| Gao et al. | A multi-level parallel tie-dye algorithm for auto-CFD | |
| Tian et al. | Efficient Parallel Dynamic Overset Unstructured Grid for Flow Simulation with Moving Bodies | |
| CN106897148A (en) | A kind of system and method for generating micro-downburst |
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: 20180116 |