CN201035617Y - Graphics rendering system for running OpenGL application program - Google Patents
Graphics rendering system for running OpenGL application program Download PDFInfo
- Publication number
- CN201035617Y CN201035617Y CNU2007201489186U CN200720148918U CN201035617Y CN 201035617 Y CN201035617 Y CN 201035617Y CN U2007201489186 U CNU2007201489186 U CN U2007201489186U CN 200720148918 U CN200720148918 U CN 200720148918U CN 201035617 Y CN201035617 Y CN 201035617Y
- Authority
- CN
- China
- Prior art keywords
- instruction stream
- video card
- application program
- unit
- main control
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The utility model provides a graph rendering system for running an OpenGL application program. A master control machine is connected with a plurality of graph rendering machines through an Ethernet network. The master control machine comprises an acquisition device for acquiring the graph storage application program instruction flow; the graph rendering machine comprises an analysis device for analyzing the instruction flow, and a display card; the instruction flow arrives at a display card through the Ethernet network and the analysis device. Through the application of the utility model, and the dominance utilization of the high performance cluster and the arithmetic capability of a microcomputer, the rendering speed of the graph is improved.
Description
Technical field
The utility model relates to the technology of playing up of computer graphics, is meant a kind of graph rendering system of the OpenGL of operation application program especially.
Background technology
Computing machine is played up very big calculated amount in the process of graphing, when particularly running into complicated figure and need play up, and the very long bottleneck of realizing its function that often becomes of its render process.And at present at high-performance computing sector, performance based on the high performance computing system of microcomputer cluster, on the basis of existing network technology, meet or exceed the performance of the computing system under traditional SMP framework, therefore can utilize the high performance arithmetic capability of microcomputer cluster to solve the problems referred to above fully.; great majority are used OpenGL (open graphic library at present; Open Graphics Library) graphics application program still uses traditional standalone to calculate and play up; do not support " hyperchannel graphical distribution formula is played up "; make the advantage of microcomputer cluster high performance computation ability can't obtain performance at all, cause the speed of graph rendering slow.
The utility model content
The purpose of this utility model provides a kind of graph rendering system of the OpenGL of operation application program, improves the speed of computing machine render graphics.
A kind of graph rendering system that moves the OpenGL application program, main control computer connects a plurality of graph rendering machines by Ethernet;
Described main control computer comprises the deriving means of shape library application program instructions stream;
Describedly play up resolver and the video card that machine comprises instruction stream;
Described instruction stream arrives described video card by described deriving means by described Ethernet and described resolver.
The described deriving means of this system comprises instruction stream interception unit, buffer, instruction stream analytic unit and the instruction stream compression unit that is linked in sequence, and described instruction stream interception unit is connected with integrated virtual demonstration map manager in the described main control computer.
The described resolver of this system comprises instruction stream decompression unit, buffer and the instruction stream performance element that is linked in sequence; Described instruction stream decompression unit connects described instruction stream compression unit by described Ethernet, and described instruction stream performance element connects described video card.
The described machine of playing up of this system includes the video card driver module that is connected with described video card, and described video card is connected with display terminal.
The described display terminal of this system splices each plays up the part figure that machine is played up.
The described display terminal of this system is a projector.
Use technology of the present utility model, overcome existing most of OpenGL graphics application programs and on unit, calculated the deficiency of playing up limited ability, the high-performance calculation advantage that the utility model not only can make the OpenGL graphics application program utilize figure to troop improves self and calculates and play up ability; And can make things convenient for to such an extent that combine with existing OpenGL graphics application program, needn't make amendment to original program source code.
Description of drawings
Fig. 1 is the utility model embodiment deriving means synoptic diagram;
Fig. 2 is the utility model embodiment resolver synoptic diagram;
Fig. 3 is for the utility model embodiment main control computer and play up machine annexation synoptic diagram;
Fig. 4 is the utility model embodiment principle of work schematic flow sheet;
The planar rectangular of Fig. 5 for needing among the utility model embodiment to play up.
Embodiment
The utility model provides a kind of graph rendering system of the OpenGL of operation application program, and this system can be divided into deriving means, Ethernet, resolver.
Wherein, deriving means comprises instruction stream interception unit (101) as shown in Figure 1, instruction stream buffer (102), instruction stream analytic unit (103), instruction stream compression unit (104).
