CN103150212B - The implementation method of Quantum mechanical calculation and device - Google Patents
The implementation method of Quantum mechanical calculation and device Download PDFInfo
- Publication number
- CN103150212B CN103150212B CN201110403559.5A CN201110403559A CN103150212B CN 103150212 B CN103150212 B CN 103150212B CN 201110403559 A CN201110403559 A CN 201110403559A CN 103150212 B CN103150212 B CN 103150212B
- Authority
- CN
- China
- Prior art keywords
- gpu
- cpu
- data
- calculative
- calculation
- 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
- 238000004364 calculation method Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 claims description 21
- 230000003750 conditioning effect Effects 0.000 claims description 7
- 238000012546 transfer Methods 0.000 abstract description 3
- 238000012800 visualization Methods 0.000 abstract description 3
- 230000005610 quantum mechanics Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Power Sources (AREA)
Abstract
The invention discloses a kind of implementation method and device of Quantum mechanical calculation, the method comprises: receive the calculation task that user is inputted by graphic interface; Data calculative in calculation task are distributed to graphic process unit GPU and central processor CPU; GPU and CPU, according to quantum-mechanical computing method, carries out Quantum mechanical calculation to the calculative data be assigned to separately.The present invention facilitates user's incoming task by providing visualization interface, and transfers to CPU and GPU jointly to complete the task of distribution, achieves the isomery of computing system, and effectively can improve the speed of calculating, quantum-mechanical numerical procedure is obtained effectively perfect.
Description
Technical field
The present invention relates to computer realm, and especially, relate to a kind of implementation method and device of Quantum mechanical calculation.
Background technology
At present, quantum mechanics (referred to as QM) computation requirement is very huge, and the relevant calculating of quantum mechanics is much all engaged in by the high-performance computer in TOP500 seniority among brothers and sisters.Quantum mechanics analog computation can solve the various problems that the basic subjects such as biology, physics, chemistry run into when theoretical research, can replace traditional experimental technique, make researchist understand the essential information of research object more efficiently, accurately.Along with improving constantly of computational complexity, people have found the method that much can promote calculated performance, wherein just comprise and adopt various accelerator to carry out the method accelerated.
In current quantum mechanics application system, some system operations speed is fast not, and some system operability are poor, and therefore, current provided Quantum mechanical calculation scheme is also perfect not, but, to this, at present effective solution is not yet proposed.
Summary of the invention
For the problem that correlation technique internediate quantum mechanics numerical procedure is perfect not, the present invention proposes a kind of implementation method and device of Quantum mechanical calculation, can user friendlyly operate, and improves the speed calculated.
Technical scheme of the present invention is achieved in that
According to an aspect of the present invention, a kind of implementation method of Quantum mechanical calculation is provided.
The method comprises: receive the calculation task that user is inputted by graphic interface; Data calculative in calculation task are distributed to graphic process unit GPU and central processor CPU; GPU and CPU, according to quantum-mechanical computing method, carries out Quantum mechanical calculation to the calculative data be assigned to separately.
Wherein, calculative data are distributed to GPU and CPU and comprises: according to the processing power of GPU and CPU, distributing respectively GPU and CPU needs data to be processed.
Further, calculative data are distributed to GPU and CPU to comprise: calculative data are distributed to GPU and CPU in batches.
The method can comprise further: carry out in the process of Quantum mechanical calculation at GPU and CPU, according to the situation of change of the processing power of GPU and CPU, and the calculative data volume that dynamic conditioning every batch distributes GPU and CPU.
Further, when distributing calculative data to GPU and CPU first, the method comprises further: the nominal processing power being obtained GPU by application programming interface API.
In addition, the quantity of CPU can be at least one; The quantity of GPU can be at least one.
According to a further aspect in the invention, a kind of implement device of Quantum mechanical calculation is provided.
This device comprises: at least one CPU; At least one GPU; Interface provides module, for providing graphic interface, and receives the calculation task that user inputted by graphic interface; Distribution module, for distributing to GPU and CPU by data calculative in calculation task; Further, GPU and CPU is used for according to quantum-mechanical computing method, carries out Quantum mechanical calculation to the calculative data be assigned to separately.
Further, distribution module is further used for the processing power according to GPU and CPU, and distributing respectively GPU and CPU needs data to be processed.
Further, distribution module is further used for calculative data to distribute to GPU and CPU in batches.
