Intelligent driving simulation data storage method and device for hybrid memory and medium

Abstract

The invention relates to an intelligent driving simulation data storage method and device oriented to a hybrid memory and a medium, and belongs to the technical field of intelligent driving. The method comprises the steps that simulation data are graded based on attributes of the simulation data, and grades corresponding to the simulation data are obtained; classifying the simulation data based on the access frequency of the simulation data to obtain types corresponding to the simulation data; and storing the simulation data in the DRAM or the NVM based on the preset weight, the grade and the type. According to the intelligent driving simulation data storage method oriented to the hybrid memory provided by the invention, the limitation on intelligent driving simulation under a single memory can be made up by utilizing a hybrid memory architecture formed by two storage media, namely, the DRAM and the NVM, and utilizing respective advantages of different storage media, so that the intelligent driving simulation can be realized. By optimizing the data storage strategy, the data processing capability of intelligent driving simulation is improved, so that the performance and efficiency of the simulation system are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a hybrid memory-oriented intelligent driving simulation data storage method, device and medium. Background Art

[0002] With the increasing development of intelligent driving, the importance of intelligent driving simulation is increasing, and it has become an indispensable tool in the development of autonomous driving technology. As technology continues to advance, simulation will play a more critical role in ensuring the safety and reliability of autonomous driving and improving development efficiency. With the continuous development of intelligent driving technology, the demand for storage in simulation systems is also growing. The scalability and flexibility of memory storage ensure that the simulation system can adapt to the development and changes of future technologies.

[0003] Intelligent driving simulation requires processing and analyzing a large amount of data, including images, radar signals, and lidar point clouds. High-speed memory storage can ensure fast reading and writing of data, meeting the simulation system's demand for data throughput. At the same time, the performance of memory storage directly affects the speed and efficiency of algorithm iteration. At present, the intelligent driving simulation system mainly adopts a single memory architecture dominated by dynamic random access memory (DRAM) in terms of memory storage, and all simulation data is stored equally in a single memory medium.

[0004] With the rapid development of intelligent driving technology, especially the increasing complexity of autonomous driving functions, the traditional single memory architecture has gradually shown its limitations. Although the single memory architecture has advantages in read and write speed, its volatility, limited storage capacity and high energy consumption refresh operation have made it difficult to meet the needs of large-scale data processing and storage. Intelligent driving simulation needs to process massive amounts of data from various sensors, which have extremely high requirements for storage and processing speed. At the same time, the simulation system also needs to have efficient data persistence, flexibility and scalability to cope with different traffic scenarios and driving conditions. In addition, the performance of the single memory architecture in energy consumption optimization is also difficult to meet the needs of modern simulation systems. With the improvement of environmental awareness and the increase in energy costs, it has become particularly important to develop low-energy simulation systems. Summary of the invention

[0005] In view of this, it is necessary to provide a hybrid memory-oriented intelligent driving simulation data storage method, device and medium to solve the problem of low simulation efficiency caused by limited storage efficiency under a single memory architecture.

[0006] In order to solve the above problems, the present invention provides a hybrid memory-oriented intelligent driving simulation data storage method, comprising: Classifying the simulation data based on the attributes of the simulation data to obtain a grade corresponding to the simulation data; Classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data; The simulation data is stored in DRAM or NVM based on a preset weight, the level, and the type.

[0007] In a possible implementation, the attributes include: Capacity share, real-time and persistence.

[0008] In a possible implementation, the types include: Highly accessed data, medium-accessed data, and cold data.

[0009] In a possible implementation, classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data includes: Determine data with an access frequency greater than or equal to a first threshold as the high-frequency access data; Determine data whose access frequency is less than or equal to a second threshold as the cold data; Determine the data whose access frequency is greater than the second threshold and less than the first threshold as the medium-hot access data; The first threshold is greater than the second threshold.

[0010] In a possible implementation, storing the simulation data in a DRAM or NVM based on a preset weight, the level, and the type includes: Calculating a score corresponding to the simulation data based on a preset weight, the level and the type; Based on the score, the simulation data is stored in DRAM or NVM.

