CN112685172A - Data processing method and system - Google Patents
Data processing method and system Download PDFInfo
- Publication number
- CN112685172A CN112685172A CN202011383321.6A CN202011383321A CN112685172A CN 112685172 A CN112685172 A CN 112685172A CN 202011383321 A CN202011383321 A CN 202011383321A CN 112685172 A CN112685172 A CN 112685172A
- Authority
- CN
- China
- Prior art keywords
- target
- display card
- resource
- display
- image rendering
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 238000009877 rendering Methods 0.000 claims abstract description 182
- 238000000034 method Methods 0.000 claims abstract description 31
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 2
- 238000004891 communication Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Landscapes
- Image Generation (AREA)
Abstract
The disclosure provides a data processing method and a data processing system, relates to the technical field of electronic information, and can solve the problem that the utilization rate of display card resources is low when images are processed. The specific technical scheme is as follows: when the display card resource scheduling platform acquires an image rendering request, converting the image rendering request into an image rendering instruction according to rendering parameters contained in the image rendering request; determining target terminal equipment and a target display card capable of processing the image rendering instruction according to resource use conditions and configuration information corresponding to the display cards in the plurality of terminal equipment in the database; and sending the image rendering instruction to the terminal equipment corresponding to the target display card, and acquiring a result after the target display card is rendered. The method and the device are used for integrating and processing the display card resources.
Description
Technical Field
The present disclosure relates to the field of electronic information technologies, and in particular, to a data processing method and system.
Background
With the advancement of science and technology, users have higher and higher requirements on the performance of hardware image processing in image processing, and in order to meet the performance requirements of users, the display cards are updated more and more rapidly. When the hardware equipment of the user can not meet the image rendering capability required by the image to be processed, the hardware equipment of the user needs to be upgraded and replaced by the display card, so that the cost of the equipment is increased undoubtedly, and meanwhile, when the low-configuration display card is replaced, the low-configuration display card is idle, so that the resource of the low-configuration display card is wasted.
Disclosure of Invention
The embodiment of the disclosure provides a data processing method and system, which can solve the problem that in the prior art, when an image is processed, the utilization rate of a display card resource is low. The technical scheme is as follows:
according to a first aspect of the embodiments of the present disclosure, a data processing method is provided, where the method is applied to a display card resource centralized scheduling platform, and includes:
acquiring image rendering request information, wherein the image rendering request information is used for requesting to render a target image and comprises target rendering information;
according to a preset conversion rule, converting the image rendering request information into an image rendering instruction, wherein the instruction is used for instructing to render a target image according to the target rendering information;
according to target processing resources required by the image rendering instruction, determining target terminal equipment matched with the target processing resources and a target display card in the target terminal equipment in a plurality of terminal equipment of a target database, and sending the image rendering instruction to the target terminal equipment, wherein the target database comprises preset processing resources corresponding to the display card in each of the plurality of equipment;
and acquiring a processing result sent by the target terminal equipment, and acquiring a rendered target image after the processing result is subjected to integration processing, wherein the processing result is a result obtained after a target display card in the target terminal equipment processes the target image according to the image rendering instruction.
In one embodiment, before obtaining the image rendering request information, the method further comprises:
acquiring display card information corresponding to each display card in each device through a display card resource agent module corresponding to each device in the devices, wherein the display card information comprises configuration information and preset processing resources;
and according to the display card information corresponding to each display card, generating the target database after registering each display.
In one embodiment, the method for converting the image rendering request information into the image rendering instruction comprises the following steps:
analyzing the image rendering request information, and acquiring target rendering information in the image rendering request information according to an analysis result, wherein the target rendering information at least comprises: configuration information, special effect information and drawing parameters;
and converting the image rendering request information to generate an image rendering instruction through the target driver and the target rendering information.
In one embodiment, the method for determining at least one target graphics card matching the target processing resource in the target database includes:
traversing the target database, and determining whether preset processing resources of the independent display card are matched with target processing resources corresponding to the target rendering instruction;
when the preset processing resource corresponding to the independent display card in the plurality of display cards is matched with the target processing resource, determining the independent display card as the target display card;
when the preset processing resources corresponding to the individual display cards in the plurality of display cards are not matched with the target processing resources, acquiring the added preset processing resources of the N display cards in each individual device;
when the preset processing resource obtained by adding the N display cards in each single device is matched with the target processing resource, determining that the N display cards in the single device are the target display cards;
when the added preset processing resources of all the display cards in each single device are not matched with the target processing resources, the added preset processing resources of P display cards in M devices are obtained, and M is more than or equal to 2;
and when the preset processing resource obtained by adding the P display cards in the M devices is matched with the target processing resource, determining that the P display cards are the target display cards.