Wherein, because main control computer (301) has been gone up integrated integrated virtual demonstration map manager (VR MapperManager, Virtual Reality Mapper Manager) technology, instruction stream interception unit (101) can utilize this virtual demonstration map manager to realize application programming interface hook (API Hook, ApplicationProgramming Interface) technology, the API instruction stream of interception graphics application program; Described application programming interface Hook Technique generally includes revises the IAT table, revises the function gauge outfit, and replaces three kinds of dynamic link libraries etc.; Described instruction stream is some instructions, and instruction stream is the pictute of above-mentioned some instructions.
Instruction stream buffer (102): be used for temporarily depositing the instruction stream of interception, instruction stream is generally deposited according to the form of formation or storehouse in instruction stream buffer (102).
Instruction stream analytic unit (103): from " instruction stream buffer (102) ", take out instruction stream, analyzed according to passage and viewpoint, and shunt;
Described viewpoint is meant the direction of observing figure, promptly when observation point is arranged in the different position of three dimensions, and the figure of being seen.
Described passage can be regarded as the zone that figure is chosen, and promptly chooses a zone, then this zone is played up.
Specifically the classification what standard to carry out instruction stream according to is set up on their own by the designer, and instruction stream analytic unit (103) sends to instruction stream compression unit (104) with sorted instruction stream.
Instruction stream compression unit (104): the instruction stream compression unit (104) that is positioned at main control computer forms packet to the sorted instruction stream packing compression that receives, and prepares transmission.
Resolver comprises instruction stream decompression unit (202), instruction stream buffer (203), instruction stream performance element (201), can also add a video card driver module (204) as shown in Figure 2.
Instruction stream decompression unit (202), and in instruction stream buffer (203) and the above-mentioned main control computer can be with a kind of unit has so just reduced the cost of research and development, making.The instruction stream decompression unit of playing up in the machine (202) is used for the packet that receives is decompressed again; The compression and the inverse process each other that decompresses.
Instruction stream performance element (201): instruction stream is resolved to the drawing for order of video card identification and submits to video card driver module (204).
Video card driver module (204): can be the driver of a software, be used to drive video card, receive the drawing for order that instruction stream performance element (201) sends, drive video card and draw.
Video card: video card hardware is drawn corresponding figure according to the drawing for order of receiving.
Each is played up after part figure that the video card of machine (303) is responsible for drawing finishes, and is spliced by display terminal (304); For example projector can be finished each partly final splicing and projection of figure by adjusting parameters such as separately view field, color, brightness.
Can be known that by foregoing description the utility model mainly comprises deriving means and two parts of resolver, as shown in Figure 3, above-mentioned both connect by network; Wherein deriving means is installed on the main control computer (301), main control computer (301) is a high-performance computer, and operation OpenGL graphics application program, be responsible for each and play up the scheduling of machine (303), based on the OpenGL distribution technique, figure playing up on main control computer (301) can be assigned to a plurality of machines (303) of playing up and play up; Resolver is positioned on the machine of playing up (303), plays up normally high performance graphics workstation of machine (303), accepts the drafting instruction stream of main control computer (301) by network, and finishes drafting by resolver, and is presented on the display terminal (304).The network switch of Ethernet (302) is used to finish main control computer (301) and plays up connection between the machine (303).
As shown in Figure 4, the deriving means on the description main control computer (301) and principle of work and the flow process of playing up the resolver cooperating on the machine (303).
On main control computer (301), only need to adopt existing VR Mapper Manager technology just can on the basis of not changing, deriving means be operated on the main control computer (301) based on the graphics application program source code of OpenGL, VR Mapper Manager is OpenGL dynamic link library (dll, dynamic link library) in " mapping or agency " of application layer, in fact, the graphics application program of OpenGL remains and calls the OpenGL built-in function and finish the task of playing up, and still keep the OpenGL interface constant to the application program that is positioned at the upper strata, realized transparent to upper level applications.Like this, for the OpenGL application program, the VR Mapper Manager that use configures just is equal to the OpenGL API that uses through reconfiguring, the instruction stream of these OpenGL API Calls is by instruction stream interception unit (101) interception, and instruction stream interception unit (101) temporarily leaves the instruction stream of intercepting in the instruction stream buffer (102); Instruction stream analytic unit (103) extracts instruction stream from instruction stream buffer (102), and is analyzed, shunts according to passage and viewpoint, and the instruction stream after the shunting is admitted to instruction stream decompression unit (202); The compression of packing of instruction stream after instruction stream decompression unit (202) will be shunted, promptly instruction stream compression in network packet one by one, the zone bit of each packet has also identified the machine of playing up (303) that receives this packet, and can send above-mentioned network packet to main control computer (301) request by Ethernet by deriving means.
Main control computer (301) sends to above-mentioned packet the network switch (302) of Ethernet, the network switch (301) then sends to the corresponding machine of playing up (303) according to the sign in the packet with each packet, after playing up machine (303) reception, handle by resolver.
The instruction stream decompression unit (202) of resolver decompresses packet, and decompression is the inverse process of compression, so the design of this unit is identical with the instruction stream decompression unit (202) of deriving means; Instruction stream after the decompression leaves in the instruction stream buffer (203); Instruction stream performance element (201) extracts instruction stream from instruction stream buffer (203), according to instruction stream, resolve to the instruction that corresponding video card can be discerned, and send to video card driver module (204); Video card driver module (204) drives video card according to this instruction and plays up graph of a correspondence.
Specifically finish the task of graphing by video card, and be presented on the display terminal (304); As previously discussed, each display terminal (304) all demonstrates the figure that the corresponding machine of playing up is responsible for playing up, and a plurality ofly plays up parallel the carrying out that machine has just realized that multichannel is drawn, and obtains the figure that ultimate demand is played up.
With shown in Figure 5 be example to playing up of planar rectangular, be without loss of generality, this planar rectangular is divided into the A that is equal to the dotted line in scheming, B, three parts of C figure is played up respectively, the instruction stream that A is played up is acquired the device interception on main control computer (301), deposit buffer in, and by analysis, the packing back sends to target by network and plays up machine (303), after this resolver of playing up machine (303) receives, be buffered in the instruction stream buffer (203), instruction stream performance element (201) extracts instruction stream from instruction stream buffer (203), according to instruction stream, resolve to the instruction that corresponding video card can be discerned, and send to video card driver module (204); Video card driver module (204) is carried out above-mentioned instruction and is played up graph of a correspondence by video card, and is presented on the display terminal (304).
Use deriving means, resolver in the utility model, can change and need not make code level to former OpenGL application program when the OpenGL application program realizes multi-path distributed playing up, the integrated environment configuration is simple.Overcome existing most of OpenGL graphics application programs and on unit, calculated the deficiency of playing up limited ability, the high-performance calculation advantage that the utility model not only can make the OpenGL graphics application program utilize figure to troop, improve and self calculate and play up ability, and can make things convenient for to such an extent that combine with existing OpenGL graphics application program.
The above only is a preferred implementation of the present utility model; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the utility model principle; can also make some improvements and modifications, these improvements and modifications also should be considered as protection domain of the present utility model.
Claims (6)
1. a graph rendering system that moves the OpenGL application program is characterized in that, main control computer connects a plurality of graph rendering machines by Ethernet;
Described main control computer comprises the deriving means of shape library application program instructions stream;
Describedly play up resolver and the video card that machine comprises instruction stream;
Described instruction stream arrives described video card by described deriving means by described Ethernet and described resolver.
2. system according to claim 1, it is characterized in that, described deriving means comprises instruction stream interception unit, buffer, instruction stream analytic unit and the instruction stream compression unit that is linked in sequence, and described instruction stream interception unit is connected with integrated virtual demonstration map manager in the described main control computer.
3. system according to claim 1 is characterized in that, described resolver comprises instruction stream decompression unit, buffer and the instruction stream performance element that is linked in sequence; Described instruction stream decompression unit connects described instruction stream compression unit by described Ethernet, and described instruction stream performance element connects described video card.
4. according to claim 1 or 3 described systems, it is characterized in that the described machine of playing up includes the video card driver module that is connected with described video card, described video card is connected with display terminal.
5. system according to claim 4 is characterized in that, described display terminal splices each plays up the part figure that machine is played up.
6. system according to claim 5 is characterized in that described display terminal is a projector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2007201489186U CN201035617Y (en) | 2007-04-25 | 2007-04-25 | Graphics rendering system for running OpenGL application program |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2007201489186U CN201035617Y (en) | 2007-04-25 | 2007-04-25 | Graphics rendering system for running OpenGL application program |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201035617Y true CN201035617Y (en) | 2008-03-12 |
Family
ID=39196508
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNU2007201489186U Expired - Fee Related CN201035617Y (en) | 2007-04-25 | 2007-04-25 | Graphics rendering system for running OpenGL application program |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201035617Y (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101477704B (en) * | 2009-01-08 | 2011-11-16 | 浪潮电子信息产业股份有限公司 | Elastic construction process based on Deadline graphic workstation rendering |
| CN102419702A (en) * | 2010-09-27 | 2012-04-18 | 北京中娱在线网络科技有限公司 | Method and device for displaying stereoscopic picture |
| CN102609971A (en) * | 2012-01-11 | 2012-07-25 | 南京大学 | Quick rendering system using embedded GPU (Graphics Processing Unit) for realizing 3D-GIS (Three Dimensional-Geographic Information System) |
-
2007
- 2007-04-25 CN CNU2007201489186U patent/CN201035617Y/en not_active Expired - Fee Related
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101477704B (en) * | 2009-01-08 | 2011-11-16 | 浪潮电子信息产业股份有限公司 | Elastic construction process based on Deadline graphic workstation rendering |
| CN102419702A (en) * | 2010-09-27 | 2012-04-18 | 北京中娱在线网络科技有限公司 | Method and device for displaying stereoscopic picture |
| CN102419702B (en) * | 2010-09-27 | 2013-12-11 | 北京中娱在线网络科技有限公司 | Method and device for displaying stereoscopic picture |
| CN102609971A (en) * | 2012-01-11 | 2012-07-25 | 南京大学 | Quick rendering system using embedded GPU (Graphics Processing Unit) for realizing 3D-GIS (Three Dimensional-Geographic Information System) |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106406977B (en) | A kind of GPU vitualization realization system and method | |
| US10109099B2 (en) | Method and apparatus for efficient use of graphics processing resources in a virtualized execution enviornment | |
| CN103761139B (en) | General purpose computation virtualization implementation method based on dynamic library interception | |
| CN104246829B (en) | Coloring is patched up in graphics process | |
| US7796130B2 (en) | PC-based computing system employing multiple graphics processing units (GPUS) interfaced with the central processing unit (CPU) using a PC bus and a hardware hub, and parallelized according to the object division mode of parallel operation | |
| US7385607B2 (en) | Scalable shader architecture | |
| CN109478310B (en) | Multi-resolution deferred shading using a texel shader in a computing environment | |
| US8111260B2 (en) | Fast reconfiguration of graphics pipeline state | |
| CN100416540C (en) | Information processing system including processors and memory managing method used in the same system | |
| US7659901B2 (en) | Application program interface for programmable graphics pipeline | |
| CN105426191B (en) | user interface display processing method and device | |
| EP1594091B1 (en) | System and method for providing an enhanced graphics pipeline | |
| US8497865B2 (en) | Parallel graphics system employing multiple graphics processing pipelines with multiple graphics processing units (GPUS) and supporting an object division mode of parallel graphics processing using programmable pixel or vertex processing resources provided with the GPUS | |
| CN108776949A (en) | Device and method for the memory management in graphics processing context | |
| DE102019114970A1 (en) | ADVANCED IMMERSIVE MEDIA PIPELINE FOR CORRECTING ARTIFACTS AND CLARITY OF OBJECTS IN COMPUTING ENVIRONMENTS | |
| CN106383965A (en) | Three-dimensional numerical atmospheric visual support system | |
| WO2017014836A1 (en) | Distributed frame buffer for scalable parallel rendering | |
| US20230360321A1 (en) | Shader binding management in ray tracing | |
| CN102651142A (en) | Image rendering method and image rendering device | |
| CN106020929A (en) | System and method for supporting 3D application in virtual environment | |
| CN201035617Y (en) | Graphics rendering system for running OpenGL application program | |
| US20180288392A1 (en) | Adjusting display content based on user distance detection | |
| US11436696B2 (en) | Apparatus and method for provisioning virtualized multi-tile graphics processing hardware | |
| CN104065953A (en) | Method and device based on display apparatus multi-view visual display control | |
| CN102857533B (en) | A kind of long-distance interactive system based on cloud computing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: BEIJING SEASTARS TECHNOLOGY CO., LTD. Free format text: FORMER OWNER: BEIJING SAISIDA TECHNOLOGY CO., LTD. Effective date: 20110104 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20110104 Address after: 100029 Beijing City, Chaoyang District Madian Yumin Road A Chinese international science and Technology Exhibition Center No. 12 Block 14 layer Patentee after: BEIJING SEASTARS TECHNOLOGY CO., LTD. Address before: 100029 Beijing City, Chaoyang District Madian Yumin Road A Chinese international science and Technology Exhibition Center No. 12 Block 14 layer Patentee before: Beijing Saisida Technology Co., Ltd. |
|
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080312 Termination date: 20140425 |