[0011] In a possible implementation, storing the simulation data in a DRAM or NVM based on the score includes: When the score is greater than or equal to a third threshold, storing the simulation data in DRAM; When the score is less than the third threshold, the simulation data is stored in the NVM.

[0012] In a possible implementation, the simulation data includes: Sensor collection data, vehicle status data, environmental model data, simulation configuration information and simulation log information.

[0013] The present invention also provides an intelligent driving simulation data storage device for hybrid memory, comprising: A grading module, used to grade the simulation data based on the attributes of the simulation data to obtain the grade corresponding to the simulation data; A classification module, used for classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data; A storage module is used to store the simulation data in DRAM or NVM based on a preset weight, the level and the type.

[0014] On the other hand, the present invention also provides an electronic device, including a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the intelligent driving simulation data storage method for hybrid memory described in any of the above implementations.

[0015] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent driving simulation data storage method for hybrid memory described in any of the above-mentioned implementation methods.

[0016] The beneficial effects of the present invention are as follows: the intelligent driving simulation data storage method, device and medium for hybrid memory provided by the present invention classify the simulation data according to its attributes, and the attributes may include importance, real-time requirements, data size, etc., so as to obtain the corresponding level of the simulation data. The level can be used to evaluate the importance of the data and classify it according to the access frequency, so that frequently accessed data can be stored in a storage medium with a faster access speed, and the score of the simulation data is calculated by presetting the weight for the level and type of the simulation data, so as to determine whether the simulation data is stored in DRAM or NVM. Data with high importance and frequent access can be stored in DRAM to utilize the high-speed access characteristics of DRAM, and data with low importance or infrequent access can be stored in NVM to save DRAM resources and utilize the non-volatile characteristics of NVM. By utilizing a hybrid memory architecture composed of two storage media, DRAM and NVM, and utilizing the respective advantages of different storage media, the limitations of intelligent driving simulation under a single memory can be compensated, and the data processing capability of the intelligent driving simulation is improved by optimizing the data storage strategy, thereby improving the performance and efficiency of the simulation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A method flow chart of an embodiment of a hybrid memory-oriented intelligent driving simulation data storage method provided by the present invention; Figure 2 A schematic diagram of the structure of an embodiment of a hybrid memory-oriented intelligent driving simulation data storage device provided by the present invention; Figure 3 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] In the description of the embodiments of the present invention, unless otherwise specified, "multiple" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone.

[0021] The descriptions of "first", "second", etc. involved in the embodiments of the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0022] Figure 1 A method flow chart of an embodiment of the hybrid memory-oriented intelligent driving simulation data storage method provided by the present invention, such as Figure 1 As shown, the intelligent driving simulation data storage method for hybrid memory includes: S101, grading the simulation data based on the attributes of the simulation data to obtain the grade corresponding to the simulation data; S102, classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data; S103 , storing the simulation data in DRAM or NVM based on a preset weight, the level and the type.

[0023] It should be noted that: the present invention proposes to adopt a hybrid memory storage architecture composed of dynamic random access memory DRAM and non-volatile memory (NVM) for the intelligent driving simulation system, referred to as DRAM-NVM hybrid memory. Under this architecture, two storage media, DRAM and NVM, coexist in the same parallel address space, and the operating system and application can directly access both through a unified memory address. DRAM has the advantage of faster speed compared to NVM, and NVM has the advantage of higher capacity compared to DRAM. The hybrid memory architecture will form a complementary advantage between these two different storage media.

[0024] Compared with the prior art, the intelligent driving simulation data storage method for hybrid memory provided in an embodiment of the present invention grades the simulation data according to its attributes, and the attributes may include importance, real-time requirements, data size, etc., so as to obtain the level corresponding to the simulation data. The level can be used to evaluate the importance of the data and classify it according to the access frequency, so that frequently accessed data can be stored in a storage medium with a faster access speed, and the score of the simulation data is calculated by presetting the weight for the level and type of the simulation data, so as to determine whether the simulation data is stored in DRAM or NVM. Data with high importance and frequent access can be stored in DRAM to utilize the high-speed access characteristics of DRAM, and data with low importance or infrequent access can be stored in NVM to save DRAM resources and utilize the non-volatile characteristics of NVM. By utilizing a hybrid memory architecture composed of two storage media, DRAM and NVM, and utilizing the respective advantages of different storage media, the limitations of intelligent driving simulation under a single memory can be compensated, and the data processing capability of the intelligent driving simulation is improved by optimizing the data storage strategy, thereby improving the performance and efficiency of the simulation system.

