CN102866423B - Seismic prestack time migration processing method and system - Google Patents
Seismic prestack time migration processing method and system Download PDFInfo
- Publication number
- CN102866423B CN102866423B CN201210339234.XA CN201210339234A CN102866423B CN 102866423 B CN102866423 B CN 102866423B CN 201210339234 A CN201210339234 A CN 201210339234A CN 102866423 B CN102866423 B CN 102866423B
- Authority
- CN
- China
- Prior art keywords
- pstm
- fft
- calculation
- cpu
- buffer zone
- 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.)
- Active
Links
- 238000013508 migration Methods 0.000 title claims abstract description 26
- 230000005012 migration Effects 0.000 title claims abstract description 26
- 238000003672 processing method Methods 0.000 title abstract 2
- 238000000034 method Methods 0.000 claims abstract description 79
- 238000004364 calculation method Methods 0.000 claims abstract description 62
- 230000008569 process Effects 0.000 claims description 26
- 230000005540 biological transmission Effects 0.000 claims description 15
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 13
- 230000003139 buffering effect Effects 0.000 abstract description 2
- 241001269238 Data Species 0.000 description 9
- 238000012423 maintenance Methods 0.000 description 6
- 238000010276 construction Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000003208 petroleum Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 102000008482 12E7 Antigen Human genes 0.000 description 1
- 108010020567 12E7 Antigen Proteins 0.000 description 1
- 101000893549 Homo sapiens Growth/differentiation factor 15 Proteins 0.000 description 1
- 101000692878 Homo sapiens Regulator of MON1-CCZ1 complex Proteins 0.000 description 1
- 102100026436 Regulator of MON1-CCZ1 complex Human genes 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to a seismic prestack time migration (PSTM) processing method and a system. The method comprises N PSTM processing procedures which are launched simultaneously, wherein each PSTM processing procedure comprises a fast fourier transform (FFT) calculation step: reading seismic trace data of a batch to be processed to have FFT calculation, alternatively caching the FFT calculation results of different batches into an idle first or an idle second cache area corresponding to the PSTM processing procedure; and a PSTM calculation step: alternately carrying out the PSTM calculation for the FFT calculation result of each batch in the first or the second buffering area corresponding to the PSTM processing procedure. The method can improve the seismic prestack time migration processing efficiency.
Description
Technical field
The present invention relates to computing machine high-performance computing sector, oil seismic exploration field, be specifically related to a kind of disposal route and system of seismic pre-stack time migration.
Background technology
In fields such as earthquake petroleum prospectings, seismic pre-stack time migration (PreStack Time Migration, PSTM) be one of most effective method of engineering construction system, the situation that vertical lateral speed change is larger can be adapted to, be applicable to the migration imaging of high spud angle.PSTM has carried out studying for many years, the nineties initial stage in last century starts Preliminary Applications, middle and later periods has played vital role in the seismic prospecting in many exploratory areas, start after entering this century and apply comparatively widely, current part process company and computing center join this technology in conventional process flow as conventional software, become to obtain to protect width information realization attributive analysis, the important step of AVO/AVA/AVP inverting and other parametric inversion and foundation.
PSTM often exports a seismic trace, is exactly a magnanimity computing.With 1 millisecond of sampling, 6 seconds data instances, the output of a seismic trace needs the input channel at least 1,000 ten thousand roads even more (migration aperture decision), and each point will do rms twice computing and twice additive operation, amplitude compensation twice multiplying.Calculating like this is got off, and realizes together skew needs 1000000 × 6000 × 2 × (square+addition+multiplication) secondary mathematical operation, calculated amount and need data volume to be processed all extremely huge! At present, people often use large-scale server cluster to carry out migration before stack process, and its principle data is first assigned on each CPU core, then solely calculated by each CPU vouching, finally result is gathered output.This way consumes a large amount of time, electric power and maintenance cost.And, along with the cycle request of people to petroleum prospecting seism processing is shorter and shorter, accuracy requirement is more and more higher, the scale of server cluster is done larger and larger, in system constructing cost, data center machine room space, internal memory and I/O bandwidth, power consumption heat radiation and electrical power limit, manageability, programming simplification, extendability, management maintenance expense etc., be all faced with huge challenge.