According to a second aspect of embodiments of the present disclosure, there is provided a data processing system, the system comprising: the system comprises a display card resource centralized scheduling platform and a plurality of terminal devices, wherein the display card resource centralized scheduling platform is connected with the plurality of devices, and the terminal devices comprise at least one display card;
the video card resource centralized scheduling platform is used for acquiring image rendering request information, and the image rendering request information is used for requesting to render a target image;
according to a preset conversion rule, converting the image rendering request information into an image rendering instruction, wherein the instruction is used for instructing to render a target image according to a preset index;
according to target processing resources required by the image rendering instruction, determining target terminal equipment matched with the target processing resources and a target display card in the terminal equipment in a target database, and sending the image rendering instruction to the target terminal equipment, wherein the target database comprises preset processing resources corresponding to each display card in the plurality of display cards;
acquiring a processing result sent by the target terminal equipment, and acquiring a rendered target image after integrating the processing result;
the target terminal device is used for receiving the image rendering instruction;
and according to the image rendering instruction, performing rendering processing on the target image through the target display card to obtain a processing result, and sending the processing result to the display card resource centralized scheduling platform.
In one embodiment, the terminal device in the system comprises a video card resource agent module which is respectively connected with each video card in the terminal device,
the display card resource agent module is used for establishing a target transmission link with the display card resource centralized scheduling platform;
acquiring the display card information of each display card in the terminal equipment, wherein the display card information comprises configuration information and preset processing resources;
and sending the display card information of each display card in the terminal equipment to the display card resource centralized scheduling platform through the target transmission link.
In one embodiment, the display card in the terminal equipment comprises a resource calculation module,
the resource calculation module is used for acquiring the current processing resource of the display card according to a preset acquisition rule;
judging whether the current processing resource of the display card is consistent with the preset processing resource of the display card;
updating the preset processing resource according to the current processing resource when the preset processing resource of the current processing resource is inconsistent;
and sending the updated preset processing resource to the video card resource agent module.
In one embodiment, the display card in the terminal device in the system comprises a display card memory, the display card memory is connected with the display card resource agent module,
the display card memory is used for acquiring a processing result sent by the target display card, wherein the processing result is a result obtained after the target display card processes the target image according to the image rendering instruction;
storing the processing result according to the target storage position information;
and sending the target storage position information to the resource agent module so that the display card resource agent module sends the processing result to the display card resource centralized scheduling platform.
In one embodiment, the system comprises a graphics card resource centralized scheduling platform for
Traversing the target database, and determining whether preset processing resources of the independent display card are matched with target processing resources corresponding to the target rendering instruction;
when the preset processing resource corresponding to the independent display card in the plurality of display cards is matched with the target processing resource, determining the independent display card as the target display card;
when the preset processing resources corresponding to the individual display cards in the plurality of display cards are not matched with the target processing resources, acquiring the added preset processing resources of the N display cards in each individual device;
when the preset processing resource obtained by adding the N display cards in each single device is matched with the target processing resource, determining that the N display cards in the single device are the target display cards;
when the added preset processing resources of all the display cards in each single device are not matched with the target processing resources, the added preset processing resources of P display cards in M devices are obtained, and M is more than or equal to 2;
and when the preset processing resource obtained by adding the P display cards in the M devices is matched with the target processing resource, determining that the P display cards are the target display cards.
In one embodiment, the display card resource centralized scheduling platform in the system is configured to determine a resource to be processed corresponding to each target display card according to the target processing resource and a preset processing resource corresponding to each target display card;
splitting the image rendering instruction according to the to-be-processed resource corresponding to each target display card to generate a sub-image rendering instruction corresponding to each target display card, wherein the sub-image rendering instruction is used for instructing to render the to-be-processed resource;
sequentially sending the sub-image rendering instruction to the terminal equipment corresponding to the target display card;
the display card resource agent module in the terminal equipment is used for receiving the sub-image rendering instruction, and the sub-image rendering instruction is matched with a target display card;
and sending the sub-image rendering instruction to the target display card. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 is a flowchart of a data processing method provided by an embodiment of the present disclosure;
fig. 1a is a schematic diagram of a logical layer structure in a data processing method according to an embodiment of the present disclosure;
FIG. 2 is a block diagram of a data processing system provided by an embodiment of the present disclosure;
FIG. 2a is a block diagram of a data processing system according to an embodiment of the present disclosure shown in FIG. 1.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
Example one
The embodiment of the present disclosure provides a data processing method, as shown in fig. 1, where the data processing method is applied to a display card resource centralized scheduling platform, and includes the following steps:
101. and acquiring image rendering request information.