Further, distribution module is further used for carrying out in the process of Quantum mechanical calculation at GPU and CPU, according to the situation of change of the processing power of GPU and CPU, and the calculative data volume that dynamic conditioning every batch distributes GPU and CPU.
The present invention facilitates user's incoming task by providing visualization interface, and transfers to CPU and GPU jointly to complete the task of distribution, achieves the isomery of computing system, and effectively can improve the speed of calculating, quantum-mechanical numerical procedure is obtained effectively perfect.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the implementation method of Quantum mechanical calculation according to the embodiment of the present invention;
Fig. 2 is the block diagram of the implement device of Quantum mechanical calculation according to the embodiment of the present invention.
Embodiment
According to embodiments of the invention, provide a kind of implementation method of Quantum mechanical calculation.
As shown in Figure 1, comprise according to the implementation method of the Quantum mechanical calculation of the embodiment of the present invention:
Step S101, receives the calculation task that user is inputted by graphic interface;
Data calculative in calculation task are distributed to graphic process unit (GraphicProcessingUnit, referred to as GPU) and central processing unit (referred to as CPU) by step S103;
Step S105, GPU and CPU, according to quantum-mechanical computing method, carry out Quantum mechanical calculation to the calculative data be assigned to separately.
Wherein, when calculative data are distributed to GPU and CPU, according to the processing power of GPU and CPU, GPU and CPU can be distributed respectively and needs data to be processed.Because calculative data volume usual in Quantum mechanical calculation is comparatively large, therefore calculative data can be distributed to GPU and CPU in batches.
Preferably, because processor is in operational process, its performance there will be certain fluctuation, therefore, carry out in the process of Quantum mechanical calculation at GPU and CPU, can according to the situation of change of the processing power of GPU and CPU, the calculative data volume that dynamic conditioning every batch distributes GPU and CPU.This makes it possible to ensure all to distribute rational task amount to GPU and CPU in each time.
In addition, because the processing power of GPU can can't be shown intuitively as the processing power of CPU, so, when distributing calculative data to GPU and CPU first, the nominal processing power of GPU can be obtained by application programming interface API, thus understand the processing power of GPU, for the distribution of data to be calculated provides reference.
When calculating, the quantity of the CPU adopted can be at least one, and the quantity of the GPU of employing is similarly at least one, and CPU and the GPU quantity separately participating in calculating can adjust according to actual needs, if desired, whole CPU or GPU also can be stopped to carry out Quantum mechanical calculation.In addition, when the quantity of CPU is multiple, can data distribution be carried out according to the processing power of each CPU and distribute the dynamic conditioning of data volume, also these CPU can be considered as a whole, similar mode can be adopted equally to carry out data for GPU and distribute and sendout adjustment.
The present invention considers that GPU is a kind of accelerator, and calculates the performance of GPU compared to CPU, finds that GPU calculated performance improves 10 ~ 100 times in part quantum mechanical application.But in the application of some quantum mechanics, GPU cannot be used to accelerate, CPU can only be adopted to calculate.Therefore, most reasonable manner is the mode adopting CPU+GPU Heterogeneous Computing, allows the task that different processor process is good at separately.Therefore, such scheme effectively can accelerate the speed calculated, and raises the efficiency.
In addition, for most of researchist, in the face of the cluster operation interface of order line, usually the study of a period of time is needed skillfully could to grasp use-pattern, high performance computing machine can be used afterwards, and the application has visual use interface by providing, the experience of user can greatly be promoted, and facilitating user to operate computing machine.
Such as, in actual applications, the solution of the present invention can support the Quantum mechanical calculation software based on Terachem, supports CPU/GPU Heterogeneous Computing, supports that the GPU of CPU and the Nvidia of Intel/AMD (or ATI etc.) carries out Heterogeneous Computing simultaneously.
In addition, the present invention can provide the visualized graphs process interface based on B/S framework, and user of service only needs by browser, just calculation task can be submitted on the high-performance computer of isomery.
In concrete operation process, the solution of the present invention can support the job scheduling of quantum mechanics under isomerous environment, realize following functions: heterogeneous device resource automatic sensing, support that many heterogeneous device calculate, support the scheduling strategy such as FIFO, resource reservation, Backfill, preemptive type, support MPI/PVM, batch processing/interactive mode waits homework type, supports many queue managements, support program applying template customization and support abundant form book keeping operation function etc., will not enumerate herein.The present invention can also realize the portal website of Quantum mechanical calculation, is completed the post-processing function such as the pretreatment function such as application data preparation and data result collection, arrangement by Simu-pack graphical interfaces.
According to embodiments of the invention, additionally provide a kind of implement device of Quantum mechanical calculation.
As shown in Figure 2, comprise according to the implement device of the Quantum mechanical calculation of the embodiment of the present invention:
At least one CPU21;
At least one GPU22 (for clearly object, illustrate only a CPU and GPU in Fig. 2, in fact, the quantity of CPU and GPU can be multiple, illustrates no longer one by one herein);
Interface provides module 23, for providing graphic interface, and receives the calculation task that user inputted by graphic interface;
Distribution module 24, is connected to CPU21, GPU22 and interface provides module 23, for data calculative in calculation task are distributed to GPU22 and CPU21;
Further, GPU and CPU is used for according to quantum-mechanical computing method, carries out Quantum mechanical calculation to the calculative data be assigned to separately.
Distribution module 24 is further used for the processing power according to GPU and CPU, and distributing respectively GPU and CPU needs data to be processed.
Distribution module 24 is further used for calculative data to distribute to GPU and CPU in batches.
Distribution module 24 is further used for carrying out in the process of Quantum mechanical calculation at GPU and CPU, according to the situation of change of the processing power of GPU and CPU, and the calculative data volume that dynamic conditioning every batch distributes GPU and CPU.
To sum up, by means of technique scheme of the present invention, user's incoming task is facilitated by providing visualization interface, and transfer to CPU and GPU jointly to complete the task of distribution, achieve the isomery of computing system, and effectively can improve the speed of calculating, quantum-mechanical numerical procedure be obtained effectively perfect.
These are only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (4)
1. an implementation method for Quantum mechanical calculation, is characterized in that, comprising:
Receive the calculation task that user is inputted by graphic interface;
Data calculative in described calculation task are distributed to graphic process unit GPU and central processor CPU;
Described GPU and described CPU, according to quantum-mechanical computing method, carries out Quantum mechanical calculation to the described calculative data be assigned to separately;
Wherein, calculative data are distributed to GPU and CPU to comprise:
According to the processing power of described GPU and described CPU, described GPU and described CPU is distributed respectively and needs data to be processed, and, calculative data are distributed in batches described GPU and described CPU; And,
Carry out in the process of Quantum mechanical calculation at described GPU and described CPU, according to the situation of change of the processing power of described GPU and described CPU, the calculative data volume that dynamic conditioning every batch distributes described GPU and described CPU.
2. implementation method according to claim 1, is characterized in that, when distributing calculative data to described GPU and described CPU first, described method comprises further:
The nominal processing power of described GPU is obtained by application programming interface API.
3. the implementation method according to any one of claim 1, is characterized in that, the quantity of described CPU is at least one; The quantity of described GPU is at least one.
4. an implement device for Quantum mechanical calculation, is characterized in that, comprising:
At least one CPU;
At least one GPU;
Interface provides module, for providing graphic interface, and receives the calculation task that user inputted by described graphic interface;
Distribution module, for distributing to described GPU and described CPU by data calculative in described calculation task;
Further, described GPU and described CPU is used for according to quantum-mechanical computing method, carries out Quantum mechanical calculation to the described calculative data be assigned to separately;
Wherein, described distribution module is further used for the processing power according to described GPU and described CPU, and distributing respectively described GPU and described CPU needs data to be processed, and, calculative data are distributed in batches described GPU and described CPU; And,
Described distribution module is further used for carrying out in the process of Quantum mechanical calculation at described GPU and described CPU, according to the situation of change of the processing power of described GPU and described CPU, the calculative data volume that dynamic conditioning every batch distributes described GPU and described CPU.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110403559.5A CN103150212B (en) | 2011-12-06 | 2011-12-06 | The implementation method of Quantum mechanical calculation and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110403559.5A CN103150212B (en) | 2011-12-06 | 2011-12-06 | The implementation method of Quantum mechanical calculation and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103150212A CN103150212A (en) | 2013-06-12 |
| CN103150212B true CN103150212B (en) | 2016-04-06 |
Family
ID=48548310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110403559.5A Active CN103150212B (en) | 2011-12-06 | 2011-12-06 | The implementation method of Quantum mechanical calculation and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103150212B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10095981B1 (en) * | 2017-03-22 | 2018-10-09 | Accenture Global Solutions Limited | Multi-state quantum optimization engine |
| US10228972B2 (en) * | 2017-06-22 | 2019-03-12 | Banuba Limited | Computer systems and computer-implemented methods for dynamically adaptive distribution of workload between central processing unit(s) and graphics processing unit(s) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1753031A (en) * | 2005-11-10 | 2006-03-29 | 北京航空航天大学 | GPU based particle system |
| WO2007033362A3 (en) * | 2005-09-13 | 2007-09-27 | Luminescent Technologies Inc | Systems, masks, and methods for photolithography |
| CN101685530A (en) * | 2008-09-23 | 2010-03-31 | 中国科学院过程工程研究所 | Method for calculating particles on GPU by utilizing multi-body interaction model |
| CN101901042A (en) * | 2010-08-27 | 2010-12-01 | 上海交通大学 | Power consumption reduction method based on dynamic task migration technology in multi-GPU system |
-
2011
- 2011-12-06 CN CN201110403559.5A patent/CN103150212B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007033362A3 (en) * | 2005-09-13 | 2007-09-27 | Luminescent Technologies Inc | Systems, masks, and methods for photolithography |
| CN1753031A (en) * | 2005-11-10 | 2006-03-29 | 北京航空航天大学 | GPU based particle system |
| CN101685530A (en) * | 2008-09-23 | 2010-03-31 | 中国科学院过程工程研究所 | Method for calculating particles on GPU by utilizing multi-body interaction model |
| CN101901042A (en) * | 2010-08-27 | 2010-12-01 | 上海交通大学 | Power consumption reduction method based on dynamic task migration technology in multi-GPU system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103150212A (en) | 2013-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rao et al. | Hedging against uncertainty: A tale of internet data center operations under smart grid environment | |
| CN103049241B (en) | A kind of method improving CPU+GPU isomery device calculated performance | |
| CN107563955A (en) | A kind of parallel map dicing method and system based on GPU | |
| Shokripour et al. | New method for scheduling heterogeneous multi-installment systems | |
| CN102902592B (en) | A kind of subregion schedule management method of PC cluster resource | |
| Zidenberg et al. | MultiAmdahl: How should I divide my heterogenous chip? | |
| CN103685309A (en) | Asynchronous request queue model oriented to map visualization tile service | |
| CN103488531B (en) | A kind of software and hardware mixed real-time application dispatching method based on polycaryon processor and FPGA | |
| CN102880510A (en) | Parallel programming method oriented to data intensive application based on multiple data architecture centers | |
| CN105320565B (en) | A kind of computer scheduling of resource method for a variety of application software | |
| CN103150212B (en) | The implementation method of Quantum mechanical calculation and device | |
| CN108429784B (en) | An energy efficiency priority cloud resource allocation and scheduling method | |
| CN105245472B (en) | A kind of inheritance power supply information priorities dispatching method | |
| CN104468710A (en) | Mixed big data processing system and method | |
| CN103942103A (en) | Method and device for generating processor scheduling policy in multi-core system and scheduling system | |
| CN103150463A (en) | Method and device for realizing molecular dynamics calculation | |
| CN104536939A (en) | Method for configurable energy-saving dispatching of multi-core embedded cache | |
| CN107197013B (en) | Energy-saving system for enhancing cloud computing environment | |
| CN109710395A (en) | Parameter optimization control method, device and distributed computing system | |
| CN110377411B (en) | Distributed cloud-oriented workflow task scheduling method and system | |
| CN102968388B (en) | Data layout's method and device thereof | |
| CN102904964A (en) | Concurrent Processing Method of Mass Concentrator Based on Completion Port Model | |
| CN107479945A (en) | A kind of resources of virtual machine dispatching method and device | |
| CN202003354U (en) | Monte-Carlo simulation-computation clustering system for fault imaging | |
| CN109271240A (en) | A kind of process scheduling method based on multicore processing |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180112 Address after: North building, No. 1699 building 215300 Jiangsu Province, Yushan city Kunshan town Zuchongzhi Road Room 405 Patentee after: Zhongke controlled Information Industry Co., Ltd. Address before: 300384 Tianjin city Xiqing District Huayuan Industrial Zone (outer ring) Haitai Huake Street No. 15 1-3 Patentee before: Sugon Information Industry Co., Ltd. |
|
| TR01 | Transfer of patent right |