Summary of the invention
A kind of disposal route and system of seismic pre-stack time migration is provided, with the problem that the treatment effeciency solving seismic pre-stack time migration is low during the technical problem to be solved in the present invention.
For solving the problems of the technologies described above, the invention provides the disposal route of a kind of seismic pre-stack time migration (PSTM), the method comprises the parallel N number of PSTM treatment scheme initiated, and each PSTM treatment scheme comprises:
FFT calculation procedure, reads the seismic channel data of pending batch and carries out FFT calculating, and the FFT result of calculation of each batch is alternately buffered into first or second buffer zone of free time corresponding to this PSTM treatment scheme;
PSTM calculation procedure, alternately carries out PSTM calculating to each batch of FFT result of calculation in the first or second buffer zone corresponding to this PSTM treatment scheme.
Alternatively, the system of application the method comprises 1 central processing unit (CPU) end and (N-1) individual integrated many core (MIC) end, wherein the FFT calculation procedure of N number for the treatment of scheme is held by CPU and is realized, PSTM calculation procedure in described N number for the treatment of scheme is held by this CPU respectively and (N-1) individual MIC holds realization, when described PSTM calculation procedure is realized by MIC end, the FFT result of calculation asynchronous transmission in first, second buffer zone is held to MIC by described CPU end.
For solving aforementioned technical problem, present invention also offers the disposal system of a kind of seismic pre-stack time migration (PSTM), this system comprises N number of parallel performance element, and each performance element comprises:
FFT computing module, for reading the seismic channel data of pending batch and carrying out FFT calculating, also for the FFT result of calculation of each batch alternately being buffered into first or second buffer zone of the free time of this performance element;
First buffer zone, the second buffer zone, for the FFT result of calculation of each batch of buffer memory;
PSTM computing module, for alternately carrying out PSTM calculating to the first buffer zone of this performance element, the FFT result of calculation of the second buffer zone.
Alternatively, this system comprises 1 central processing unit (CPU) end and (N-1) individual integrated many core (MIC) end, corresponding, wherein the FFT computing module of N number of performance element holds realization with first, second buffer zone by CPU, and the PSTM computing module of described N number of performance element is held by this CPU respectively and (N-1) individual MIC end realizes; Described system also comprises the asynchronous transmission module that CPU end realizes, when the PSTM computing module of described performance element is realized by MIC end, described asynchronous transmission module, for the PSTM computing module FFT result of calculation asynchronous transmission in first, second buffer zone of this performance element realized to MIC end.
The inventive method and system adopt the mode of parallel processing, effectively can improve the efficiency of seismic pre-stack time migration process, in addition, FFT calculating section is placed on CPU end to perform, the parallel computation while that PSTM calculating section adopting CPU and MIC, and greatly reduces power consumption, decreases machine room construction cost and management, operation, maintenance cost, and this method realizes simple, the cost of development of needs is low.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of the disposal route embodiment 1 of seismic pre-stack time migration of the present invention;
Fig. 2 is for realizing pre-stack time migration acceleration logic structural drawing;
Fig. 3 is the building-block of logic that asynchronous parallel realizes;
Fig. 4 is the schematic diagram of the disposal route embodiment 2 of seismic pre-stack time migration of the present invention;
Fig. 5 is the modular structure schematic diagram of the disposal system of seismic pre-stack time migration.