[0025] In some embodiments of the present invention, the simulation data includes: Sensor collection data, vehicle status data, environmental model data, simulation configuration information and simulation log information.

[0026] The relevant simulation data involved in intelligent driving simulation may include sensor acquisition data, vehicle status data, environmental model data, simulation configuration information, and simulation log information.

[0027] Specifically, the sensor collection data may include sensor data including lidar, millimeter-wave radar, forward-looking integrated device, etc.; vehicle status data may include speed, acceleration, direction, etc.; environmental model data may include map information, traffic conditions, etc.; simulation configuration information, data that has been set before the simulation begins; simulation logs and result record information may include system operation status information, abnormal event records and analysis reports, etc.

[0028] In some embodiments of the present invention, the attributes include: Capacity share, real-time and persistence.

[0029] The data classifier classifies the data into levels according to the capacity ratio, real-time performance and persistence of the above simulation data. Data with a larger capacity ratio is more suitable for NVM media with larger storage capacity; data with high real-time requirements are more suitable for storage in DRAM media with faster access efficiency; data with high persistence requirements are more suitable for storage in non-volatile NVM media.

[0030] Therefore, data is divided into Level 1, Level 2, Level 3, Level 4 and Level 5, representing the demand for DRAM memory storage. Level 1 data has the highest demand for DRAM, and Level 5 data has the lowest demand for DRAM.

[0031] The hybrid memory-oriented intelligent driving simulation data storage method provided in an embodiment of the present invention can classify the data according to the attributes of the simulation data (such as capacity ratio, real-time and persistence), and can evaluate the importance and urgency of the simulation data, thereby determining its priority in the storage system.

[0032] In some embodiments of the present invention, the types include: Highly accessed data, medium-accessed data, and cold data.

[0033] In some embodiments of the present invention, classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data includes: Determine data with an access frequency greater than or equal to a first threshold as the high-frequency access data; Determine data whose access frequency is less than or equal to a second threshold as the cold data; Determine the data whose access frequency is greater than the second threshold and less than the first threshold as the medium-hot access data; The first threshold is greater than the second threshold.

[0034] The access pattern analyzer can be used to divide simulation data into three types according to access frequency: high-hot access data HotnessH, medium-hot access data HotnessM, and cold data HotnessC.

[0035] Since some data may have periodic popularity, in order to ensure the real-time and accuracy of data popularity, the access pattern analyzer uses the LFU algorithm to calculate the data access frequency and update the data popularity category in real time, and sets the first threshold ThresholdH and the second threshold ThresholdL.

[0036] Data whose calculated access frequency is greater than or equal to the first threshold ThresholdH is determined as high-heat access data HotnessH, data whose calculated access frequency is less than or equal to the second threshold ThresholdL is determined as cold data HotnessC, and the remaining data is medium-heat access data HotnessM, that is, data whose calculated access frequency is greater than the second threshold ThresholdL and less than the first threshold ThresholdH.

[0037] Hot data is more suitable to be stored in faster DRAM memory.

[0038] The intelligent driving simulation data storage method for hybrid memory provided by an embodiment of the present invention classifies simulation data according to the access frequency so as to store frequently accessed data in a storage medium with a faster access speed, thereby reducing access delay.

[0039] In some embodiments of the present invention, storing the simulation data in DRAM or NVM based on the preset weight, the level and the type includes: Calculating a score corresponding to the simulation data based on a preset weight, the level and the type; Based on the score, the simulation data is stored in DRAM or NVM.

[0040] The storage policy decider can be used to dynamically adjust the distribution of data in two different storage media, DRAM and NVM, according to the data level fed back by the data classifier and the data heat category fed back by the access pattern analyzer.

[0041] Through preset weights, the storage policy decision maker assigns weights to the five data levels: Score_Level1=5, Score_Level2=4, Score_Level3=3, Score_Level4=2, and Score_Level5=1.

[0042] At the same time, the three data heat categories are assigned weights of Score_HotnessH=4, Score_HotnessC=2, and Score_HotnessC=1.

[0043] By periodically calculating the sum of the data weights (Score_LevelN + Score_HotnessN), the higher the score, the higher the priority for being stored in the DRAM medium.

[0044] When the scores are the same, data with a larger heat weight score will have a higher priority. The storage policy will eventually dynamically adjust the storage location of data in DRAM-NVM hybrid memory according to the priority, so as to fully utilize the advantages of hybrid memory and improve storage efficiency.

[0045] The hybrid memory-oriented intelligent driving simulation data storage method provided by the embodiment of the present invention reduces access delay and improves the performance of the simulation system by storing high-frequency access and high-importance data in DRAM. It utilizes the non-volatile characteristics of NVM to protect important data from being lost in unexpected situations such as power failures, and improves the utilization rate of storage resources by reasonably allocating the storage space of DRAM and NVM.

[0046] In some embodiments of the present invention, storing the simulation data in DRAM or NVM based on the score includes: When the score is greater than or equal to a third threshold, storing the simulation data in DRAM; When the score is less than the third threshold, the simulation data is stored in the NVM.

[0047] Optionally, the simulation data score can be calculated periodically, and by pre-setting a third threshold, the simulation data with a score greater than or equal to the third threshold can be stored in DRAM, and the simulation data with a score less than the third threshold can be stored in NVM, thereby dynamically adjusting the storage location of the simulation data in the DRAM-NVM hybrid memory, so as to fully utilize the advantages of the hybrid memory and improve storage efficiency.

[0048] The hybrid memory-oriented intelligent driving simulation data storage method provided by the embodiment of the present invention preferentially stores data with higher scores (i.e., data with high importance and frequent access) in DRAM to utilize the high-speed access characteristics of DRAM, reduce access delays, and improve the performance of the simulation system. Data with lower scores (such as data with lower importance or infrequent access) is stored in NVM to save DRAM resources and utilize the non-volatile characteristics of NVM to ensure the security of data storage.

[0049] The hybrid memory-oriented intelligent driving simulation data storage method provided in the embodiment of the present invention adopts a hybrid memory architecture in order to improve the efficiency and reliability of intelligent driving simulation, while reducing energy consumption and improving data processing capabilities. The hybrid memory architecture combines the advantages of different storage media such as DRAM, solid-state drives, and phase change memories to provide more powerful and flexible data storage for intelligent driving simulation. This architecture can not only improve data access speed, but also optimize data storage and access strategies through intelligent data management algorithms, achieve a balance between performance and cost, and meet the growing technical needs of intelligent driving simulation.

[0050] This paper proposes a hybrid memory-oriented intelligent driving simulation data management solution, aiming to solve the problem of insufficient simulation efficiency caused by limited storage efficiency under a single memory architecture. It aims to fully utilize the advantages of the hybrid memory architecture through innovative data management methods to improve the efficiency and reliability of the intelligent driving simulation system.

[0051] Secondly, in order to fully utilize the advantages of DRAM and NVM, the present invention mainly focuses on the DRAM-NVM hybrid memory architecture and proposes a data management solution for intelligent driving simulation under this architecture. The method includes the following steps: First, the data classifier is used to identify the type and mark the attributes of the simulation data. Second, the access pattern analyzer is used to analyze the data access pattern to predict the data access trend. Then, the storage policy decision maker dynamically adjusts the distribution of data in different storage media based on the data classification and access pattern analysis results.

[0052] The present invention is particularly suitable for intelligent driving simulation systems that need to process a large amount of sensor data and traffic scene information. In the following, in combination with specific scenarios, taking urban traffic flow simulation as an example, the intelligent driving simulation data storage method for hybrid memory provided by the present invention is described in detail.

[0053] First, collect all kinds of data required for urban traffic flow simulation, including sensor data such as traffic cameras and traffic flow detectors; vehicle status data including vehicle speed, acceleration, driving direction, etc.; environmental model data including city maps, traffic signals, road conditions, etc.; simulation configuration information including the number of vehicles, traffic rules, etc.; simulation logs and result records including system status, abnormal events and analysis reports during the simulation process.

[0054] The data classifier classifies data into different levels according to its capacity, real-time and persistence: Data with large capacity and high real-time requirements, such as real-time video streams from traffic cameras, belongs to Level 1. Data with a large volume and medium real-time requirements, such as real-time data from traffic flow detectors, belongs to Level 2. Data with medium capacity and high persistence requirements, such as vehicle status data, belongs to Level 3; Data with a small capacity and medium persistence requirements, such as simulation configuration information, belongs to Level 4; Data with a small capacity and high persistence requirements, such as simulation logs and result recording information, belongs to Level 5.

[0055] Subsequently, the access pattern analyzer monitors the access frequency of various data generated in the urban traffic flow simulation in real time. Using historical access records and real-time monitoring data, the Least Frequently Used (LFU) algorithm is integrated into the simulation system to track and calculate the access frequency of data. Whenever data is accessed, its access frequency is updated.

[0056] The LFU algorithm can help identify the least frequently accessed data items and predict which data sets will face access peaks, such as identifying traffic flow data during peak hours.

[0057] According to the access frequency calculated by the LFU algorithm, the data is classified into one of HotnessH, HotnessM and HotnessC.

[0058] Finally, based on the feedback from the data classifier, the five levels of data generated in the urban traffic flow simulation are assigned different weights (Score_Level1 to Score_Level5) according to their capacity share, real-time and persistence.

[0059] According to the three types of heat data fed back by the access pattern analyzer, weighted scores (Score_HotnessH, Score_HotnessM, Score_HotnessC) are assigned respectively. The storage policy decision maker will periodically calculate the sum of the weighted scores of each type of data, that is, Score_Total = Score_LevelN + Score_HotnessN.

[0060] This total score determines the priority of storing relevant data in DRAM or NVM in the urban traffic flow simulation. Data with higher weight scores will be stored in DRAM media for fast access, while data with lower weight scores are more suitable for storage in NVM media to ensure data persistence.

[0061] The hybrid memory-oriented intelligent driving simulation data storage method provided in an embodiment of the present invention is different from the traditional single memory architecture. The present invention proposes to design a hybrid memory architecture composed of two storage media, DRAM and NVM, for intelligent driving simulation, while utilizing the respective advantages of different storage media to make up for the limitations of intelligent driving simulation under single memory.

[0062] In addition, the data management method proposed by the present invention for hybrid memory for intelligent driving simulation can not only improve the data processing capability of intelligent driving simulation, but also help to effectively reduce the energy consumption of the system through the hybrid memory architecture, and has broad application prospects and practical value.

[0063] In order to better implement the intelligent driving simulation data storage method for hybrid memory in the embodiment of the present invention, based on the intelligent driving simulation data storage method for hybrid memory, the embodiment of the present invention also provides an intelligent driving simulation data storage device for hybrid memory. Figure 2 A structural diagram of an embodiment of a hybrid memory-oriented intelligent driving simulation data storage device provided by the present invention, such as Figure 2 As shown, the hybrid memory-oriented intelligent driving simulation data storage device 200 includes: A grading module 210, configured to grade the simulation data based on the attributes of the simulation data to obtain a grade corresponding to the simulation data; A classification module 220, configured to classify the simulation data based on the access frequency of the simulation data to obtain a type corresponding to the simulation data; The storage module 230 is configured to store the simulation data in a DRAM or NVM based on a preset weight, the level, and the type.

[0064] The intelligent driving simulation data storage device 200 for hybrid memory provided in the above embodiment can implement the technical solution described in the above embodiment of the intelligent driving simulation data storage method for hybrid memory. The specific implementation principles of the above modules or units can refer to the corresponding contents in the embodiment of the intelligent driving simulation data storage method for hybrid memory, which will not be repeated here.

[0065] like Figure 3 As shown, the present invention also provides an electronic device 300. The electronic device 300 includes a processor 301, a memory 302 and a display 303. Figure 3 Only some components of the electronic device 300 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0066] In some embodiments, the processor 301 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 302, such as the intelligent driving simulation data storage method for hybrid memory in the present invention.

[0067] In some embodiments, the processor 301 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 301 may be local or remote. In some embodiments, the processor 301 may be implemented in a cloud platform. In some embodiments, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0068] In some embodiments, the memory 302 may be an internal storage unit of the electronic device 300, such as a hard disk or memory of the electronic device 300. In other embodiments, the memory 302 may also be an external storage device of the electronic device 300, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 300.

[0069] Furthermore, the memory 302 may include both an internal storage unit of the electronic device 300 and an external storage device. The memory 302 is used to store application software installed in the electronic device 300 and various data.

[0070] In some embodiments, the display 303 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an organic light-emitting diode (OLED) touch device, etc. The display 303 is used to display information on the electronic device 300 and to display a visual user interface. The components 301-303 of the electronic device 300 communicate with each other via a system bus.

[0071] In one embodiment, when the processor 301 executes the hybrid memory-oriented intelligent driving simulation data storage program in the memory 302, the following steps may be implemented: Classifying the simulation data based on the attributes of the simulation data to obtain a grade corresponding to the simulation data; Classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data; The simulation data is stored in DRAM or NVM based on a preset weight, the level, and the type.

[0072] It should be understood that: when the processor 301 executes the hybrid memory-oriented intelligent driving simulation data storage program in the memory 302, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the corresponding method embodiment above.

[0073] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 300 mentioned, and the electronic device 300 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic device may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 300 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0074] Correspondingly, an embodiment of the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps or functions of the intelligent driving simulation data storage method for hybrid memory provided in the above-mentioned method embodiments can be implemented.

[0075] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0076] The above describes in detail the intelligent driving simulation data storage method, device and medium for hybrid memory provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A hybrid memory-oriented intelligent driving simulation data storage method, characterized in that: include: Classifying the simulation data based on the attributes of the simulation data to obtain a grade corresponding to the simulation data; Classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data; The simulation data is stored in DRAM or NVM based on a preset weight, the level, and the type.

2. The hybrid memory-oriented intelligent driving simulation data storage method according to claim 1, characterized in that: The properties include: Capacity share, real-time and persistence.

3. The hybrid memory-oriented intelligent driving simulation data storage method according to claim 1, characterized in that: The types include: Highly accessed data, medium-accessed data, and cold data.

4. The hybrid memory-oriented intelligent driving simulation data storage method according to claim 3 is characterized in that: The classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data includes: Determine data with an access frequency greater than or equal to a first threshold as the high-frequency access data; Determine data whose access frequency is less than or equal to a second threshold as the cold data; Determine the data whose access frequency is greater than the second threshold and less than the first threshold as the medium-hot access data; The first threshold is greater than the second threshold.

5. The hybrid memory-oriented intelligent driving simulation data storage method according to claim 1, characterized in that: The storing the simulation data in DRAM or NVM based on the preset weight, the level and the type includes: Calculating a score corresponding to the simulation data based on a preset weight, the level and the type; Based on the score, the simulation data is stored in DRAM or NVM.

6. The hybrid memory-oriented intelligent driving simulation data storage method according to claim 5 is characterized in that: Based on the score, storing the simulation data in DRAM or NVM includes: When the score is greater than or equal to a third threshold, storing the simulation data in DRAM; When the score is less than the third threshold, the simulation data is stored in the NVM.

7. The intelligent driving simulation data storage method for hybrid memory according to any one of claims 1 to 6, characterized in that: The simulation data includes: Sensor collection data, vehicle status data, environmental model data, simulation configuration information and simulation log information.

8. An intelligent driving simulation data storage device for hybrid memory, characterized in that: include: A grading module, used to grade the simulation data based on the attributes of the simulation data to obtain the grade corresponding to the simulation data; A classification module, used for classifying the simulation data based on the access frequency of the simulation data to obtain the type corresponding to the simulation data; A storage module is used to store the simulation data in DRAM or NVM based on a preset weight, the level and the type.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the intelligent driving simulation data storage method for hybrid memory as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for storing intelligent driving simulation data for hybrid memory as described in any one of claims 1 to 7 is implemented.

Images