The image rendering request information is for requesting a rendering of the target image.
The display card resource centralized scheduling platform in the method provided by the disclosure can be used as an independent rendering resource pool to provide image rendering capability for terminal equipment or a cloud server, or other equipment needing image rendering. In other examples, the multi-graphics card resource scheduling platform can also be directly integrated in a terminal device or a server to be used as a rendering resource.
Before processing the image rendering request, the method further includes registering the display cards in the plurality of terminal devices, and generating a target database according to a registration result:
acquiring display card information corresponding to each display card in each terminal device through a display card resource agent module of each terminal in the plurality of terminal devices, wherein the display card information comprises configuration information and preset processing resources;
and according to the display card information corresponding to each display card, generating the target database after registering each display.
Through the display card registration of a plurality of terminal equipment, the display card processing resources of the plurality of terminal equipment can be planned comprehensively, and the display card resource utilization efficiency in the terminal equipment is improved. If the processing resources required for rendering the target image cannot be met by the display card in the local device a, the processing resources can be retrieved from the target database and the device B capable of processing the target image can be found, so that the rendering task can be completed through the device a and the device B, idle display cards in the device a are avoided, and the rendering task of the target image can also be completed.
In a specific example, the process of registering the display cards in the plurality of terminal devices may include:
step one, the display card resource agent module initiates a display card resource registration request to a display card resource centralized scheduling platform based on the information of all the display card resources detected by the local equipment.
Each terminal device is provided with a display card resource agent module which is used for communicating a display card resource scheduling platform and display card resources to realize the communication between the display card resource scheduling platform and the display card resources, and meanwhile, the display card resource agent is also used for directly carrying out other business communication with the display card resource scheduling platform. The display card resource is provided by at least one display card in the terminal equipment.
In actual implementation, the graphics card resource agent in the terminal device may detect the graphics cards configured for the terminal device in advance, and initiate a registration request to the graphics card resource scheduling platform according to the detected graphics card information of all the graphics cards, where the registration request carries the graphics card information of each graphics card.
Specifically, the display card information of each display card includes: basic information and configuration information; wherein, the basic information includes but is not limited to: chip type, DAC information, BIOS information, adapter character string; configuration information includes, but is not limited to: the main frequency of the video card, the frequency of the video memory, the total available graphic memory, the special video memory, the system video memory and the shared system memory.
And step two, the display card resource centralized scheduling platform registers the display card according to the received display card information.
Specifically, the display card resource centralized scheduling platform allocates a display card number to the current display card to be registered according to the registration request information sent by the display card resource agent module, records the display card information of the corresponding display card in the local database, associates the display card information with the corresponding number,
meanwhile, heartbeat connection is kept between the display card resource centralized scheduling platform and the display card resource agent, and the resource use state of the corresponding display card is obtained in real time from the display card resource agent and recorded.
102. And converting the image rendering request information into an image rendering instruction according to a preset conversion rule.
The image rendering instructions in the method provided by the present disclosure are for instructing rendering of a target image according to target rendering information.
In the method provided by the present disclosure, converting the image rendering request information into an image rendering instruction includes:
analyzing the image rendering request information, and acquiring target rendering information in the image rendering request information according to an analysis result, wherein the target rendering information at least comprises: configuration information, special effect information and drawing parameters;
and converting the image rendering request information to generate an image rendering instruction through the target driver and the target rendering information.
Specifically, after the graphics card resource scheduling platform receives the image rendering request, the graphics card resource scheduling platform translates the image rendering request through a corresponding display driver in the operating system to obtain a plurality of image rendering instructions. After the image rendering instructions are sent to the display card, corresponding image rendering operation can be executed in the display card.
Referring to fig. 1a, the principle of performing conversion and translation on an image rendering request by a display driver is described by taking a Windows system as an example for performing 3D rendering. The Windows Display Driver Model (WDDM) is a display driver architecture supported by Windows.
Fig. 1a is a logic diagram of a conventional WDDM driver model, which calls display acceleration software, such as a related interface function of D3D RT (Direct 3D runtime), when an application detects a 3D rendering requirement, and sends a configuration parameter (configuration parameter for a graphics card), a special effect parameter (parameter related to a special effect), a drawing parameter, and the like related to rendering to D3D RT.
And D3D RT will interact with the user mode Display driver to obtain the Display card status and transmit the drawing instruction (the drawing instruction is generated according to the drawing parameter) to the user mode Display driver, the user mode Display driver will put the translated instruction into the memory after translating the received instruction, so as to facilitate the Display miniport driver to read the translated instruction,
meanwhile, the user mode Display driver notifies the D3D RT of completion of instruction translation, and at this time, the D3D RT notifies a Display miniport driver (Display miniport driver) to obtain the translated instruction in the memory through a Graphics kernel subsystem (Direct Graphics kernel subsystem or dxgkrnl. On the other hand, the D3D RT passes the configuration parameters and special effects parameters received from the application layer to the graphics kernel subsystem. The graphics kernel subsystem is a system file for the operating system to complete image display management, and participates in all image generation management of the system.
103. And according to the target processing resource required by the image rendering instruction, determining a target display card matched with the target processing resource in a target database, and sending the image rendering instruction to the target display card.
The target database comprises preset processing resources corresponding to each display card in each terminal device.
In the method provided by the disclosure, the display card resource scheduling platform keeps heartbeat connection with each display card resource agent module corresponding to each display card, so that the resource use condition of each display card is obtained from each display card resource agent, and simultaneously, the display card resource scheduling platform locally updates the resource use condition of each display card bound with the display card in real time, so that when an image rendering request is received, a distribution strategy of the image rendering instruction is determined according to the resource use conditions of all the current display cards.
And after the image distribution strategy is determined, the video card resource scheduling platform distributes corresponding image rendering instructions to video card resource agents corresponding to the video cards in the terminal equipment, and the video card resource agents distribute computing resources and the video memory to execute specific image rendering operation.
When a target display card matched with the target processing resource is determined in a target database, a display card resource scheduling platform sets a load balancing strategy to enable the target processing resource required by an image rendering instruction to be matched with the preset processing resource of the display card, and the specific steps can include:
traversing the target database, and determining whether preset processing resources of the independent display card are matched with target processing resources corresponding to the target rendering instruction;
when the preset processing resource corresponding to the independent display card in the plurality of display cards is matched with the target processing resource, determining the independent display card as the target display card;
when the preset processing resources corresponding to the individual display cards in the plurality of display cards are not matched with the target processing resources, acquiring the added preset processing resources of the N display cards in each individual device;
when the preset processing resource obtained by adding the N display cards in each single device is matched with the target processing resource, determining that the N display cards in the single device are the target display cards;
when the added preset processing resources of all the display cards in each single device are not matched with the target processing resources, the added preset processing resources of P display cards in M devices are obtained, and M is more than or equal to 2;
and when the preset processing resource obtained by adding the P display cards in the M devices is matched with the target processing resource, determining that the P display cards are the target display cards.
Specific examples are listed here for explanation: since the number of graphics cards used is inversely related to the ultimate processing efficiency, the fewer the number of sets split is actually the better. The resolution is generally carried out in the following manner:
firstly, judging whether a current image rendering request can be processed independently through one display card in a plurality of display cards, and if so, distributing the current image rendering request to the display card for processing;
if the independent display card can not process the image rendering request, judging whether the corresponding display card meets the processing requirements of the two parts after the image rendering request is split into the two parts (by grouping the image rendering instructions);
if yes, respectively distributing the two groups of split image rendering instructions to corresponding display cards;
if not, trying to split the current image rendering request into three groups, and judging whether three display cards capable of meeting the processing requirements of the current image rendering request exist;
if yes, determining the three display cards, and distributing corresponding image rendering instructions to the display cards; and so on until the image rendering instruction is processed.
Aiming at the scheme, the disclosure further provides an optimization scheme aiming at determining the target display card in the plurality of terminals:
specifically, if there are two or more than two display cards located on the same terminal device, the two or more than two display cards located on the same terminal device are preferentially selected to perform current image rendering. For example, it is determined that the current image rendering instruction is split into three groups, and the combinations of the graphics cards that can meet the processing are respectively: the display card 1, the display card 2 and the display card 7; or the display card 1, the display card 3 and the display card 7; and if the display card 1 and the display card 2 are positioned on the same terminal device, the first allocation scheme is preferentially selected.
According to the method, after the target display card and the allocation strategy are determined, all current image rendering instructions are split, so that the current image rendering instructions are split into multiple groups, and the display card allocated to each group for processing is determined.
By using a plurality of display cards in a centralized manner in the distribution method, the problem that the low-configuration display cards in the prior art need to be replaced or upgraded when the low-configuration display cards cannot process images can be solved, and the use efficiency of the display cards is improved.
Meanwhile, based on the selection strategy, the terminal equipment related to the selected display card combination can be ensured to be as few as possible, and therefore the image rendering efficiency can be improved.
In the method provided by the present disclosure, when sending an image rendering instruction to a target graphics card, the method includes:
determining a resource to be processed corresponding to each target display card according to the target processing resource and a preset processing resource corresponding to each target display card;
splitting the image rendering instruction according to the to-be-processed resource corresponding to each target display card to generate a sub-image rendering instruction corresponding to each target display card, wherein the sub-image rendering instruction is used for instructing to render the to-be-processed resource;
and sequentially sending a sub-image rendering instruction corresponding to each target display card to the terminal equipment corresponding to the target display card.
For example, the target image may be divided into M blocks according to the processing capability of the target graphics card, and each image block is the resource to be processed.
104. And acquiring a processing result sent by the target terminal equipment, and acquiring a rendered target image after integrating the processing result.
The processing result in the method provided by the present disclosure is a result obtained by the target terminal device after the target graphics card in the terminal device processes the target image according to the image rendering instruction.
After a display card in the terminal equipment in the method obtains an image rendering instruction, pixel information is calculated and processed through GPU (graphics processing Unit) calculation resources in the display card, a processing result (a pixel value and a display position on a screen) is written into a corresponding display memory, and storage position information is returned to a display card resource agent module; and the video card resource agent module reads a corresponding processing result from the video memory through the storage position information and sends the read processing result to the video card resource scheduling platform.
And the display card resource scheduling platform collects the processing results returned by the terminal equipment through the display card resource agent module to obtain a final output image.
And the display card resource scheduling platform returns the final output image to the equipment initiating the image rendering request.
The data processing method provided by the embodiment of the disclosure is applied to a display card resource scheduling platform, and when the display card resource scheduling platform acquires an image rendering request, the image rendering request is converted into an image rendering instruction according to rendering parameters contained in the image rendering request; determining target terminal equipment and a target display card capable of processing the image rendering instruction according to resource use conditions and configuration information corresponding to the display cards in the plurality of terminal equipment in the database; and sending the image rendering instruction to the terminal equipment corresponding to the target display card, and acquiring a result after the target display card is rendered.
Through the processing scheme provided by the disclosure, the display card resources of a plurality of terminal devices can be concentrated, and greater calculation and rendering capabilities are released through unified scheduling, so that the display card resource cost is greatly saved, the use efficiency of the display card resources is improved, and the resource waste of the display card is avoided.
Example two
Based on the data processing method described in the embodiment corresponding to fig. 1, the following is an embodiment of the system of the present disclosure, which may be used to execute an embodiment of the method of the present disclosure.
An embodiment of the present disclosure provides a data processing system, as shown in fig. 2, the data processing system 20 includes:
the system comprises a display card resource centralized scheduling platform 201 and a plurality of terminal devices 202, wherein the display card resource centralized scheduling platform 201 is connected with the plurality of terminal devices 202, the display card resource centralized scheduling platform 201 is connected with the plurality of devices 202, and the terminal devices 202 comprise at least one display card;
the video card resource centralized scheduling platform is used for acquiring image rendering request information, and the image rendering request information is used for requesting to render a target image;
according to a preset conversion rule, converting the image rendering request information into an image rendering instruction, wherein the instruction is used for instructing to render a target image according to a preset index;
according to target processing resources required by the image rendering instruction, determining target terminal equipment matched with the target processing resources and a target display card in the terminal equipment in a target database, and sending the image rendering instruction to the target terminal equipment, wherein the target database comprises preset processing resources corresponding to each display card in the plurality of display cards;
acquiring the processing result sent by the target display card, integrating the processing result, acquiring a rendered target image,
the target terminal device is used for receiving the image rendering instruction;
and according to the image rendering instruction, rendering the target image through the target display card, and sending a processing result to the display card resource centralized scheduling platform.
In an alternative embodiment, as shown in fig. 2a, the centralized scheduling system for multiple graphics cards provided by the present disclosure mainly includes: the system comprises a display card resource centralized scheduling platform and a plurality of devices, wherein each device comprises at least one display card which is used for providing display card resources for the display card resource centralized scheduling platform; the display card resource scheduling platform and the equipment can be connected through a PCIE bus. Specifically, each terminal device at least includes: the system comprises a display card resource agent module and display card resources; the display card resource agent module is used for communicating the display card resource scheduling platform with the display card resources to realize the communication between the display card resource scheduling platform and the display card, and is also used for directly carrying out other business communication with the display card resource scheduling platform; the display card resource comprises at least one display card.
In one embodiment, the terminal device 202 in the system includes a graphics card resource agent module, which is respectively connected with each graphics card in the terminal device,
the display card resource agent module is used for establishing a target transmission link with the display card resource centralized scheduling platform;
acquiring the display card information of each display card in the terminal equipment, wherein the display card information comprises configuration information and preset processing resources;
and sending the display card information of each display card in the terminal equipment to the display card resource centralized scheduling platform through the target transmission link.
In one embodiment, the graphics card in the terminal device 202 in the system includes a resource calculation module,
the resource calculation module is used for acquiring the current processing resource of the display card according to a preset acquisition rule;
judging whether the current processing resource of the display card is consistent with the preset processing resource of the display card;
updating the preset processing resource according to the current processing resource when the preset processing resource of the current processing resource is inconsistent;
and sending the updated preset processing resource to the video card resource agent module.
In one embodiment, the display card in the terminal device 202 in the system includes a display card memory, the display card memory is connected to the display card resource agent module,
the display card memory is used for acquiring a processing result sent by the target display card, wherein the processing result is a result obtained after the target display card processes the target image according to the image rendering instruction;
storing the processing result according to the target storage position information;
and sending the target storage position information to the resource agent module so that the display card resource agent module sends the processing result to the display card resource centralized scheduling platform.
In one embodiment, the system comprises a graphics card resource centralized scheduling platform for
Traversing the target database, and determining whether preset processing resources of the independent display card are matched with target processing resources corresponding to the target rendering instruction;
when the preset processing resource corresponding to the independent display card in the plurality of display cards is matched with the target processing resource, determining the independent display card as the target display card;
when the preset processing resources corresponding to the individual display cards in the plurality of display cards are not matched with the target processing resources, acquiring the added preset processing resources of the N display cards in each individual device;
when the preset processing resource obtained by adding the N display cards in each single device is matched with the target processing resource, determining that the N display cards in the single device are the target display cards;
when the added preset processing resources of all the display cards in each single device are not matched with the target processing resources, the added preset processing resources of P display cards in M devices are obtained, and M is more than or equal to 2;
and when the preset processing resource obtained by adding the P display cards in the M devices is matched with the target processing resource, determining that the P display cards are the target display cards.
In one embodiment, the display card resource centralized scheduling platform in the system is configured to determine a resource to be processed corresponding to each target display card according to the target processing resource and a preset processing resource corresponding to each target display card;
splitting the image rendering instruction according to the to-be-processed resource corresponding to each target display card to generate a sub-image rendering instruction corresponding to each target display card, wherein the sub-image rendering instruction is used for instructing to render the to-be-processed resource;
sequentially sending the sub-image rendering instruction to the terminal equipment corresponding to the target display card;
the display card resource agent module in the terminal equipment is used for receiving the sub-image rendering instruction, and the sub-image rendering instruction is matched with a target display card;
and sending the sub-image rendering instruction to the target display card.
The data processing system provided by the embodiment of the disclosure comprises a display card resource scheduling platform and a plurality of display cards, wherein when the display card resource scheduling platform acquires an image rendering request, the image rendering request is converted into an image rendering instruction according to parameters contained in the image rendering request; and determining a target display card capable of processing the image rendering instruction according to the resource use condition and the configuration information corresponding to the plurality of display cards in the database. And sending the image rendering instruction to the target display card, and acquiring a rendered result of the target display card.
Based on the data processing method described in the embodiment corresponding to fig. 1, an embodiment of the present disclosure further provides a computer-readable storage medium, for example, the non-transitory computer-readable storage medium may be a Read Only Memory (ROM), a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like. The storage medium stores computer instructions for executing the data processing method described in the embodiment corresponding to fig. 1, which is not described herein again.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
Claims (10)
1. A data processing system, comprising: the system comprises a display card resource centralized scheduling platform and a plurality of terminal devices, wherein the display card resource centralized scheduling platform is connected with the plurality of terminal devices, and the terminal devices comprise at least one display card;
the video card resource centralized scheduling platform is used for acquiring image rendering request information, and the image rendering request information is used for requesting to render a target image;
according to a preset conversion rule, converting the image rendering request information into an image rendering instruction, wherein the instruction is used for instructing to render a target image according to a preset index;
determining a target display card matched with the target processing resource in a target database according to the target processing resource required by the image rendering instruction, and sending the image rendering instruction to target terminal equipment corresponding to the target display card, wherein the target database comprises a preset processing resource corresponding to each display card in each terminal equipment;
acquiring a processing result sent by the target terminal equipment, and acquiring a rendered target image after integrating the processing result;
the target terminal device is used for receiving the image rendering instruction;
and according to the image rendering instruction, rendering the target image through the target display card, acquiring a processing result, and sending the processing result to the display card resource centralized scheduling platform.
2. The system of claim 1, wherein the terminal device comprises a graphics card resource agent module respectively connected with each graphics card in the terminal device,
the display card resource agent module is used for establishing a target transmission link with the display card resource centralized scheduling platform;
acquiring display card information of each display card in the terminal equipment, wherein the display card information comprises configuration information and preset processing resources;
and sending the display card information of each display card in the terminal equipment to the display card resource centralized scheduling platform through the target transmission link.
3. The system of claim 1, wherein the display card in the terminal device comprises a resource calculation module,
the resource calculation module is used for acquiring the current processing resource of the display card according to a preset acquisition rule;
judging whether the current processing resource of the display card is consistent with the preset processing resource of the display card;
when the current processing resource is inconsistent with the preset processing resource, updating the preset processing resource according to the current processing resource;
and sending the updated preset processing resource to the video card resource agent module.
4. The system according to claim 2, wherein the display card in the terminal device comprises a display card memory, the display card memory is connected with the display card resource agent module,
the graphics card memory is used for acquiring a processing result sent by the target graphics card, wherein the processing result is a result obtained after the target graphics card processes the target image according to the image rendering instruction;
storing the processing result according to the target storage position information;
and sending the target storage position information to the resource agent module so that the display card resource agent module sends the processing result to the display card resource centralized scheduling platform.
5. The system of claim 1, wherein the graphics card resource centralized scheduling platform is configured to perform centralized scheduling on graphics card resources
Traversing the target database, and determining whether preset processing resources of an individual display card are matched with target processing resources corresponding to the target rendering instruction;
when a preset processing resource corresponding to an individual display card in the plurality of display cards is matched with the target processing resource, determining the individual display card as the target display card;
when the preset processing resources corresponding to the individual display cards in the plurality of display cards are not matched with the target processing resources, acquiring the added preset processing resources of the N display cards in each individual device;
when the preset processing resource obtained by adding the N display cards in each individual device is matched with the target processing resource, determining that the N display cards in the individual device are target display cards;
when the added preset processing resources of all the display cards in each single device are not matched with the target processing resources, obtaining the added preset processing resources of P display cards in M devices, wherein M is more than or equal to 2;
and when the preset processing resource obtained by adding the P display cards in the M devices is matched with the target processing resource, determining that the P display cards are the target display cards.
6. The system according to claim 1, wherein a graphics card resource centralized scheduling platform in the system is configured to determine, according to the target processing resources and preset processing resources corresponding to each target graphics card, to-be-processed resources corresponding to each target graphics card;
according to the to-be-processed resource corresponding to each target display card, splitting the image rendering instruction to generate a sub-image rendering instruction corresponding to each target display card, wherein the sub-image rendering instruction is used for instructing to render the to-be-processed resource;
sequentially sending the sub-image rendering instruction to terminal equipment corresponding to the target display card;
the display card resource agent module in the terminal equipment is used for receiving the sub-image rendering instruction, and the sub-image rendering instruction is matched with a target display card;
and sending the sub-image rendering instruction to the target display card.
7. A data processing method is applied to a display card resource centralized scheduling platform, and comprises the following steps:
acquiring image rendering request information, wherein the image rendering request information is used for requesting to render a target image, and the rendering request information comprises target rendering information;
according to a preset conversion rule, converting the image rendering request information into an image rendering instruction, wherein the instruction is used for instructing to render a target image according to the target rendering information;
according to target processing resources required by the image rendering instruction, determining target terminal equipment matched with the target processing resources and a target display card in the target terminal equipment in a plurality of terminal equipment of a target database, and sending the image rendering instruction to the target terminal equipment, wherein the target database comprises preset processing resources corresponding to the display card in each of the plurality of equipment;
and acquiring a processing result sent by the target terminal equipment, and acquiring a rendered target image after the processing result is subjected to integration processing, wherein the processing result is a result obtained after a target display card in the target terminal equipment processes the target image according to the image rendering instruction.
8. The method of claim 7, wherein prior to obtaining image rendering request information, the method further comprises:
acquiring display card information corresponding to each display card in each device through a display card resource agent module corresponding to each device in the devices, wherein the display card information comprises configuration information and preset processing resources;
and according to the display card information corresponding to each display card, after registering each display, generating the target database.
9. The method of claim 7, wherein converting the image rendering request information into image rendering instructions comprises:
analyzing the image rendering request information, and acquiring target rendering information in the image rendering request information according to an analysis result, wherein the target rendering information at least comprises: configuration information, special effect information and drawing parameters;
and converting the image rendering request information to generate an image rendering instruction through a target driver and target rendering information.
10. The method of claim 8, wherein determining at least one target graphics card in the target database that matches the target processing resource comprises:
traversing the target database, and determining whether preset processing resources of an individual display card are matched with target processing resources corresponding to the target rendering instruction;
when a preset processing resource corresponding to an individual display card in the plurality of display cards is matched with the target processing resource, determining the individual display card as the target display card;
when the preset processing resources corresponding to the individual display cards in the plurality of display cards are not matched with the target processing resources, acquiring the added preset processing resources of the N display cards in each individual device;
when the preset processing resource obtained by adding the N display cards in each individual device is matched with the target processing resource, determining that the N display cards in the individual device are target display cards;
when the added preset processing resources of all the display cards in each single device are not matched with the target processing resources, obtaining the added preset processing resources of P display cards in M devices, wherein M is more than or equal to 2;
and when the preset processing resource obtained by adding the P display cards in the M devices is matched with the target processing resource, determining that the P display cards are the target display cards.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011383321.6A CN112685172A (en) | 2020-11-30 | 2020-11-30 | Data processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011383321.6A CN112685172A (en) | 2020-11-30 | 2020-11-30 | Data processing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112685172A true CN112685172A (en) | 2021-04-20 |
Family
ID=75447089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011383321.6A Pending CN112685172A (en) | 2020-11-30 | 2020-11-30 | Data processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112685172A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113436056A (en) * | 2021-07-21 | 2021-09-24 | 挂号网（杭州）科技有限公司 | Rendering method, rendering device, electronic equipment and storage medium |
-
2020
- 2020-11-30 CN CN202011383321.6A patent/CN112685172A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113436056A (en) * | 2021-07-21 | 2021-09-24 | 挂号网（杭州）科技有限公司 | Rendering method, rendering device, electronic equipment and storage medium |
| CN113436056B (en) * | 2021-07-21 | 2023-02-17 | 挂号网（杭州）科技有限公司 | Rendering method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110227259B (en) | Data processing method, device, server and system | |
| CN112488907A (en) | Data processing method and system | |
| CN112784989B (en) | Inference system, inference method, electronic device, and computer storage medium | |
| US9063793B2 (en) | Virtual server and virtual machine management method for supporting zero client by providing host interfaces from classified resource pools through emulation or direct connection modes | |
| CN108074210B (en) | Object acquisition system and method for cloud rendering | |
| CN108073350B (en) | Object storage system and method for cloud rendering | |
| CN111860853B (en) | Online prediction system, device, method and electronic device | |
| CN111343262B (en) | Distributed cluster login method, device, equipment and storage medium | |
| CN111159046B (en) | Test method, test device, electronic equipment, test system and storage medium | |
| CN111679911A (en) | Management method, device, equipment and medium for GPU (graphics processing Unit) card in cloud environment | |
| CN112256414A (en) | Method and system for connecting multiple computing storage engines | |
| CN112269628A (en) | Resource scheduling system and method | |
| CN115292020A (en) | Data processing method, device, equipment and medium | |
| CN108920274A (en) | Performance optimization and device for image processing server end | |
| CN116204307A (en) | Federal learning method and federal learning system compatible with different computing frameworks | |
| CN116860391A (en) | GPU computing power resource scheduling method, device, equipment and medium | |
| CN112685172A (en) | Data processing method and system | |
| CN114820272A (en) | Data interaction method and device, storage medium and electronic equipment | |
| CN107734050B (en) | Load machine distribution method, computing equipment and load machine distribution system | |
| CN112286688B (en) | Memory management and use method, device, equipment and medium | |
| CN114040189A (en) | Multimedia test method, device, storage medium and electronic equipment | |
| CN111654539B (en) | Cloud-protogenesis-based Internet of things operating system construction method and system and electronic equipment | |
| CN111813541B (en) | Task scheduling method, device, medium and equipment | |
| CN116339927B (en) | Equipment determining method, device, storage medium and electronic device | |
| US20110131271A1 (en) | Apparatus and method for allocating and releasing imaging device in virtualization system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |