675 results about "Graphic card" patented technology

The invention relates to the technical field of graphics card power consumption adjustment, and discloses a method, a device and equipment for dynamically adjusting the power consumption of a graphics card component. The method comprises the following steps: carrying out multi-channel data acquisition and working load characteristic analysis on operation parameters of a graphics card GPU core, a video memory and a power management unit to obtain a component temperature state evaluation matrix; executing component-level power consumption distribution based on multi-objective optimization to obtain a power consumption distribution scheme of each component; inputting the power consumption distribution scheme of each component into a power consumption-frequency-voltage mapping model, and calculating an optimal frequency voltage configuration parameter through a binary search algorithm to obtain a frequency voltage control instruction; performing multi-step prediction to obtain a module power consumption demand prediction value; according to the method, the frequency voltage combination with the highest performance efficiency can be found under the condition that the power consumption constraint is met, and accurate power consumption control is achieved.

The invention discloses an industrial personal computer and multi-graphics card collaborative parallel computation acceleration system, which relates to the technical field of industrial resource allocation and parallel computation, and comprises a resource monitoring and predicting module, a resource management module and a prediction type resource preparation module, the task splitting and collaborative execution module comprises a task splitting module, a cross-node collaborative module and a collaborative operation engine; the intelligent scheduling and dynamic resource allocation module comprises an intelligent scheduler, a dynamic resource allocation module and a conflict avoidance module. According to the method, the GPU video memory utilization rate, the core utilization rate, the temperature, the video memory fragment rate, the available video memory total amount, the CPU core total utilization rate, the load condition and the idle core number index are collected in real time through a resource monitoring module, and the video memory capacity, the GPU core occupancy rate and the CPU load requirement are predicted in advance before a task is submitted in combination with a gradient boosting decision tree and a neural network prediction model; and resources are reserved, so that the scheduling delay is remarkably reduced, and the scheduling hit rate is improved.

The invention discloses a processor, a graphics card, equipment and a resource allocation method and device, and belongs to the technical field of computer resource management. The processor comprises a thread group assembling unit, a thread group scheduling execution unit and at least two processing units, the thread group assembling unit is used for assembling the tasks into one or more thread groups based on the attribute information of the tasks in a task assembling stage; and allocating computing resources for one or more thread groups with the thread group granularity, wherein each thread group is allocated with a processing unit; the thread group scheduling execution unit is used for scheduling one or more thread groups to the processing unit corresponding to each thread group for execution according to a unified scheduling rule, and the unified scheduling rule is a scheduling rule shared by different types of tasks; and the at least two processing units are used for executing the scheduled thread groups. And the computing resources are allocated according to the thread group granularity, so that the execution efficiency of the processor is improved.

The invention relates to the technical field of graphics card performance monitoring, and discloses a graphics card component performance monitoring method, device and equipment, and the method comprises the steps: setting a temperature sensor array for a GPU chip, a video memory, a power management module and a radiator of a graphics card, and carrying out the temperature data collection and preprocessing, and obtaining a temperature distribution matrix; establishing a thermal network topological structure and performing finite element discretization calculation to obtain a thermal network state space equation; partial least square regression analysis is carried out to obtain m temperature control areas; the temperature state vector and the load state vector of the temperature control area are input into a mixed Q network structure, an epsilon-greedy strategy and a Q learning updating rule are executed, and a heat dissipation control parameter sequence is obtained.By means of the method, accurate calculation of a temperature coupling coefficient and design of a temperature decoupling controller are achieved, the problem of temperature coupling between assemblies is effectively solved, and the reliability of heat dissipation control is improved. And the heat dissipation control strategy can be adaptively adjusted to adapt to different working load conditions.

The invention discloses a real-time dynamic rendering method based on multi-layer ray tracing and cache collaboration, and relates to the technical field of three-dimensional rendering of an energy station, and the method comprises the steps: converting a scene grid of the energy station into a voxelized grid, generating a global distance field, and generating grid surface data at the same time; a mixed ray tracing mode including software ray tracing and hardware ray tracing is combined to process static and dynamic objects respectively. Through a mixed ray tracing mode, the advantages of software ray tracing and hardware ray tracing are combined, illumination changes in a dynamic scene can be processed in real time, the software ray tracing utilizes a global distance field to quickly respond to intersection query of rays and the scene, the method is suitable for complex static scenes, and the real-time performance of the scene is improved. According to hardware ray tracing, high-precision reflection and shadow of a dynamic object are processed through the hardware acceleration capacity of an RTX display card, the real-time performance and dynamic performance of rendering are improved, and the requirement of an energy field station scene for real-time dynamic rendering is met.