Embodiment
In order to improve defect of the prior art, analyze as follows:
(1) performance bottleneck of PSTM is located;
The computation partition of PSTM: analyze PSTM program, according to function, PSTM is divided into FFT calculating section and PSTM calculating section, the working time of whole PSTM program is that FFT calculates and PSTM calculates sum working time;
Select different imaging spaces (as body is biased into image space, line is biased into image space) and input channel data scale operation PSTM program, write down FFT respectively to calculate and PSTM calculates working time of two parts, find that PSTM calculates and account for more than 90% of whole PSTM program runtime, such as: select an individuality to be biased into image space, its side line number is 91, it is 963 that the CMP of every bar side line counts, input channel number of channels is 110000, size is 70G, run PSTM program, find that PSTM calculates and account for 97% of whole PSTM program runtime, illustrate that PSTM calculating section is the performance bottleneck part of whole PSTM.
(2) parallel parsing of PSTM calculating section serial algorithm:
By analyzing PSTM calculating section serial algorithm, can parallel processing between seismic channel data, and in each seismic channel data, it will process L_NUM bar side line CMP_NUM the CMP point corresponding with every bar side line, and namely CMP to be dealt with counts as L_NUM*CMP_NUM, and each point can realize walking abreast, particularly, aforementioned body is biased in the example of image space, can walk abreast between 110000 track datas, also can walk abreast between 91*963 CMP point.。
(3) Parallel Design of PSTM calculating section serial algorithm:
Different seismic traces can be placed on CPU or MIC equipment, realizes the parallel of equipment room, and such as: CPU and MIC gets 100 track datas at every turn and carry out PSTM calculating, and who is complete, just takes off 100 track datas, CPU and MIC calculates and reaches dynamic load leveling.For each track data, CMP point according to each equipment the Thread Count that rises walk abreast.
Embodiment 1
The disposal route embodiment 1 of seismic pre-stack time migration of the present invention (PSTM), the method comprises the parallel N number of PSTM treatment scheme initiated, and as shown in Figure 1, each PSTM treatment scheme comprises the following steps:
FFT calculation procedure, reads the seismic channel data of pending batch and carries out FFT calculating, and the FFT result of calculation of each batch is alternately buffered into first or second buffer zone of free time corresponding to this PSTM treatment scheme;
The memory capacity of first, second buffer zone described is equal.
PSTM calculation procedure, alternately carries out PSTM calculating to each batch of FFT result of calculation in the first or second buffer zone corresponding to this PSTM treatment scheme.
Preferably, the system applying the method can comprise:
Central processing unit (CPU) end, described platform comprises cpu chip;
N-1 integrated many core (MIC) end; It is Intel Company's exploitation that this MIC end comprises MIC chip, for many core chips of high performance parallel computation.It develops from existing Xeon processor product basis, and it aims at very-high performance and calculates and raw new architecture.Formal product based on MIC framework is Xeon Phi.It is in computer architecture, not for replacing CPU, but exist as coprocessor.The x86 core that MIC chip has more than 50 to simplify usually, each core supports 4 hardware threads, and the number of tasks of executed in parallel can reach more than 200, provide the computing power of highly-parallel, its pair of smart peak performance reaches 1TFlops.
Utilize MIC can improve the calculated performance of PSTM, meet the demand of oil seismic exploration data process, and the method can reduce machine room construction cost and management, operation, maintenance cost.
Understandably, MIC end is held with CPU and is connected by connector.
Preferably, the FFT calculation procedure of N number for the treatment of scheme is held by CPU and is realized, PSTM calculation procedure in described N number for the treatment of scheme is held by this CPU respectively and N-1 MIC holds realization, when described PSTM calculation procedure is realized by MIC end, the FFT result of calculation asynchronous transmission in first, second buffer zone is held to MIC by described CPU end.
Have employed the technology of double buffering in this embodiment, effectively improving system effectiveness when effectively utilizing internal memory.
Known through above analysis, Fast Fourier Transform (FFT) (FFT) not consuming time can be calculated based on aforementioned system and be placed on the execution of CPU end, and executed in parallel while of the consuming time and extraordinary PSTM calculating section of concurrency being placed on CPU end and MIC end, CPU and MIC not only carries out cooperated computing, also calculate simultaneously, and the data between CPU and MIC are transmitted by asynchronous system, reach to calculate and hide with the mutual of IO, PSTM is accelerated in final realization, improve its handling property, thus meet the demand of oil seismic exploration data process, and reduce machine room construction cost and management, run, maintenance cost.
Particularly: CPU holds employing 2 pieces of CPU (16 core), start 16 OpenMP threads, (3 threads are FFT and calculate 16 threads, 13 threads are PSTM and calculate, wherein 2 threads utilize the very short time to carry out MIC scheduling) a parallel computation 91*963 CMP point, MIC holds employing 2 pieces of MIC (KNF), each MIC has 30 cores, starts 120 threads.120 thread parallels calculate 91*963 CMP point.
In order to make the object, technical solutions and advantages of the present invention more clear, below in conjunction with drawings and Examples, the present invention is described in detail below.
The PSTM of different seismic traces calculates and can be placed on CPU or MIC equipment, realizes the parallel PSTM overall logic structural design of equipment room; PSTM overall logic structure adopts 2 pieces of CPU+2 block MIC isomery cooperated computing CPU to hold and is responsible for process scheduling, FFT calculates, and participates in PSTM calculating.MIC end is only responsible for PSTM and is calculated, and CPU and MIC realizes cooperated computing and jointly calculates.CPU has held 3 calculation procedure (being equivalent to said PSTM treatment scheme above).Calculation procedure 1 is responsible for control CUP and is carried out the calculating that FFT and MIC1 carries out PSTM, and calculation procedure 2 is responsible for control CUP and is carried out the calculating that FFT and MIC2 carries out PSTM, and calculation procedure 3 is responsible for the calculating that control 2 pieces of CPU carry out FFT calculating and PSTM.Building-block of logic as shown in Figure 2.
Play two threads in each calculation procedure, open up 2 buffer zones, thread 1 is responsible for reading seismic channel data, and starts to carry out multiple tracks FFT calculating, and thread 2 is responsible for PSTM and is calculated.PSTM calculating calculates can walk abreast with read data, FFT, realizes mutually hiding, and the data transmission of CPU to MIC adopts asynchronous system, also can be hidden in PSTM calculating.
Particularly, thread 1 reads the 1st batch of 100 road seismic channel data, starts to carry out 100 road FFT and calculates, and result is left in buffer zone 1 after running through.While thread 2 starts to take 100 track datas to carry out PSTM calculating from buffer zone 1, thread 1 starts again to read 2 batches of 100 track datas, run through and start to carry out these 100 a collection of road FFT and calculate afterwards, and result is left in buffer zone 2, while thread 2 starts to take the 2nd batch data from buffer zone 2, PSTM calculates, thread 1 starts again to read the 3rd batch data, so analogize, calculate so PSTM calculates can walk abreast with read data, FFT, realize mutually hiding, and the data transmission of CPU to MIC adopts asynchronous system, also can be hidden in PSTM calculating.The asynchronous computing design of whole CPU and MIC as shown in Figure 3.
Embodiment 2
The disposal route of seismic pre-stack time migration of the present invention (PSTM), as shown in Figure 4, the method embodiment comprises:
Step 401: data allocation step, initiatively distribution or the application according to each PSTM treatment scheme distribute the seismic channel data of each PSTM treatment scheme process, terminate until seismic channel data is distributed;
Described data allocation step is held by described CPU and is realized.
Particularly, can on the basis of embodiment 1, after each process (PSTM treatment scheme) completes the FFT calculating of 100 track datas, just to the application of CPU end, CPU end will send out the first address position that application process needs 100 track datas got, and application process just reads 100 track datas according to this position.
Step 402: parallel initiate N number of PSTM treatment scheme, particularly, identical with embodiment 1 of this PSTM treatment scheme.
Closer, described FFT calculation procedure circulation performs until the seismic channel data process distributed terminates, and described PSTM calculation procedure circulation performs until process the FFT result of calculation in first, second buffer zone.
When PSTM calculation procedure is realized by MIC end, the FFT result of calculation asynchronous transmission in first, second buffer zone is held to MIC by described CPU end.
Step 403: aggregation step, gathers the PSTM result of calculation of each PSTM treatment scheme.
Preferably, described aggregation step is held by CPU and is realized.
In order to realize anterior-posterior approach, present invention also offers the disposal system embodiment of a kind of seismic pre-stack time migration (PSTM), as shown in Figure 5, this system embodiment comprises N number of parallel performance element, and each performance element comprises:
FFT computing module, for reading the seismic channel data of pending batch and carrying out FFT calculating, also for the FFT result of calculation of each batch alternately being buffered into first or second buffer zone of the free time of this performance element;
First buffer zone, the second buffer zone, for the FFT result of calculation of each batch of buffer memory;
PSTM computing module, for alternately carrying out PSTM calculating to the first buffer zone of this performance element, the FFT result of calculation of the second buffer zone.
This system comprises 1 central processing unit (CPU) end and (N-1) individual integrated many core (MIC) end, corresponding, wherein the FFT computing module of N number of performance element holds realization with first, second buffer zone by CPU, and the PSTM computing module of described N number of processing execution unit is held by this CPU respectively and (N-1) individual MIC end realizes.Preferably, N=3
Preferably, described system also comprises the asynchronous transmission module that CPU end realizes, when the PSTM computing module of described performance element is realized by MIC end, described asynchronous transmission module, for the PSTM computing module FFT result of calculation asynchronous transmission in first, second buffer zone of this performance element realized to MIC end.
Preferably, described system also comprises holds by CPU the data allocation unit realized, for active or the seismic channel data of distributing each performance element process according to the application of each performance element.
Described FFT computing module circulation performs until the seismic channel data process distributed terminates, and described PSTM computing module circulation performs until process the FFT result of calculation in first, second buffer zone.
Described system also comprises collection unit, is held realize, for gathering the seismic channel data after each performance element process by CPU.
Alternatively, every batch of seismic channel data comprises 100 roads, and the memory capacity of first, second buffer zone described is equal.
In order to verify the performance of the present invention program, to test 91 surveys line, 963 CMP (common midpoints on every bar survey line, Common Midpoint) point, inputting 110000 track datas, to carry out migration imaging be example, and under original CPU isomorphism system, the time that PSTM spends in single-threaded serial mode is for 76053s, and native system working time is 1075s, performance promotes greatly.
From technical scheme of the present invention, the present invention is by being divided into seismic pre-stack time migration FFT calculating section and PSTM calculating section, by finding that PSTM calculating section is very consuming time the working time of comparing two calculating sections, account for more than 90% of whole PSTM program, so PSTM calculating section is the performance bottleneck of PSTM, then the concurrency that PSTM calculates is analyzed, and design parallel mode, FFT calculating section is placed on CPU end and performs, PSTM calculating section adopts CPU and MIC parallel computation simultaneously, whole PSTM calculates and realizes asynchronous system, read seismic channel data, FFT calculates, PSTM calculates and the data of CPU to MIC are transmitted, can mutually hide.By test, utilize MIC greatly to accelerate the performance of PSTM, so not only meet the demand of oil seismic exploration process, and greatly reduce power consumption, decrease machine room construction cost and management, operation, maintenance cost, and this method realizes simple, the cost of development of needs is low.
Claims (10)
1. a disposal route for seismic pre-stack time migration (PSTM), is characterized in that, the method comprises the parallel N number of PSTM treatment scheme initiated, and each PSTM treatment scheme comprises the following steps:
FFT calculation procedure, reads the seismic channel data of pending batch and carries out FFT calculating, and the FFT result of calculation of each batch is alternately buffered into first or second buffer zone of free time corresponding to this PSTM treatment scheme;
PSTM calculation procedure, alternately carries out PSTM calculating to each batch of FFT result of calculation in the first or second buffer zone corresponding to this PSTM treatment scheme;
The system of application the method comprises 1 central processor CPU end and N-1 integrated many core MIC holds, wherein the FFT calculation procedure of N number for the treatment of scheme is held by CPU and is realized, PSTM calculation procedure in described N number for the treatment of scheme is held by this CPU respectively and N-1 MIC holds realization, when described PSTM calculation procedure is realized by MIC end, the FFT result of calculation asynchronous transmission in first, second buffer zone is held to MIC by described CPU end.
2. the method for claim 1, it is characterized in that: N is 3, what described CPU held enables 16 threads, wherein 3 threads are for performing the FFT calculation procedure of each treatment scheme, 13 threads are used for described PSTM calculation procedure, and first, second MIC end enables PSTM calculation procedure described in 120 thread execution respectively.
3. the method for claim 1, it is characterized in that: before described PSTM treatment scheme, the method also comprises data allocation step, and initiatively distribution or the application according to each PSTM treatment scheme distribute the seismic channel data of each PSTM treatment scheme process, terminate until seismic channel data is distributed.
4. method as claimed in claim 3, is characterized in that: described FFT calculation procedure circulation performs until the seismic channel data process distributed terminates, and described PSTM calculation procedure circulation performs until process the FFT result of calculation in first, second buffer zone.
5. method as claimed in claim 3, it is characterized in that, after described PSTM treatment scheme, the method also comprises aggregation step, the PSTM result of calculation of each PSTM treatment scheme is gathered.
6. method as claimed in claim 5, is characterized in that: described data allocation step and described aggregation step are held by CPU and realized.
7. a disposal system for seismic pre-stack time migration (PSTM), is characterized in that, this system comprises N number of parallel performance element, and each performance element comprises:
FFT computing module, for reading the seismic channel data of pending batch and carrying out FFT calculating, also for the FFT result of calculation of each batch alternately being buffered into first or second buffer zone of the free time of this performance element;
First buffer zone, the second buffer zone, for the FFT result of calculation of each batch of buffer memory;
PSTM computing module, for alternately carrying out PSTM calculating to the first buffer zone of this performance element, the FFT result of calculation of the second buffer zone;
This system comprises 1 central processor CPU end and N-1 integrated many core MIC holds, corresponding, wherein the FFT computing module of N number of performance element holds realization with first, second buffer zone by CPU, and the PSTM computing module of described N number of performance element is held by this CPU respectively and N-1 MIC holds realization; Described system also comprises the asynchronous transmission module that CPU end realizes, when the PSTM computing module of described performance element is realized by MIC end, described asynchronous transmission module, for the PSTM computing module FFT result of calculation asynchronous transmission in first, second buffer zone of this performance element realized to MIC end.
8. system as claimed in claim 7, is characterized in that: described system also comprises holds by CPU the data allocation unit realized, for active or the seismic channel data of distributing each performance element process according to the application of each performance element.
9. system as claimed in claim 7, is characterized in that: described FFT computing module circulation performs until the seismic channel data process distributed terminates, and described PSTM computing module circulation performs until process the FFT result of calculation in first, second buffer zone.
10. system as claimed in claim 7, is characterized in that: described system also comprises collection unit, is held realize, for gathering the seismic channel data after each performance element process by CPU.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210339234.XA CN102866423B (en) | 2012-09-13 | 2012-09-13 | Seismic prestack time migration processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210339234.XA CN102866423B (en) | 2012-09-13 | 2012-09-13 | Seismic prestack time migration processing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102866423A CN102866423A (en) | 2013-01-09 |
| CN102866423B true CN102866423B (en) | 2015-07-22 |
Family
ID=47445403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210339234.XA Active CN102866423B (en) | 2012-09-13 | 2012-09-13 | Seismic prestack time migration processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102866423B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106291688B (en) * | 2016-07-25 | 2019-02-15 | 中国石油天然气股份有限公司 | Method and device for processing post-stack seismic data |
| CN110927784B (en) * | 2019-11-29 | 2020-09-29 | 中国地质大学（武汉） | Frequency domain Gaussian beam Green function calculation method based on OpenMP |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102243321A (en) * | 2011-03-15 | 2011-11-16 | 浪潮(北京)电子信息产业有限公司 | Method and system for processing seismic pre-stack time migration |
| WO2011146161A1 (en) * | 2010-05-19 | 2011-11-24 | Exxonmobil Upstream Research Company | Method and system for checkpointing during simulations |
-
2012
- 2012-09-13 CN CN201210339234.XA patent/CN102866423B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011146161A1 (en) * | 2010-05-19 | 2011-11-24 | Exxonmobil Upstream Research Company | Method and system for checkpointing during simulations |
| CN102243321A (en) * | 2011-03-15 | 2011-11-16 | 浪潮(北京)电子信息产业有限公司 | Method and system for processing seismic pre-stack time migration |
Non-Patent Citations (1)
| Title |
|---|
| "芯革命新未来 英特尔开启MIC众核时代";中关村在线 范平;《中关村服务器在线》;20110901;正文第7段，第13-20段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102866423A (en) | 2013-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guo et al. | Gpu-accelerated subgraph enumeration on partitioned graphs | |
| CN102243321B (en) | Method and system for processing seismic pre-stack time migration | |
| Spafford et al. | The tradeoffs of fused memory hierarchies in heterogeneous computing architectures | |
| CN103049241B (en) | A kind of method improving CPU+GPU isomery device calculated performance | |
| Guo et al. | Coupled ensemble flow line advection and analysis | |
| CN102253919A (en) | Parallel numerical simulation method and system based on GPU and CPU cooperative operation | |
| CN106776455B (en) | Single-machine multi-GPU communication method and device | |
| Rojek et al. | Adaptation of fluid model EULAG to graphics processing unit architecture | |
| Dong et al. | Implementing molecular dynamics simulation on the Sunway TaihuLight system with heterogeneous many‐core processors | |
| CN109657197A (en) | A kind of pre-stack depth migration calculation method and system | |
| CN102866423B (en) | Seismic prestack time migration processing method and system | |
| Lei et al. | Optimization and acceleration of flow simulations for CFD on CPU/GPU architecture | |
| CN103076627A (en) | Smoothing optimization method of velocity model | |
| Hou et al. | Full tensor gravity gradiometry data inversion: Performance analysis of parallel computing algorithms | |
| CN105700998A (en) | Method and device for monitoring and analyzing performance of parallel programs | |
| Yoshimi et al. | An FPGA-based tightly coupled accelerator for data-intensive applications | |
| CN103294639A (en) | CPU+MIC mixed heterogeneous cluster system for achieving large-scale computing | |
| Ciznicki et al. | Elliptic solver performance evaluation on modern hardware architectures | |
| CN101963916A (en) | Compilation processing method and device | |
| CN103018776B (en) | A kind of method utilizing CPU/MIC cooperated computing to accelerate seismic pre-stack time migration | |
| Fan et al. | Model aggregation method for data parallelism in distributed real-time machine learning of smart sensing equipment | |
| CN104536938A (en) | Method and system for calculating earthquake pre-stack time migration | |
| Loghin et al. | A time–energy performance analysis of MapReduce on heterogeneous systems with GPUs | |
| Sane et al. | Scalable in situ computation of Lagrangian representations via local flow maps | |
| Qureshi et al. | Genome sequence alignment-design space exploration for optimal performance and energy architectures |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |