CN102750663A - Method, device and system for geographic information data processing based on graphics processing unit (GPU) - Google Patents
Method, device and system for geographic information data processing based on graphics processing unit (GPU) Download PDFInfo
- Publication number
- CN102750663A CN102750663A CN2011102495594A CN201110249559A CN102750663A CN 102750663 A CN102750663 A CN 102750663A CN 2011102495594 A CN2011102495594 A CN 2011102495594A CN 201110249559 A CN201110249559 A CN 201110249559A CN 102750663 A CN102750663 A CN 102750663A
- Authority
- CN
- China
- Prior art keywords
- precision floating
- point data
- geographic information
- data
- main body
- 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
Images
Landscapes
- Instructional Devices (AREA)
Abstract
An embodiment of the invention discloses a method, a device and a system for geographic information data processing based on a graphics processing unit (GPU). The method includes the following steps: receiving main body single-precision floating point data and error single-precision floating point data which are respectively obtained through double-precision floating point data dividing operated by a main machine processor; operating the main body single-precision floating point data and the error single-precision floating point data according to a preset operation rule and leading the operation result to generate unified single-precision floating point data; and processing geographic information data according to the unified single-precision floating point data. In the embodiment, the double-precision floating point data are divided by the main machine processor, the GPU operates the divided single-precision floating point data, the geographic information data are processed according to the operation result, geographic information data processing with the GPU is achieved, and processing efficiency is improved.
Description
Technical field
The present invention relates to the geographic information data process field, relate in particular to a kind of method, equipment and system that handles based on the geographic information data of GPU.
Background technology
In the GIS-Geographic Information System, the processing of geographic information data usually will be used GPU (Graphic Processing Unit, the graphic process unit) programming of video card.GPU is a notion with respect to CPU, because the processing of (particularly domestic system, the fan of recreation) figure becomes more and more important in the computing machine in modern times, needs a special graph core processor.GPU makes video card reduce the dependence to CPU, and carries out the part work of CPU originally, especially when graphic processing data, 3D graphics process.
The data volume of geographic information data is very big, and very complicated, especially the three-dimensional geographic information data.The three-dimensional geographic information data comprise complex spatial information, when the three-dimensional geographic information data are handled, will be referred to the double-precision floating point data.In the IEEE754 international standard, floating data commonly used has two kinds of forms:
(1) single-precision floating-point data (Single): be used for representing to have the real number of fraction part, the science that generally is used for is calculated.Take 4 bytes (32) storage space, comprise 1 of sign bit, 8 of exponents, 23 of mantissa.Its numerical range is 3.4E-38～3.4E+38, and single-precision floating-point data has 7 decimal system significant figure at most.
(2) double-precision floating point data (double): with 8 bytes (64) storage space, comprise 1 of sign bit, 11 of exponents, 52 of mantissa.Its numerical range is 1.7E-308～1.7E+308, and the double-precision floating point data have 15 or 16 decimal system significant figure at most.
In realizing process of the present invention, the inventor finds to exist in the prior art following shortcoming:
Existing a lot of video card is not supported the double-precision floating point data, only supports single-precision floating-point data, does not satisfy and handles the requirement of geographic information data to the data precision.
Summary of the invention
The embodiment of the invention provides a kind of method, equipment and system that handles based on the geographic information data of GPU; Through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result, the geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
The embodiment of the invention provides a kind of method of handling based on the geographic information data of GPU, may further comprise the steps:
Receive host-processor and split main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
Operation rule according to predetermined carries out computing to said main body single-precision floating-point data and said error single-precision floating-point data, generates main body single-precision floating-point data operation result and error single-precision floating-point data operation result;
Generate the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
Handle said geographic information data according to said normalizing single-precision floating-point data.
Said double-precision floating point data are 64 floating datas.
Also comprise:
Sign bit and rank that said main body single-precision floating-point data is said double-precision floating point data;
Said error single-precision floating-point data is the magnitude portion of said double-precision floating point data.
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
The embodiment of the invention provides a kind of equipment of handling based on the geographic information data of GPU, comprising:
The single-precision floating-point data receiving element is used to receive host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
The single-precision floating-point data arithmetic element; Be used for said main body single-precision floating-point data and said error single-precision floating-point data being carried out computing, generate main body single-precision floating-point data operation result and error single-precision floating-point data operation result according to predetermined operation rule;
Normalizing single-precision floating-point data generation unit generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
The geographic information data processing unit is used for handling said geographic information data according to said normalizing single-precision floating-point data.
Said double-precision floating point data are 64 floating datas.
Also comprise:
Sign bit and rank that said main body single-precision floating-point data is said double-precision floating point data;
Said error single-precision floating-point data is the magnitude portion of said double-precision floating point data.
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
The embodiment of the invention provides the embodiment of the invention that a kind of system that handles based on the geographic information data of GPU is provided, and comprising:
Single precision GPU comprises:
The single-precision floating-point data receiving element is used to receive host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
The single-precision floating-point data arithmetic element; Be used for said main body single-precision floating-point data and said error single-precision floating-point data being carried out computing, generate main body single-precision floating-point data operation result and error single-precision floating-point data operation result according to predetermined operation rule;
Normalizing single-precision floating-point data generation unit generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
The geographic information data processing unit is used for handling said geographic information data according to said normalizing single-precision floating-point data.
Host-processor comprises:
The geographic information data acquiring unit is used to obtain geographic information data, and said geographic information data comprises the double-precision floating point data;
Double-precision floating point data split cells is used to split said double-precision floating point data and obtains main body single-precision floating-point data and error single-precision floating-point data;
Data transmission unit is used for said main body single-precision floating-point data and said error single-precision floating-point data are sent to said single precision GPU.
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art; To do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below; Obviously, the accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills; Under the prerequisite of not paying creative work property, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is a kind of method flow diagram of handling based on the geographic information data of GPU in the embodiment of the invention;
Fig. 2 is a kind of method particular flow sheet of handling based on the geographic information data of GPU in the embodiment of the invention;
Fig. 3 is a kind of equipment structure chart of handling based on the geographic information data of GPU in the embodiment of the invention;
Fig. 4 is a kind of system construction drawing of handling based on the geographic information data of GPU in the embodiment of the invention.
Embodiment
To combine the accompanying drawing in the embodiment of the invention below, the technical scheme in the embodiment of the invention is carried out clear, intactly description, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills are not making the every other embodiment that is obtained under the creative work prerequisite, all belong to the scope of the present invention's protection.
The embodiment of the invention one provides a kind of method of handling based on the geographic information data of GPU, and is as shown in Figure 1, may further comprise the steps:
Step S101 receives host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain.
Step S102 carries out computing according to predetermined operation rule to said main body single-precision floating-point data and said error single-precision floating-point data, generates main body single-precision floating-point data operation result and error single-precision floating-point data operation result.
Step S103 generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data.
Step S104 handles said geographic information data according to said normalizing single-precision floating-point data.
Said double-precision floating point data are 64 floating datas.
Also comprise:
Sign bit and rank that said main body single-precision floating-point data is said double-precision floating point data;
Said error single-precision floating-point data is the magnitude portion of said double-precision floating point data.
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the GPU is handled, and has improved the efficient of handling.
The embodiment of the invention two provides a kind of method of handling based on the geographic information data of GPU, and is as shown in Figure 2, may further comprise the steps:
Step S201, host-processor split said double-precision floating point data and obtain main body single-precision floating-point data and error single-precision floating-point data, and said double-precision floating point data are 64 floating datas.
If the double-precision floating point data are d, two single-precision floating-point datas after then splitting are f1, f2, and wherein f1 is used for representing the sign bit and the rank of double-precision floating point data, the magnitude portion of f2 double-precision floating point data.Said host-processor is torn part concrete formula of said double-precision floating point data (1), (2) open:
fCpu1＝d (1)
fCpu2＝(d-f1)*1e6 (2)
Said host process sends to single precision GPU to split result.
Step S202 receives host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain.
Step S203; Operation rule according to predetermined carries out computing to said main body single-precision floating-point data and said error single-precision floating-point data, generates main body single-precision floating-point data operation result fResult.x and error single-precision floating-point data operation result fResult.y.
Step S204 generates normalizing single-precision floating-point data f according to said main body single-precision floating-point data operation result fResult.x and said error single-precision floating-point data fResult.y, like formula (3):
f＝fResult.x+fResult.y*1e-6 (3)
Step S205 handles said geographic information data according to said normalizing single-precision floating-point data f.
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
The embodiment of the invention three provides a kind of equipment of handling based on the geographic information data of GPU, and is as shown in Figure 3, comprising:
Single-precision floating-point data receiving element 301 is used to receive host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
Single-precision floating-point data arithmetic element 302; Be used for said main body single-precision floating-point data and said error single-precision floating-point data being carried out computing, generate main body single-precision floating-point data operation result and error single-precision floating-point data operation result according to predetermined operation rule;
Normalizing single-precision floating-point data generation unit 303 generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
Geographic information data processing unit 304 is used for handling said geographic information data according to said normalizing single-precision floating-point data.
Said double-precision floating point data are 64 floating datas.
Also comprise:
Sign bit and rank that said main body single-precision floating-point data is said double-precision floating point data;
Said error single-precision floating-point data is the magnitude portion of said double-precision floating point data.
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
The embodiment of the invention four provides a kind of system that handles based on the geographic information data of GPU, and is as shown in Figure 4, comprising:
Single-precision floating-point data receiving element 401 is used to receive host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
Single-precision floating-point data arithmetic element 402; Be used for said main body single-precision floating-point data and said error single-precision floating-point data being carried out computing, generate main body single-precision floating-point data operation result and error single-precision floating-point data operation result according to predetermined operation rule;
Normalizing single-precision floating-point data generation unit 403 generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
Geographic information data processing unit 404 is used for handling said geographic information data according to said normalizing single-precision floating-point data.
Host-processor 410 comprises:
Geographic information data acquiring unit 411 is used to obtain geographic information data, and said geographic information data comprises the double-precision floating point data;
Double-precision floating point data split cells 412 is used to split said double-precision floating point data and obtains main body single-precision floating-point data and error single-precision floating-point data;
The beneficial effect that the technical scheme of the embodiment of the invention is brought is following: through host-processor is split the double-precision floating point data; Single precision GPU carries out computing to the single-precision floating-point data after splitting; Handle geographic information data according to operation result; The geographic information data that is implemented under the single precision GPU is handled, and has improved the efficient of handling.
Through the description of above embodiment, those skilled in the art can be well understood to the present invention and can realize through hardware, also can realize by the mode that software adds necessary general hardware platform; Based on such understanding; Technical scheme of the present invention can be come out with the embodied of software product, this software product can be stored in a non-volatile memory medium (can be CD-ROM, USB flash disk; Portable hard drive etc.) in; Comprise some instructions with so that computer equipment (can be personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.
In a word, the above is merely preferred embodiment of the present invention, is not to be used to limit protection scope of the present invention.All within spirit of the present invention and principle, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (7)
1. method of handling based on the geographic information data of GPU, said geographic information data comprises the double-precision floating point data, it is characterized in that, may further comprise the steps:
Receive host-processor and split main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
Operation rule according to predetermined carries out computing to said main body single-precision floating-point data and said error single-precision floating-point data, generates main body single-precision floating-point data operation result and error single-precision floating-point data operation result;
Generate the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
Handle said geographic information data according to said normalizing single-precision floating-point data.
2. a kind of according to claim 1 method of handling based on the geographic information data of GPU is characterized in that said double-precision floating point data are 64 floating datas.
3. a kind of according to claim 1 method of handling based on the geographic information data of GPU is characterized in that, also comprises:
Sign bit and rank that said main body single-precision floating-point data is said double-precision floating point data;
Said error single-precision floating-point data is the magnitude portion of said double-precision floating point data.
4. an equipment of handling based on the geographic information data of GPU is characterized in that, comprising:
The single-precision floating-point data receiving element is used to receive host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
The single-precision floating-point data arithmetic element; Be used for said main body single-precision floating-point data and said error single-precision floating-point data being carried out computing, generate main body single-precision floating-point data operation result and error single-precision floating-point data operation result according to predetermined operation rule;
Normalizing single-precision floating-point data generation unit generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
The geographic information data processing unit is used for handling said geographic information data according to said normalizing single-precision floating-point data.
5. like the said a kind of equipment of handling based on the geographic information data of GPU of claim 4, it is characterized in that said double-precision floating point data are 64 floating datas.
6. like the said a kind of equipment of handling based on the geographic information data of GPU of claim 4, it is characterized in that, also comprise:
Sign bit and rank that said main body single-precision floating-point data is said double-precision floating point data;
Said error single-precision floating-point data is the magnitude portion of said double-precision floating point data.
7. a system that handles based on the geographic information data of GPU is characterized in that, comprising:
GPU comprises:
The single-precision floating-point data receiving element is used to receive host-processor and splits main body single-precision floating-point data and the error single-precision floating-point data that said double-precision floating point data obtain;
The single-precision floating-point data arithmetic element; Be used for said main body single-precision floating-point data and said error single-precision floating-point data being carried out computing, generate main body single-precision floating-point data operation result and error single-precision floating-point data operation result according to predetermined operation rule;
Normalizing single-precision floating-point data generation unit generates the normalizing single-precision floating-point data according to said main body single-precision floating-point data operation result and said error single-precision floating-point data;
The geographic information data processing unit is used for handling said geographic information data according to said normalizing single-precision floating-point data.
Host-processor comprises:
The geographic information data acquiring unit is used to obtain geographic information data, and said geographic information data comprises the double-precision floating point data;
Double-precision floating point data split cells is used to split said double-precision floating point data and obtains main body single-precision floating-point data and error single-precision floating-point data;
Data transmission unit is used for said main body single-precision floating-point data and said error single-precision floating-point data are sent to said GPU.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102495594A CN102750663A (en) | 2011-08-26 | 2011-08-26 | Method, device and system for geographic information data processing based on graphics processing unit (GPU) |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102495594A CN102750663A (en) | 2011-08-26 | 2011-08-26 | Method, device and system for geographic information data processing based on graphics processing unit (GPU) |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102750663A true CN102750663A (en) | 2012-10-24 |
Family
ID=47030814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011102495594A Pending CN102750663A (en) | 2011-08-26 | 2011-08-26 | Method, device and system for geographic information data processing based on graphics processing unit (GPU) |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102750663A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107124286A (en) * | 2016-02-24 | 2017-09-01 | 深圳市知穹科技有限公司 | A kind of mass data high speed processing, the system and method for interaction |
| CN110737470A (en) * | 2017-04-28 | 2020-01-31 | 英特尔公司 | Computational optimization of low-precision machine learning operations |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5487022A (en) * | 1994-03-08 | 1996-01-23 | Texas Instruments Incorporated | Normalization method for floating point numbers |
| CN101158892A (en) * | 2007-11-16 | 2008-04-09 | 中国科学院计算技术研究所 | Floating-point data converting device and method |
| CN101452571A (en) * | 2007-12-07 | 2009-06-10 | 辉达公司 | Graphical rendering pipeline having double precision fused multiply-add functional unit(dfma) |
| CN101866345A (en) * | 2009-01-30 | 2010-10-20 | 纳夫特克北美有限责任公司 | Be used to exchange the method and system of the location conten data under the different data format |
| CN102122240A (en) * | 2011-01-20 | 2011-07-13 | 东莞市泰斗微电子科技有限公司 | Data type conversion circuit |
-
2011
- 2011-08-26 CN CN2011102495594A patent/CN102750663A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5487022A (en) * | 1994-03-08 | 1996-01-23 | Texas Instruments Incorporated | Normalization method for floating point numbers |
| CN101158892A (en) * | 2007-11-16 | 2008-04-09 | 中国科学院计算技术研究所 | Floating-point data converting device and method |
| CN101452571A (en) * | 2007-12-07 | 2009-06-10 | 辉达公司 | Graphical rendering pipeline having double precision fused multiply-add functional unit(dfma) |
| CN101866345A (en) * | 2009-01-30 | 2010-10-20 | 纳夫特克北美有限责任公司 | Be used to exchange the method and system of the location conten data under the different data format |
| CN102122240A (en) * | 2011-01-20 | 2011-07-13 | 东莞市泰斗微电子科技有限公司 | Data type conversion circuit |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107124286A (en) * | 2016-02-24 | 2017-09-01 | 深圳市知穹科技有限公司 | A kind of mass data high speed processing, the system and method for interaction |
| CN110737470A (en) * | 2017-04-28 | 2020-01-31 | 英特尔公司 | Computational optimization of low-precision machine learning operations |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10073817B1 (en) | Optimized matrix multiplication using vector multiplication of interleaved matrix values | |
| US9146901B2 (en) | Vector floating point argument reduction | |
| US9268528B2 (en) | System and method for dynamically reducing power consumption of floating-point logic | |
| GB2586559A (en) | Enhanced low precision binary floating-point formatting | |
| EP4328744A3 (en) | Computer processor for higher precision computations using a mixed-precision decomposition of operations | |
| US10592208B2 (en) | Very low precision floating point representation for deep learning acceleration | |
| US9304898B2 (en) | Hardware-based array compression | |
| US11620105B2 (en) | Hybrid floating point representation for deep learning acceleration | |
| CN116700663B (en) | Floating point number processing method and device | |
| US9430190B2 (en) | Fused multiply add pipeline | |
| WO2022057502A1 (en) | Method and device for implementing dot product operation, electronic device, and storage medium | |
| CN102750663A (en) | Method, device and system for geographic information data processing based on graphics processing unit (GPU) | |
| Markstein | The new IEEE-754 standard for floating point arithmetic | |
| US10671347B2 (en) | Stochastic rounding floating-point multiply instruction using entropy from a register | |
| CN116700665B (en) | Method and device for determining floating point number square root reciprocal | |
| CN111313905A (en) | Floating point number conversion method and device | |
| CN113778523B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN110199255B (en) | Combining execution units to compute a single wide scalar result | |
| CN113805844B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN110427172A (en) | Floating number processing method, device, equipment and computer readable storage medium | |
| US9087398B2 (en) | System and method for compressing bounding box data and processor incorporating the same | |
| CN117370722A (en) | Matrix multiplication operation processing method in parallel computing hardware and related equipment | |
| Nehmeier et al. | On Implementing the IEEE Interval Standard P1788 | |
| US20130318138A1 (en) | Apparatus and method for performing decimal division | |
| CN115951858A (en) | Data processor, data processing method and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20121024 |
|
| RJ01 | Rejection of invention patent application after publication |