The invention discloses a multi-model dynamic image analysis system, and relates to the technical field of artificial intelligence. The system adopts a heterogeneous computing architecture and comprises a preprocessing model, a multi-modal analysis model, an anisotropic correction model, a large language model, a task knowledge base embedding model and a KV cache management system. An incongruous task reconstruction mechanism based on hardware isomerism is constructed, a large language reasoning model is nested for analysis, and efficient reuse of computing resources is achieved by using a cross-model KV cache optimization technology. The system is provided with multiple graphics cards and a double-channel NUMA, and a multi-mode large language model carries out unsupervised or weak supervised training on a target detection model in the operation of the system. According to the system, through a dynamic load scheduling strategy, a mechanism for reasoning by using a processor and a memory enhanced display card in a full-load mode is innovatively designed, and three-section real-time streaming enhanced output is achieved. According to the system, through a dynamic load scheduling strategy, a mechanism for reasoning by using a processor and a memory enhanced graphics card in a full-load mode is innovatively designed.

The invention relates to the technical field of computers, and provides a method, a device and equipment for switching multiple display cards under a domestic platform and a medium. The method comprises the following steps: responding to power-on of a domestic platform, initializing firmware, and then calling a UEFI configuration interface of the domestic platform to obtain configuration information of a target graphics card set by a user; judging whether the configuration information of the target graphics card set by the user is consistent with the configuration information of the default graphics card or not; and in response to inconsistency of the configuration information of the two, restarting the domestic platform to switch the configuration information of the default graphics card into the configuration information of the target graphics card, then initializing the target graphics card, starting an operating system and loading a driver corresponding to the target graphics card to complete graphics card switching. According to the scheme, the user can select the display card to be switched more conveniently and visually according to needs, an excellent man-machine interaction interface is provided, the user does not need to understand very difficultly, and the method for switching the multiple display cards under the domestic platform, which is simpler, more visual and easier to operate, is provided for the user.

The invention relates to a data processing method, a processor, a chip, a display card and electronic equipment, and relates to the field of intelligent computing, and the method comprises the following steps: reading a part of a first matrix from a local memory or a register of a computing unit in response to M tensor computing engines in a tensor computing engine cluster, reading the second matrix from the tensor memory, and executing matrix operation by the M tensor calculation engines according to part of the first matrix and the second matrix respectively to obtain operation results corresponding to the M tensor calculation engines; and the M tensor calculation engines respectively write operation results corresponding to the M tensor calculation engines into registers of the M tensor calculation engines. According to the embodiment of the invention, the computing power can be increased under the condition that the sizes and the bandwidths of the local memory and the register in the computing unit are not increased.

The invention discloses a three-dimensional rendering method and system based on digital twinning. The method comprises the steps of obtaining model parameters, constructing an initial model, rendering textures, separating and compressing data, loading the compressed data in parallel, blending an optimization model, and presenting the optimization model on a digital platform. According to the WebGL rendering method, by optimizing the number of model patches and the number of vertexes, multiplexing geometric data, adopting a unified texture map and compressing the model and the texture file, the occupation of WebGL rendering on a memory is reduced, and the rendering efficiency is improved. By utilizing the method and the device, the burden of the graphics card and the CPU is effectively reduced, and the frame rate and the rendering effect in a localized platform environment are improved.

The invention provides a text processing method and system based on a deep learning model attention mechanism, and the method comprises the steps: obtaining a target text inputted by a user, and the target text comprises characters inputted by the user; performing vectorization expression on the target text to obtain a target matrix; on the basis of a preset attention processing mechanism, attention calculation is carried out on the target matrix to obtain attention output, and the attention processing mechanism comprises quantitative dimension reduction processing and table look-up calculation processing; based on the attention output, an output text is obtained through calculation of a multi-layer perception mechanism, and the output text is output to the outside. According to the text processing method and system based on the deep learning model attention mechanism, the calculation overhead of Softmax can be remarkably reduced, calculation resources of graphics card parallel calculation can be released, and then the reasoning efficiency of the whole model is improved so as to accelerate text generation.

The invention relates to the technical field of performance testing, and discloses a performance testing method, device and equipment for a graphics card, and the method comprises the steps: carrying out the quaternion stage decoupling of a rendering pipeline of the graphics card, and obtaining a vertex shader decoupling unit, a geometric shader decoupling unit, a rasterization unit decoupling unit and a pixel shader decoupling unit; stage isolation performance tests are respectively carried out to obtain first performance test data of the vertex shader, second performance test data of the geometric shader, third performance test data of the rasterization unit and fourth performance test data of the pixel shader; the first performance test data, the second performance test data, the third performance test data and the fourth performance test data are comprehensively analyzed to obtain performance bottleneck positioning results of all stages of the rendering pipeline, the coupling states of all the stages of the rendering pipeline can be effectively separated, and accurate adjustment and control over the running states of all the stages are achieved.

The invention discloses a Monte Carlo rapid dose calculation method. The method comprises the following steps: loading simulation parameters; obtaining particle source phase space data of the radiotherapy accelerator; obtaining human body anatomical structure image data, and reconstructing a human body three-dimensional matrix; gridding the three-dimensional human body matrix; inputting Monte Carlo dose calculation parameters, and constructing a refined Monte Carlo dose calculation physical model; according to the method, the input data is dynamically compressed by utilizing the structured sparse characteristic of the NVIDIA video card, so that the effective utilization rate of the video memory is improved, and the continuous operation of a calculation task is ensured; the MIG instance management technology is combined with multi-dimensional scheduling and cross-instance collaboration, so that the GPU resource utilization rate is improved, and the advantages are obvious especially in a multi-task parallel scene. The GPU Monte Carlo computing system and the GPU Monte Carlo computing method work cooperatively, the hardware performance of the NVIDIA Amperee and the hardware performance of a graphics card of a next generation of architecture are brought into full play, the Monte Carlo dose computing efficiency is improved to a new level, the computing time is shortened, and the GPU Monte Carlo computing requirement of particle transport is powerfully supported.

The invention relates to the technical field of program control, in particular to a method and system for optimizing and improving the performance of a graphics card based on big data, and the method comprises the following steps: collecting shader compiling time data, initial video memory distribution mode parameters and pipeline state setting sequence logs in a starting stage of a target application program; according to the method, the shader compiling time data, the initial video memory allocation mode parameters and the pipeline state setting log in the application program starting stage are collected, and the multi-task GPU activity feature set is constructed, so that the dynamic requirements of different application starting behaviors on hardware resources can be quantified, and the accurate prediction of the resource occupation mode is realized. Task operation states are grouped based on numerical clustering, classification indexes are established, and a typical GPU resource demand time sequence profile is generated after feature data of tasks of the same kind are aggregated, so that a subsequent resource pre-allocation strategy can be adaptively adjusted according to a historical behavior mode, and the video memory bandwidth scrambling risk in a high-fluctuation scene is reduced.

The present invention provides an image processing method, device and system, the method comprising: determining a first image according to resolution information of a pre-collected source device, pre-collected cone modeling parameters and pre-collected sample images; performing deformation processing on the first image to obtain a cone calibration map and cone calibration parameters; performing geometric correction on the cone calibration map to obtain a second image and geometric correction parameters; based on the cone calibration parameters and geometric correction parameters, correcting a pre-acquired original video image to obtain a target video image. This technology is applied to the scene of image virtual simulation. This solution adopts a combination of software and hardware, and is suitable for situations where the performance of the source device is not high, or the graphics card is not sufficient to support fusion processing, or when the source device has confidentiality requirements, to ensure the stability of the image deformation processing process, and the frame processing delay meets the requirements, which can improve the interactive response speed.

The invention provides a cloud desktop 3D multi-screen display method, system and device and a medium. The method comprises the steps that a virtual graphics card is created in a server physical machine, a server virtual machine GPU drive type is configured, and a universal port and a plurality of stream ports are created; the server virtual machine creates a plurality of stream pushing processing threads according to the number of stream ports, the stream pushing processing threads send the display output size of a streaming media equipment screen connected with the client to the GPU driver, and the GPU driver obtains image data according to the display output size, encodes the image data and returns encoded code streams to the stream pushing processing threads; and the stream pushing processing thread sends the coded code stream to a dynamic library of a server physical machine through a stream port and sends the code stream to a client through the dynamic library, and the client processes the coded code stream and then performs 3D multi-screen display. According to the embodiment of the invention, the model selection range of the CPU and the GPU can be widened, and the safety is improved.

A graphics card lifting device comprises an elongated lifting member having a front end and a rear end. A groove track is provided between the two ends. A first portion and a second portion are provided at the front end and the rear end of the elongated lifting member respectively. The second portion is movable along the groove track. When one end of a graphics card is disposed between the front end and the rear end, a spacing between the front end and the rear end is adjusted, so that the one end of the graphics card is supported by the first portion and the second portion.

A graphics card including a circuit board module, a first heat dissipation fin, and a pair of fans is provided. The circuit board module includes a circuit board and a heat source. The circuit board has first to fourth sides surrounding the heat source. The first and second sides are opposite sides. The third and fourth sides are opposite sides. The first heat dissipation fin is in thermal contact with the heat source and has multiple channels communicating with the first to fourth sides. The fans disposed on the first and second sides respectively have first flow outlets facing the first heat dissipation fin and generate flows towards the first heat dissipation fin through the first flow outlets. The flows meet and squeeze in the channels to form turbulent flows and flow out of the graphics card through the third and fourth sides respectively. A computer host is also provided.

The invention discloses an automatic texture color mapping method for large-scale point cloud data. The automatic texture color mapping method comprises the following steps: firstly, realizing geometric alignment of geographic texture data and point cloud data; then realizing release of intra-core buffer by means of a mass point cloud space segmentation and retrieval management technology, and segmenting the mass point cloud into a plurality of regional point cloud blocks; carrying out filtering processing on the regional point cloud blocks to filter out discrete flying spots; secondly, considering correction constraints of different texture source information to realize seamless generation of geographic texture data; secondly, efficient point cloud preliminary color assignment is achieved by means of the spatial relation between geometric mapping and local aggregation of regional point cloud blocks and geographic texture data; secondly, eliminating the texture information (color) of the wrong coloring point cloud by using the self-shielding judgment result of the point cloud; and finally, realizing spatial merging of regional point cloud block data sets by using an efficient out-of-core file merging technology. According to the method, the dependence on hardware equipment (such as a large memory and a high graphics card) is avoided, and the block processing of massive point clouds is efficiently realized.

The invention provides a graph card-based hybrid scheduling rendering method and system, and the method comprises the steps: dividing a frame buffer area, binding the divided frame buffer area with a task, completing the rendering of each card through a hybrid task scheduling strategy, and finally synthesizing and outputting a complete graph; a frame buffer area of a rendering graph is divided into a plurality of graph cards according to display criticality, each graph card is bound with an independent rendering task and a video memory area, a unique task division and resource allocation mode is formed, and the mode is different from a traditional overall rendering mode; and in combination with a hybrid scheduling strategy of high-priority task preemptive scheduling and dynamic card task parallel rendering, the problems of rendering task priority conflict and resource utilization are effectively solved.

Disclosed in the present application are resource configuration methods, a resource configuration system, an electronic device and a storage medium. A resource configuration method comprises: acquiring configuration requirement information corresponding to a computing power service; on the basis of the configuration requirement information, performing resource configuration on a general-purpose server and a graphics card resource pool to obtain a configuration result, wherein the general-purpose server is obtained on the basis of processor resource configuration, and the graphics card resource pool comprises a plurality of graphics processing units that support a hot-swapping function; and on the basis of the configuration result, providing a first instance for the computing power service, wherein the first instance comprises a processor resource allocated by the general-purpose server and a processor resource allocated by the graphics card resource pool. The present application solves the technical problems in the related art of low resource utilization rate and poor flexibility of the entire unit caused by heterogeneous models being constrained by the fixed configuration of processor resources of the entire unit.

The invention relates to the technical field of material detection, and discloses a graphics card welding spot pseudo soldering automatic detection method based on image recognition, and the method comprises the steps: firstly comparing a graphics card visible light image after gray level normalization with an infrared thermal distribution diagram after noise suppression, and obtaining the surface texture and thermodynamic characteristics of a welding spot; and then splicing the two to obtain an associated feature weight matrix, extracting texture and thermodynamic feature vectors of a welding spot region from the associated feature weight matrix, and splicing the texture and thermodynamic feature vectors into a fusion feature matrix according to channel dimensions. And finally, positioning pseudo soldering spot coordinates in combination with the standard soldering spot feature library, and generating a graphics card soldering spot pseudo soldering defect report according to the coordinates and the texture fracture index. According to the process, through multi-feature fusion and comparison, pseudo soldering accurate detection and report generation are achieved; according to the method, the accuracy and the efficiency of insufficient solder detection of the graphics card welding spots based on image recognition can be improved, and a high-precision automatic solution is provided for quality detection of the graphics card welding spots.

The invention relates to the technical field of video card power management, and discloses a power management method, device and equipment for a video card and a storage medium. The method comprises the following steps: collecting time sequence data of SM unit frequency of a graphics card, GDDR video memory bandwidth and VRM output voltage in real time, and establishing a GPU power consumption topology model; based on the GPU power consumption topology model, extracting a time sequence feature vector of the load change of the rendering pipeline; constructing a sparse representation matrix based on the time sequence feature vector; performing power consumption gradient threshold judgment and delay type classification by using the sparse representation matrix, and identifying a delay fault type in the current graphics card power management system; and adjusting a switching parameter of the GPU, a VRM compensation current and a fan rotating speed according to the identified delay fault type, and updating a delay detection threshold value and a repair strategy strength parameter. According to the invention, three different delay fault types of frequency climbing, voltage regulation and temperature control current limiting can be accurately distinguished, and continuous optimization of power management performance is realized.

The invention discloses a video card information monitoring system, a server, a monitoring method and a storage medium. Comprising a graphics card and a substrate management controller, the display card comprises a command register, a data register and a processor; the baseboard management controller generates a request instruction containing a first type of operation codes and at least one parameter according to a monitoring demand, sends the request instruction to the command register, reads a status bit from the command register, and reads data from the data register or writes the data according to the request instruction; the command register receives a request instruction sent by the substrate management controller, stores an operation code, a parameter and a status bit, and feeds the status bit back to the substrate management controller; the processor analyzes and executes the request instruction in the command register, and updates the status bit in the command register; the data register stores data written by the baseboard management controller or the processor and allows the baseboard management controller or the processor to read the data, and the problem that the number of monitorable information is limited due to I2C address limitation is solved.

A computer case with a graphics card holder is provided, which includes a computer case body and a graphics card holder. The graphics card holder includes a first support member, a second support member, and a third support member. The first support member is slidably connected to a fixing member, the second support member is vertically provided at one end of the first support member, the second support member is provided with a graphics card adapter cable. One side of the third support member away from the first support member is fixedly connected to the computer case body. By setting the graphics card holder and fixing component, the computer case can stably fix the graphics card. The setting of the graphics card adapter cable allows the graphics card to be placed vertically, thereby enhancing the heat dissipation performance of the graphics card.

The invention discloses a computer host internal heat dissipation structure which comprises a case, a mainboard part in the case and a CPU part and a display card part on the mainboard part, a heat insulation plate is installed at the top of the case, a semiconductor chilling plate is arranged at the end of the case, the cold end of the case faces inwards, the hot end of the case faces upwards, and the cold end of the chilling plate is provided with a double-circulation shunt heat dissipation structure to guide air exchange between the interior and the exterior of the case; the heat dissipation structure is simple in design, internal heat is efficiently absorbed and discharged in a double-circulation mode, the temperature of a case is reduced, stable operation of a system is ensured, centralized cooling is achieved through optimized pipeline design, heat loss is effectively isolated through a heat preservation interlayer, the overall heat dissipation efficiency is improved, in addition, a humidity isolation structure and a waterproof breathable film provide additional waterproof guarantee, and the service life is prolonged. According to the computer cooling system, condensed water drops are prevented from flowing out in the cooling process, safe operation of the system in the environment is ensured, electronic components are prevented from being damaged, the efficiency, safety and reliability of the cooling system are jointly improved, and stable and efficient operation of the computer host in various working environments is ensured.

1. The name of this design product: server channel fan frame. 2. Purpose of this design product: It is used to be installed on the chassis to dissipate heat for the motherboard, graphics card and hard disk inside the chassis. 3. The key point of the design of this product lies in its shape. 4. The picture or photo that best illustrates the design points: Stereoscopic drawing 1.

The invention discloses a graphics card optimization method and system based on intelligent rendering acceleration, and relates to the technical field of graphics card intelligent rendering. Physical topology of an input scene is analyzed, and an acting force network and a causal event chain between objects are constructed; extracting a spatial-temporal characteristic matrix based on the physical topology of the input scene, dividing the input scene into a static region and a dynamic region, and binding cross-scale physical attributes based on the spatial-temporal characteristic matrix to generate layered scene fingerprints; performing historical fingerprint matching on the layered scene fingerprints, and performing static region multiplexing and dynamic region light path propagation prediction based on a matching result to generate a primary prediction rendering frame; analyzing the primary rendering frame based on a causal event chain to identify an abnormal frame and a compliance frame; performing grading processing on the abnormal frame and the predicted rendering frame to obtain a rendering result, and dynamically updating a physical rule base and a causal event chain according to the rendering result; by analyzing the scene physical topology and the causal event chain, intelligent acceleration of the graphics card rendering process is realized.

The invention discloses a K8s cluster large model reasoning GPU sharing adaptive dynamic scheduling method, belongs to the field of artificial intelligence, can adaptively configure GPU resources required by large model reasoning, and improves the correctness and use efficiency of GPU configuration in a cluster. According to the adaptive method, GPU resources in the cluster can be correctly configured according to the model type, the reasoning or accelerated reasoning mode, the model parameter quantity and the precision; in addition, the to-be-reasoned task quantity and the reasoning service state are detected in real time, a regression model is designed to predict request processing time according to the request length, the graphics card computing power, the graphics card utilization rate and historical request response time, and according to the regression model, when the request flow is too large or hardware fails, reasoning examples are additionally deployed, and the reasoning efficiency is improved. According to the method, the task processing throughput and the stability and reliability of the system are improved, and when the request quantity is sharply reduced, the deployment reasoning instances are reduced, so that GPU resources are saved, and the use efficiency of the GPU in the system is further improved.

The utility model relates to the technical field of computer graphics card testing, in particular to a computer graphics card plugging and unplugging testing device which comprises a base, and a limiting block is installed in the center of the top of the base. When one end of the connecting rod rotates along the outside of the rotating disc, the movable push rod is driven to reciprocate on the limiting block so as to drive the computer graphics cards to be repeatedly inserted and pulled for testing, and the adjusting bolt is rotated to drive the pressing plate to descend to limit and fix the computer graphics cards of different sizes. When a triangular block moves, a reset spring is synchronously extruded, and a limiting rod is driven to move and be connected into a limiting hole, so that sockets of different sizes can be conveniently replaced, and the problem that a traditional graphics card insertion and extraction test device is inconvenient to test computer graphics cards of different sizes and models and is low in practicability is effectively solved.

The invention relates to the technical field of graphics card resource scheduling, and discloses a graphics card resource load balancing control method and system. The method comprises the following steps: constructing a sample set containing actual GPU video memory occupation and utilization rate; constructing a GPU video memory prediction function and a GPU utilization rate prediction function; the schedulability of the current task is judged through a soft constraint matching mechanism; gPU target allocation for executing the task; and carrying out dynamic scheduling by adopting a multi-stage KV cache block recovery mechanism. In the prior art, only static configuration or a single index is adopted to carry out GPU load estimation, and especially in the scene of large Token length uncertainty and obvious KV cache dynamic expansion in an LLM class model reasoning stage, accurate video memory resource scheduling control is difficult to realize. Due to the fact that the multivariable prediction model and the hybrid priority scheduling mechanism are introduced, the load stability of the GPU in the large language model reasoning task is improved.