Intelligent computing center mirror image layered caching method and device oriented to popularity

Abstract

The invention relates to the technical field of intelligent computing centers, intelligent computing centers and computing power infrastructures, and provides an intelligent computing center mirror image layering caching method and device oriented to the Plevatory computing power, and the method comprises the steps: determining a plurality of mirror image layers divided based on a container mirror image, and determining the use frequency of each mirror image layer, the corresponding use frequency is increased by one time every time the mirror image layer is accessed in the preset unit time; the mirror image layering type is any one of the following items: high-frequency mirror image layering, intermediate-frequency mirror image layering and low-frequency mirror image layering; and monitoring the respective environment state of each mirror image layer in real time, and determining whether to execute mirror image layer migration operation or not based on the environment state, the type of the mirror image layer and the use frequency. Therefore, through the mirror image layering dynamic scheduling mechanism provided by the invention, the network bandwidth consumption during mirror image layering loading can be remarkably reduced, the economic cost is greatly reduced, and powerful support is provided for popularization of the popularity.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent computing centers, intelligent computing centers and computing power infrastructure, and particularly relates to an intelligent computing center mirror hierarchical caching method and device for inclusive computing power. Background Art

[0002] With the rapid development of artificial intelligence technology, "intelligent computing centers" and "intelligent computing centers" have emerged as the times require.

[0003] An "intelligent computing center" refers to a facility that provides the required computing power, data and algorithms for artificial intelligence applications (such as artificial intelligence deep learning model development, model training and model inference scenarios) by using large-scale heterogeneous computing power resources, including general computing power and intelligent computing power. The intelligent computing center covers facilities, hardware, software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0004] The "intelligent computing center" includes, but is not limited to, the "intelligent computing center".

[0005] An "intelligent computing center", that is, an artificial intelligence computing center, is a type of computing power infrastructure based on artificial intelligence theory, adopting an artificial intelligence computing architecture, and providing computing power services, data services and algorithm services required for artificial intelligence applications.

[0006] "Computing power" is the core of "intelligent computing centers" and "intelligent computing centers". It is the ability of computer devices or computing / data centers to process information, the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, the computing ability to achieve the output of the target result by processing information data, and a new type of productive force integrating information computing power, network carrying capacity and data storage capacity. It mainly provides services to society through computing power infrastructure.

[0007] In the rapid development of modern cloud computing and containerization technologies, mirror hierarchical files, as the core components of applications and services, play an important role. These mirror hierarchical files usually adopt a hierarchical storage method and are managed in container management platforms such as Harbor (an open-source mirror repository). However, this hierarchical storage design has some limitations. Specifically, each layer of the mirror is fixedly stored in a specific type of file system, resulting in inability to dynamically adjust according to the usage frequency. This static storage method is particularly prominent in the inference and training processes of large models. In the inference and training processes of large models, the basic mirror layers are highly reused, and the occupied space of the mirror hierarchical files is large. Due to their fixed storage, when users load the mirror layers, it not only takes a long time but also consumes a large amount of network bandwidth. This situation not only affects the user experience but also increases the economic cost, restricting the popularization and application of inclusive computing power.

[0008] In summary, since the emergence of the intelligent computing center, the static characteristics of the mirror layer storage have led to the consumption of a large amount of network bandwidth during the inference and training of large models. The loading of mirror layers (especially the highly reused basic mirror layers) causes a large amount of waste of computing power resources, with high economic costs, restricting the popularization and application of inclusive computing power. This problem has become a technical problem that urgently needs to be solved in the intelligent computing center. Summary of the Invention

[0009] The present invention provides an intelligent computing center mirror layer caching method and device for inclusive computing power to solve the technical problem that since the emergence of the intelligent computing center, the static characteristics of the mirror layer storage have led to the consumption of a large amount of network bandwidth during the inference and training of large models, with high economic costs, restricting the popularization and application of inclusive computing power.

[0010] To solve the above technical problems, the present invention is implemented as follows: In a first aspect, the present invention provides an intelligent computing center mirror layer caching method for inclusive computing power, and the method includes: Step S1: Determine a plurality of mirror layers divided based on the container image, and respectively determine the usage frequency of each mirror layer. Wherein, each time a mirror layer is accessed within a preset unit time, the corresponding usage frequency is increased by one; the types of the mirror layers are any one of the following: high-frequency mirror layer, medium-frequency mirror layer, and low-frequency mirror layer; the usage frequency of the high-frequency mirror layer is greater than or equal to a first preset threshold; the usage frequency of the medium-frequency mirror layer is greater than or equal to a second preset threshold but less than the first preset threshold; the usage frequency of the low-frequency mirror layer is less than the second preset threshold; the high-frequency mirror layer is stored on the local high-performance storage device of the intelligent computing center; the medium-frequency mirror layer is stored on the remote high-performance storage device of the intelligent computing center; the low-frequency mirror layer is stored on the remote large-capacity storage device of the intelligent computing center; Step S2: Real-time monitor the environmental status of each of the mirror layers, and determine whether to perform a mirror layer migration operation based on the environmental status, the type of the mirror layer, and the usage frequency, where the mirror layer migration operation includes: a mirror layer upgrade migration operation or a mirror layer downgrade migration operation.

[0011] Optionally, when the type of the mirror layer is the medium-frequency mirror layer, step S2 includes: Step S21: When the environmental status is network idle and the usage frequency of the medium-frequency mirror layer increases to the first preset threshold, perform a mirror layer upgrade migration operation; Among them, the execution of the mirror layer hierarchical upgrade and migration operation includes: migrating the intermediate frequency mirror layer from the remote high-performance storage device of the intelligent computing center to the local high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the intermediate frequency mirror layer in the remote high-performance storage device.

[0012] Optionally, during the execution of the mirror layer migration and upgrade operation, the method further includes: preheating the high-frequency mirror layer, where the preheating is: when the network is idle, loading the high-frequency mirror layer in parallel to the local high-performance storage devices of multiple nodes of the intelligent computing center.

[0013] Optionally, when the type of the mirror layer is the intermediate frequency mirror layer, step S2 further includes: Step S22: When the environmental state is that the storage capacity of the remote high-performance storage device where the intermediate frequency mirror layer is currently located is greater than or equal to a third preset threshold, and the usage frequency of the intermediate frequency mirror layer is less than the second preset threshold, execute the mirror layer downgrade and migration operation; Among them, the execution of the mirror layer downgrade and migration operation includes: updating the index of the intermediate frequency mirror layer to point the storage path of the intermediate frequency mirror layer to the remote large-capacity storage device; asynchronously deleting the original mirror layer file of the intermediate frequency mirror layer in the remote high-performance storage device.

[0014] Optionally, when there are multiple intermediate frequency mirror layers for which the mirror layer downgrade and migration operation is executed, asynchronously deleting the original mirror layer file of the intermediate frequency mirror layer in the remote high-performance storage device includes: asynchronously and parallelly deleting the respective original mirror layer files of the multiple intermediate frequency mirror layers in the remote high-performance storage device.

[0015] Optionally, when the type of the mirror layer is the high-frequency mirror layer, step S2 includes: Step S23: When the environmental state is that the storage capacity of the local high-performance storage device where the high-frequency mirror layer is currently located is greater than or equal to a fourth preset threshold, and the usage frequency of the high-frequency mirror layer is less than the first preset threshold, execute the mirror layer downgrade and migration operation; Among them, the execution of the mirror layer downgrade and migration operation includes: updating the index of the high-frequency mirror layer to point the storage path of the high-frequency mirror layer to the remote high-performance storage device; asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device.

[0016] Optionally, when there are multiple high-frequency image layers for performing the mirror layer downgrade migration operation, asynchronously deleting the original image layer files of the high-frequency image layers in the local high-performance storage device includes: asynchronously and parallelly deleting the original image layer files corresponding to each of the multiple high-frequency image layers in the local high-performance storage device.

[0017] Optionally, when the type of the image layer is the low-frequency image layer, step S2 includes: Step S24: When the environment state is network idle and the usage frequency of the low-frequency image layer increases to the second preset threshold, perform an image layer upgrade migration operation; Wherein, performing the image layer upgrade migration operation includes: migrating the low-frequency image layer from the remote mass storage device of the intelligent computing center to the remote high-performance storage device of the intelligent computing center, and retaining the original image layer file of the low-frequency image layer in the remote mass storage device; updating the index of the low-frequency image layer to point the storage path of the low-frequency image layer to the remote high-performance storage device.

[0018] In a second aspect, the present invention provides an intelligent computing center image layer caching device for inclusive computing power, the device includes: A determination module, configured to execute step S1: determine multiple image layers divided based on a container image, and respectively determine the usage frequency of each image layer, wherein each time an image layer is accessed within a preset unit time, the corresponding usage frequency is increased by one; the type of the image layer is any one of the following: high-frequency image layer, medium-frequency image layer, and low-frequency image layer; the usage frequency of the high-frequency image layer is greater than or equal to a first preset threshold; the usage frequency of the medium-frequency image layer is greater than or equal to a second preset threshold but less than the first preset threshold; the usage frequency of the low-frequency image layer is less than the second preset threshold; the high-frequency image layer is stored on the local high-performance storage device of the intelligent computing center; the medium-frequency image layer is stored on the remote high-performance storage device of the intelligent computing center; the low-frequency image layer is stored on the remote mass storage device of the intelligent computing center; An execution module, configured to execute step S2: real-time monitor the environment state of each image layer, and determine whether to perform an image layer migration operation based on the environment state, the type of the image layer, and the usage frequency, wherein the image layer migration operation includes: an image layer upgrade migration operation or an image layer downgrade migration operation.

[0019] Optionally, when the type of the image layer is the medium-frequency image layer, step S2 includes: Step S21: When the environmental state is network idle and the usage frequency of the intermediate-frequency mirror layer increases to the first preset threshold, perform a mirror layer upgrade and migration operation; Among them, the execution of the mirror layer upgrade and migration operation includes: migrating the intermediate-frequency mirror layer from the remote high-performance storage device of the intelligent computing center to the local high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the intermediate-frequency mirror layer in the remote high-performance storage device.

[0020] Optionally, during the process of performing the mirror layer migration and upgrade operation, the execution module is further configured to warm up the high-frequency mirror layer, where the warm-up is: when the network is idle, parallelly load the high-frequency mirror layer into the local high-performance storage devices of multiple nodes of the intelligent computing center.

[0021] Optionally, when the type of the mirror layer is the intermediate-frequency mirror layer, step S2 further includes: Step S22: When the environmental state is that the storage capacity of the remote high-performance storage device where the intermediate-frequency mirror layer is currently located is greater than or equal to the third preset threshold and the usage frequency of the intermediate-frequency mirror layer is less than the second preset threshold, perform a mirror layer downgrade and migration operation; Among them, the execution of the mirror layer downgrade and migration operation includes: updating the index of the intermediate-frequency mirror layer to point the storage path of the intermediate-frequency mirror layer to the remote large-capacity storage device; asynchronously deleting the original mirror layer file of the intermediate-frequency mirror layer in the remote high-performance storage device.

[0022] Optionally, when there are multiple intermediate-frequency mirror layers for which the mirror layer downgrade and migration operation is performed, asynchronously deleting the original mirror layer files of the intermediate-frequency mirror layers in the remote high-performance storage device includes: asynchronously and parallelly deleting the respective original mirror layer files of the multiple intermediate-frequency mirror layers in the remote high-performance storage device.

[0023] Optionally, when the type of the mirror layer is the high-frequency mirror layer, step S2 includes: Step S23: When the environmental state is that the storage capacity of the local high-performance storage device where the high-frequency mirror layer is currently located is greater than or equal to the fourth preset threshold and the usage frequency of the high-frequency mirror layer is less than the first preset threshold, perform a mirror layer downgrade and migration operation; Among them, the execution of the mirror layer downgrade and migration operation includes: updating the index of the high-frequency mirror layer to point the storage path of the high-frequency mirror layer to the remote high-performance storage device; asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device.

[0024] Optionally, when there are multiple high-frequency image layers for performing the mirror layer downgrade migration operation, asynchronously deleting the original image layer files of the high-frequency image layers in the local high-performance storage device includes: asynchronously and parallelly deleting the original image layer files corresponding to each of the multiple high-frequency image layers in the local high-performance storage device.

[0025] Optionally, when the type of the image layer is the low-frequency image layer, step S2 includes: Step S24: When the environment state is network idle and the usage frequency of the low-frequency image layer increases to the second preset threshold, perform an image layer upgrade migration operation; Wherein, performing the image layer upgrade migration operation includes: migrating the low-frequency image layer from the remote mass storage device of the intelligent computing center to the remote high-performance storage device of the intelligent computing center, and retaining the original image layer file of the low-frequency image layer in the remote mass storage device; updating the index of the low-frequency image layer to point the storage path of the low-frequency image layer to the remote high-performance storage device.

[0026] In a third aspect, the present invention provides a server, including: a processor, a memory, and a program stored on the memory and executable on the processor, where when the program is executed by the processor, it implements the steps of an intelligent computing center image layer caching method for inclusive computing power as described in the first aspect above.

[0027] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, where when the computer program is executed by a processor, it implements the steps of an intelligent computing center image layer caching method for inclusive computing power as described in the first aspect above.

[0028] In a fifth aspect, the present invention provides a computer program product, including computer instructions, where when the computer instructions are executed by a processor, they implement the steps of an intelligent computing center image layer caching method for inclusive computing power as described in the first aspect above.

[0029] In the present invention, a mirror hierarchical classification and migration mechanism based on usage frequency is first established. The mirror layers are divided into three categories: high-frequency mirror layers, medium-frequency mirror layers, and low-frequency mirror layers. By monitoring the environmental status and usage frequency of the mirror layers, the upgrade or downgrade migration of the mirror layers between different performance storage devices is automatically triggered. This dynamic scheduling mechanism enables the high-frequency accessed mirror layers to be stored on the local high-performance storage devices of the intelligent computing center, while the low-frequency accessed mirror layers can be automatically transferred to remote large-capacity storage devices. While ensuring the mirror layer loading efficiency, it can significantly reduce the consumption of network bandwidth and save computing power resources.

[0030] In summary, through the mirror layer dynamic scheduling mechanism provided by the present invention, the network bandwidth consumption during mirror layer loading can be significantly reduced, the waste of computing power resources and economic costs are greatly reduced, providing strong support for the popularization of inclusive computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 is a flowchart of an intelligent computing center mirror layer caching method for inclusive computing power provided by the present invention; Figure 2 is a structural block diagram of an intelligent computing center mirror layer caching device for inclusive computing power provided by the present invention; Figure 3 is a schematic structural diagram of an electronic device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] First, the technical terms related to the present invention will be briefly described below.

[0034] The "computing power" described in the present invention refers to: the ability of a computer device or a computing / data center to process information, the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, the computing ability to achieve the output of a target result by processing information data, and a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, which mainly provides services to society through computing power infrastructure.

[0035] The "computational power" (Computational Power, CP) described in the present invention refers to: the ability of a data center server to process data and achieve result output, which is a comprehensive indicator for measuring the computing ability of a data center and includes general computing ability, supercomputing ability, and intelligent computing ability. The commonly used measurement unit is the number of floating-point operations per second (FLOPS, 1 EFLOPS = 10^18 FLOPS), and the larger the value, the stronger the comprehensive computing ability. It is estimated that 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream laptops. The calculation formula is: CP = CP_general + CP_intelligent + CP_super.

[0036] The "carrying capacity" (Network Power, NP) described in the present invention refers to: the performance of the data transmission ability of computing power facilities, which is a comprehensive ability including network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc., and involves network transmission inside and between data centers, and is a comprehensive indicator for measuring network transmission scheduling ability.

[0037] The "storage power" (Storage Power, SP) described in the present invention refers to: the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon, which is a comprehensive indicator for measuring the data storage ability of a data center and includes external storage devices such as storage arrays and server internal storage devices. The commonly used measurement unit for storage capacity is exabyte (EB, 1 EB = 2^60 bytes), the commonly used measurement unit for performance is the number of read and write operations per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB), and the disaster recovery ratio is an important manifestation of security and reliability.

[0038] The "computing power infrastructure" described in the present invention refers to: a new type of information infrastructure integrating information computing power, network carrying capacity, and data storage capacity, which can realize the centralized computing, storage, transmission, and application of information.

[0039] The "new information infrastructure" described in the present invention refers to: mainly including network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, satellite Internet, etc., computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, supercomputing centers, etc., and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0040] The "computing power" described in the present invention includes: general computing power, intelligent computing power, and super computing power.

[0041] The "general computing power" described in the present invention refers to: the computing power provided by servers based on CPU (Central Processing Unit) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0042] The "intelligent computing power" described in the present invention refers to: for various artificial intelligence innovation applications, a computing platform deployed on a large scale based on dedicated chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit), such as natural language processing, machine vision, etc.

[0043] The "super computing power" described in the present invention refers to: mainly the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and processes extremely complex or data-intensive problems through a dedicated operating system. It is mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, gene analysis, etc.

[0044] The "intelligent computing center" described in the present invention refers to: a facility that mainly provides the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios of artificial intelligence deep learning model development, model training, and model inference) by using large-scale heterogeneous computing power resources, including general computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0045] The "intelligent computing center" described in the present invention includes but is not limited to the "intelligent computing center".

[0046] The "intelligent computing center" described in the present invention, namely the artificial intelligence computing center, is a type of computing power infrastructure that provides computing power services, data services, and algorithm services required for artificial intelligence applications based on artificial intelligence theory and using an artificial intelligence computing architecture.

[0047] The "computing power center" described in the present invention refers to a facility mainly composed of infrastructure such as wind, fire, water, and electricity and IT software and hardware devices, with computing power, carrying capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0048] The "supercomputing center" described in the present invention refers to a supercomputing data center, which is a data center based on supercomputers or large-scale computing clusters and can provide functions such as large-scale computing, storage, and network services, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling, and genome sequencing.

[0049] The "computing power resources" described in the present invention refer to the technologies and facilities required for the development of the digital society with information computing, transmission, storage, and application capabilities, including but not limited to computing resources such as CPUs and GPUs, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and support and guarantee resources such as wind, fire, water, and electricity.

[0050] The "inclusive computing power" described in the present invention refers to providing appropriate and effective computing power services for all social strata and groups with computing power service needs at an affordable cost based on the requirements of equal opportunity and the principle of commercial sustainability.

[0051] The "local high-performance storage device" described in the present invention refers to a low-latency, high-throughput storage device deployed locally at the nodes of an intelligent computing center, usually using NVMe SSD (Non-Volatile Memory Express Solid State Drive) or memory storage technology, providing microsecond-level access latency and million-level IOPS (Input / Output Operations Per Second) performance, and is specifically used to carry high-frequency mirror layering, such as the core components of the AI (Artificial Intelligence) framework.

[0052] The "remote high-performance storage device" described in the present invention refers to a distributed high-performance storage cluster built with all-flash arrays, with the ability to share data across nodes, and the latency is controlled within milliseconds, usually used to store medium-frequency mirror layering.

[0053] The "remote large-capacity storage device" described in the present invention refers to a high-density storage system built based on mechanical hard disks, providing PB-level capacity through an object storage interface, and is usually used to store low-frequency mirror layering (such as historical training model backup data).

[0054] It should be noted that the local high-performance storage device has a faster read / write speed and usually has a capacity of 1T; the remote high-performance storage device has a medium read / write speed and usually has a capacity of 500T; while the remote large-capacity storage device has a slower read / write speed and usually has a capacity of 100P.

[0055] Figure 1 Disclosed is an intelligent computing center mirror hierarchical caching method for inclusive computing power, as Figure 1 shown, the method includes: Step S1: Determine multiple mirror hierarchies divided based on container images, and respectively determine the usage frequency of each mirror hierarchy; Wherein, each time a mirror hierarchy is accessed within a preset unit time, the corresponding usage frequency is increased by one; the type of the mirror hierarchy is any one of the following: high-frequency mirror hierarchy, medium-frequency mirror hierarchy, and low-frequency mirror hierarchy; the usage frequency of the high-frequency mirror hierarchy is greater than or equal to a first preset threshold; the usage frequency of the medium-frequency mirror hierarchy is greater than or equal to a second preset threshold but less than the first preset threshold; the usage frequency of the low-frequency mirror hierarchy is less than the second preset threshold; the high-frequency mirror hierarchy is stored on the local high-performance storage device of the intelligent computing center; the medium-frequency mirror hierarchy is stored on the remote high-performance storage device of the intelligent computing center; the low-frequency mirror hierarchy is stored on the remote large-capacity storage device of the intelligent computing center; Step S2: Real-time monitor the environmental status of each mirror hierarchy, and determine whether to perform a mirror hierarchy migration operation based on the environmental status, the type of the mirror hierarchy, and the usage frequency; Wherein, the mirror hierarchy migration operation includes: a mirror hierarchy upgrade migration operation or a mirror hierarchy downgrade migration operation.

[0056] It should be noted that by constructing an intelligent mirror hierarchy management method, the dynamic matching of storage resources and access requirements can be realized within the intelligent computing center: First, according to the access count statistics within a preset time window, the mirror hierarchies are accurately divided into three categories: high-frequency (usage frequency ≥ first preset threshold), medium-frequency (second preset threshold ≤ usage frequency < first preset threshold), and low-frequency (usage frequency < second preset threshold), and are respectively deployed in the local high-performance storage, remote high-performance storage, and remote large-capacity storage devices; by continuously monitoring key parameters such as the real-time access frequency, storage device load status, and network transmission delay of each mirror hierarchy, the hierarchical migration mechanism can be automatically triggered. Thus, not only is it ensured that the high-frequency mirror hierarchy always obtains the optimal storage location, but also the amount of data transmitted across the network is significantly reduced by dynamically adjusting the storage level, which can significantly reduce the network bandwidth consumption during mirror hierarchy loading, greatly reduce the waste of computing power resources and economic costs, and provide strong support for the popularization of inclusive computing power.

[0057] In a possible implementation, when the type of the mirror layer is the intermediate-frequency mirror layer, step S2 includes: Step S21: When the environmental state is network idle and the usage frequency of the intermediate-frequency mirror layer increases to a first preset threshold, perform a mirror layer upgrade and migration operation; Among them, performing the mirror layer upgrade and migration operation includes: migrating the intermediate-frequency mirror layer from the remote high-performance storage device of the intelligent computing center to the local high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the intermediate-frequency mirror layer in the remote high-performance storage device.

[0058] It should be noted that when the type of the intermediate-frequency mirror layer is identified, the mirror layer upgrade and migration operation is performed according to the following conditions. When it is monitored that the environmental state is network idle and the cumulative usage frequency of the intermediate-frequency mirror layer reaches the first preset threshold (i.e., the standard of the high-frequency mirror layer, for example: the usage frequency reaches 100 times / unit time), the mirror layer upgrade and migration operation is immediately started. The core process of this operation is to completely copy the intermediate-frequency mirror layer from the remote high-performance storage device of the intelligent computing center to the local high-performance storage device, while ensuring that the original mirror layer file in the remote high-performance storage device continues to be retained without deletion. This design completes the migration of the mirror layer during the network idle period, avoiding the contention of bandwidth resources, enabling the subsequent access to the high-frequency mirror layer to be quickly responded to through local storage, and effectively preventing the waste of resources caused by repeated transmission by retaining the remote original mirror layer file.

[0059] In a possible implementation, during the process of performing the mirror layer migration and upgrade operation, the method further includes: preheating the high-frequency mirror layer, where preheating is: when the network is idle, loading the high-frequency mirror layer in parallel to the local high-performance storage devices of multiple nodes of the intelligent computing center.

[0060] It should be noted that when the mirror layer upgrade and migration operation is performed, a preheating mechanism for the high-frequency mirror layer is synchronously implemented. This mechanism transfers the high-frequency mirror layer to be upgraded to the local high-performance storage devices of multiple computing nodes in the intelligent computing center simultaneously during the network idle time period. Through this parallel loading method, the distribution and deployment of hot data can be completed in advance during the storage layer adjustment, so that after the high-frequency mirror layer completes the migration and upgrade, distributed copies can be formed in the local storage devices of multiple computing nodes. This preheating operation makes full use of the idle bandwidth resources during the network idle period, avoiding transmission congestion caused by concentrated access during peak hours, and realizing data reading nearby through multi-node local storage, which can achieve the technical effect of significantly reducing the access latency of the high-frequency mirror layer.

[0061] In a possible implementation, when the type of the mirror layer is the intermediate-frequency mirror layer, step S2 further includes: Step S22: When the storage capacity of the remote high-performance storage device where the intermediate-frequency mirror tier is currently located in the environmental state is greater than or equal to the third preset threshold, and the usage frequency of the intermediate-frequency mirror tier is less than the second preset threshold, perform the mirror tier downgrade migration operation; Among them, performing the mirror tier downgrade migration operation includes: updating the index of the intermediate-frequency mirror tier to point the storage path of the intermediate-frequency mirror tier to the remote large-capacity storage device; asynchronously deleting the original mirror tier file of the intermediate-frequency mirror tier in the remote high-performance storage device.

[0062] It should be noted that for the storage resource optimization scenario of the intermediate-frequency mirror tier, the downgrade migration mechanism can be triggered by dual conditions: when it is monitored that the storage capacity of the remote high-performance storage device where the intermediate-frequency mirror tier is currently located reaches or exceeds the third preset threshold (i.e., the storage space is in a tense state, for example: reaching 70%), and the actual usage frequency of this intermediate-frequency mirror tier has dropped to the low-frequency mirror tier standard (i.e., lower than the second preset threshold), the mirror tier downgrade migration operation is automatically performed. That is: first, update the index of the intermediate-frequency mirror tier to seamlessly switch the access path of the intermediate-frequency mirror tier to the remote large-capacity storage device to ensure data access continuity; then, gradually clean up the original mirror tier files in the remote high-performance storage device in an asynchronous deletion manner to avoid an instantaneous impact on the storage system performance caused by the synchronous deletion operation.

[0063] In a possible implementation manner, when there are multiple intermediate-frequency mirror tiers for which the mirror tier downgrade migration operation is performed, asynchronously deleting the original mirror tier files of the intermediate-frequency mirror tiers in the remote high-performance storage device includes: asynchronously and parallelly deleting the respective original mirror tier files of the multiple intermediate-frequency mirror tiers in the remote high-performance storage device.

[0064] It should be noted that when it is necessary to perform the downgrade migration operation on multiple intermediate-frequency mirror tiers simultaneously, a parallel resource release strategy can be adopted, that is, for all intermediate-frequency mirror tiers that meet the downgrade conditions, after completing the batch update of the indexes of these mirror tiers to point to the remote large-capacity storage device, immediately through the asynchronous parallel deletion mechanism, perform the deletion operation on the original mirror tier files of the multiple intermediate-frequency mirror tiers in the remote high-performance storage device at the same time. Thus, it not only ensures the immediate availability of data access after the storage path is switched, but also can improve the speed of reclaiming the storage space through parallel operations.

[0065] In a possible implementation manner, when the type of the mirror tier is the high-frequency mirror tier, step S2 includes: Step S23: When the storage capacity of the local high-performance storage device where the high-frequency mirror tier is currently located in the environmental state is greater than or equal to the fourth preset threshold, and the usage frequency of the high-frequency mirror tier is less than the first preset threshold, perform the mirror tier downgrade migration operation; Among them, performing the mirror layer downgrade migration operation includes: updating the index of the high-frequency mirror layer to point the storage path of the high-frequency mirror layer to a remote high-performance storage device; asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device.

[0066] In a possible implementation manner, when there are multiple high-frequency mirror layers for which the mirror layer downgrade migration operation is performed, asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device includes: asynchronously and parallelly deleting the respective original mirror layer files of the multiple high-frequency mirror layers in the local high-performance storage device.

[0067] It should be noted that for the dynamic management scenario of the storage resources of the high-frequency mirror layer, the downgrade migration mechanism is triggered by the dual indicators of capacity and frequency, that is, when the current storage capacity of the local high-performance storage device where the high-frequency mirror layer is located reaches or exceeds the fourth preset threshold (indicating that the local storage resources are tight, for example: reaching 70%), and the actual usage frequency of this high-frequency mirror layer has been lower than its classification standard (that is, less than the first preset threshold), the mirror layer downgrade migration operation is automatically performed. The key steps of this operation include: first, accurately update the index information of this high-frequency mirror layer, seamlessly switch its storage path to the remote high-performance storage device, and ensure that subsequent access requests are redirected without perception; subsequently, gradually clear the original mirror layer file in the local high-performance storage device by means of asynchronous deletion, avoiding the instantaneous pressure on the local storage performance caused by synchronous deletion. When there are multiple high-frequency mirror layers that meet the downgrade conditions, an asynchronous parallel deletion mechanism can be adopted to simultaneously perform the deletion operation on the original mirror layer files of the multiple high-frequency mirror layers in the local high-performance storage device. Thus, on the premise of ensuring business continuity, the space recovery efficiency of the local high-performance storage device can be improved.

[0068] In a possible implementation manner, when the type of the mirror layer is a low-frequency mirror layer, step S2 includes: Step S24: When the environmental state is network idle and the usage frequency of the low-frequency mirror layer increases to the second preset threshold, perform the mirror layer upgrade migration operation; Among them, performing the mirror layer upgrade migration operation includes: migrating the low-frequency mirror layer from the remote large-capacity storage device of the intelligent computing center to the remote high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the low-frequency mirror layer in the remote large-capacity storage device; updating the index of the low-frequency mirror layer to point the storage path of the low-frequency mirror layer to the remote high-performance storage device.

[0069] It should be noted that for the scenario of improving the access popularity of the low-frequency mirror layer, the upgrade operation can be triggered through the linkage mechanism of network status and usage frequency: when it is detected that the current network is in an idle state and the usage frequency of the low-frequency mirror layer continuously increases to the medium-frequency mirror layer standard (i.e., reaches the second preset threshold, for example: reaches 10 times / unit time), the mirror layer upgrade and migration operation is immediately executed. The core process of this operation includes two key actions: first, the low-frequency mirror layer is completely copied from the remote large-capacity storage device in the intelligent computing center to the remote high-performance storage device, while keeping the original mirror layer file in the remote large-capacity storage device without deletion to ensure data integrity; subsequently, the index information of this mirror layer is accurately updated, and its storage path is switched to the new copy location of the remote high-performance storage device. By using the idle network time to complete data transmission, it not only avoids the migration operation from occupying the normal service bandwidth, but also ensures data backup and reliability through the design of retaining the original mirror layer file.

[0070] Another brief description of the use of the mirror layer is as follows. When a mirror layer access request is initiated, first, it is detected whether the target mirror layer exists in the local high-performance storage device. If it exists, the local copy is directly called, and the access latency is very low at this time; if the target mirror layer is not stored locally, an access request is automatically sent to the remote high-performance storage device. If the medium-frequency mirror layer copy does not exist either, finally, the low-frequency mirror layer is loaded from the remote large-capacity storage device. This stepped access strategy can not only ensure the optimal access performance of high-frequency hot data, but also maximize the utilization rate of storage resources through the step-by-step fallback mechanism.

[0071] In the present invention, first, a mirror layer classification and migration mechanism based on usage frequency is established. The mirror layer is divided into three categories: high-frequency mirror layer, medium-frequency mirror layer, and low-frequency mirror layer. By monitoring the environmental status and usage frequency of the mirror layer, the upgrade or downgrade migration of the mirror layer between different performance storage devices is automatically triggered. This dynamic scheduling mechanism enables the mirror layer with high-frequency access to be stored on the local high-performance storage device in the intelligent computing center, while the mirror layer with low-frequency access can be automatically transferred to the remote large-capacity storage device. While ensuring the mirror layer loading efficiency, it can significantly reduce the consumption of network bandwidth and save computing power resources.

[0072] In summary, through the mirror layer dynamic scheduling mechanism provided by the present invention, the network bandwidth consumption during mirror layer loading can be significantly reduced, the waste of computing power resources and economic costs are greatly reduced, providing strong support for the popularization of inclusive computing power.

[0073] Figure 2 Shows an intelligent computing center mirror layer caching device for inclusive computing power provided according to the present invention, as Figure 2As shown, device 20 includes: A determination module 201, configured to execute step S1: determine multiple mirror layers divided based on a container image, and respectively determine the usage frequency of each mirror layer; Wherein, each time a mirror layer is accessed within a preset unit time, the corresponding usage frequency is incremented by one; the type of the mirror layer is any one of the following: high-frequency mirror layer, medium-frequency mirror layer, and low-frequency mirror layer; the usage frequency of the high-frequency mirror layer is greater than or equal to a first preset threshold; the usage frequency of the medium-frequency mirror layer is greater than or equal to a second preset threshold but less than the first preset threshold; the usage frequency of the low-frequency mirror layer is less than the second preset threshold; the high-frequency mirror layer is stored on a local high-performance storage device of the intelligent computing center; the medium-frequency mirror layer is stored on a remote high-performance storage device of the intelligent computing center; the low-frequency mirror layer is stored on a remote large-capacity storage device of the intelligent computing center; An execution module 202, configured to execute step S2: real-time monitor the environmental status of each mirror layer, and determine whether to perform a mirror layer migration operation based on the environmental status, the type of the mirror layer, and the usage frequency, wherein the mirror layer migration operation includes: a mirror layer upgrade migration operation or a mirror layer downgrade migration operation.

[0074] In a possible implementation manner, when the type of the mirror layer is a medium-frequency mirror layer, step S2 includes: Step S21: When the environmental status is network idle and the usage frequency of the medium-frequency mirror layer increases to the first preset threshold, perform a mirror layer upgrade migration operation; Wherein, performing the mirror layer upgrade migration operation includes: migrating the medium-frequency mirror layer from a remote high-performance storage device of the intelligent computing center to a local high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the medium-frequency mirror layer in the remote high-performance storage device.

[0075] In a possible implementation manner, during the process of performing the mirror layer upgrade migration operation, the execution module is further configured to preheat the high-frequency mirror layer, wherein preheating is: when the network is idle, parallelly load the high-frequency mirror layer into the local high-performance storage devices of multiple nodes of the intelligent computing center.

[0076] In a possible implementation manner, when the type of the mirror layer is a medium-frequency mirror layer, step S2 further includes: Step S22: When the environmental status is that the storage capacity of the remote high-performance storage device where the medium-frequency mirror layer is currently located is greater than or equal to a third preset threshold, and the usage frequency of the medium-frequency mirror layer is less than the second preset threshold, perform a mirror layer downgrade migration operation; Among them, performing the mirror layer downgrade migration operation includes: updating the index of the intermediate-frequency mirror layer to point the storage path of the intermediate-frequency mirror layer to a remote large-capacity storage device; asynchronously deleting the original mirror layer file of the intermediate-frequency mirror layer in the remote high-performance storage device.

[0077] In a possible implementation, when there are multiple intermediate-frequency mirror layers for which the mirror layer downgrade migration operation is performed, asynchronously deleting the original mirror layer file of the intermediate-frequency mirror layer in the remote high-performance storage device includes: asynchronously and parallelly deleting the respective original mirror layer files of the multiple intermediate-frequency mirror layers in the remote high-performance storage device.

[0078] In a possible implementation, when the type of the mirror layer is a high-frequency mirror layer, step S2 includes: Step S23: When the environmental state is that the storage capacity of the local high-performance storage device where the high-frequency mirror layer is currently located is greater than or equal to a fourth preset threshold and the usage frequency of the high-frequency mirror layer is less than a first preset threshold, perform the mirror layer downgrade migration operation; Among them, performing the mirror layer downgrade migration operation includes: updating the index of the high-frequency mirror layer to point the storage path of the high-frequency mirror layer to a remote high-performance storage device; asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device.

[0079] In a possible implementation, when there are multiple high-frequency mirror layers for which the mirror layer downgrade migration operation is performed, asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device includes: asynchronously and parallelly deleting the respective original mirror layer files of the multiple high-frequency mirror layers in the local high-performance storage device.

[0080] In a possible implementation, when the type of the mirror layer is a low-frequency mirror layer, step S2 includes: Step S24: When the environmental state is that the network is idle and the usage frequency of the low-frequency mirror layer increases to a second preset threshold, perform the mirror layer upgrade migration operation; Among them, performing the mirror layer upgrade migration operation includes: migrating the low-frequency mirror layer from the remote large-capacity storage device of the intelligent computing center to the remote high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the low-frequency mirror layer in the remote large-capacity storage device; updating the index of the low-frequency mirror layer to point the storage path of the low-frequency mirror layer to the remote high-performance storage device.

[0081] In the present invention, a mirror hierarchical classification and migration mechanism based on usage frequency is first established. The mirror layers are divided into three categories: high-frequency mirror layers, medium-frequency mirror layers, and low-frequency mirror layers. By monitoring the environmental status and usage frequency of the mirror layers, the upgrade or downgrade migration of the mirror layers between different performance storage devices is automatically triggered. This dynamic scheduling mechanism enables the high-frequency accessed mirror layers to be stored on the local high-performance storage devices of the intelligent computing center, while the low-frequency accessed mirror layers can be automatically transferred to the remote large-capacity storage devices. While ensuring the mirror layer loading efficiency, it can significantly reduce the consumption of network bandwidth and save computing power resources.

[0082] In summary, through the mirror layer dynamic scheduling mechanism provided by the present invention, the network bandwidth consumption during mirror layer loading can be significantly reduced, and the waste of computing power resources and economic costs are greatly reduced, providing strong support for the popularization of inclusive computing power.

[0083] Please refer to Figure 3 , the present invention also provides an electronic device 30, including a processor 301, a memory 302, and a computer program stored on the memory 302 and executable on the processor 301. When the computer program is executed by the processor 301, it implements the steps of the above-mentioned intelligent computing center mirror layer caching method for inclusive computing power and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0084] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-mentioned intelligent computing center mirror layer caching method for inclusive computing power and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. Among them, the computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0085] The present invention also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the steps of the above-mentioned intelligent computing center mirror layer caching method for inclusive computing power and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0086] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising such element.

[0087] Through the description of the above embodiments, those skilled in the art can clearly understand that the above method can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the method described in the present invention.

[0088] The present invention has been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims, and all of them fall within the protection scope of the present invention.

Claims

1. An intelligent computing center mirror layer caching method for inclusive computing power, characterized in that, The method includes: Step S1: Determine multiple mirror layers divided based on the container image, and respectively determine the usage frequency of each mirror layer; Wherein, each time a mirror layer is accessed within a preset unit time, the corresponding usage frequency is increased by one; the types of the mirror layers are any one of the following: high-frequency mirror layer, medium-frequency mirror layer, and low-frequency mirror layer; the usage frequency of the high-frequency mirror layer is greater than or equal to a first preset threshold; the usage frequency of the medium-frequency mirror layer is greater than or equal to a second preset threshold but less than the first preset threshold; the usage frequency of the low-frequency mirror layer is less than the second preset threshold; the high-frequency mirror layer is stored on the local high-performance storage device of the intelligent computing center; the medium-frequency mirror layer is stored on the remote high-performance storage device of the intelligent computing center; the low-frequency mirror layer is stored on the remote large-capacity storage device of the intelligent computing center; Step S2: Real-time monitor the environmental state of each of the mirror layers, and determine whether to perform a mirror layer migration operation based on the environmental state, the type of the mirror layer, and the usage frequency, wherein the mirror layer migration operation includes: a mirror layer upgrade migration operation or a mirror layer downgrade migration operation.

2. The method according to claim 1, wherein When the type of the mirror layer is the medium-frequency mirror layer, step S2 includes: Step S21: When the environmental state is network idle and the usage frequency of the medium-frequency mirror layer increases to the first preset threshold, perform a mirror layer upgrade migration operation; Wherein, performing the mirror layer upgrade migration operation includes: migrating the medium-frequency mirror layer from the remote high-performance storage device of the intelligent computing center to the local high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the medium-frequency mirror layer in the remote high-performance storage device.

3. The method according to claim 2, wherein During the process of performing the mirror layer upgrade migration operation, the method further includes: preheating the high-frequency mirror layer, wherein the preheating is: when the network is idle, parallelly loading the high-frequency mirror layer into the local high-performance storage devices of multiple nodes of the intelligent computing center.

4. The method according to claim 1, wherein When the type of the mirror layer is the medium-frequency mirror layer, step S2 further includes: Step S22: When the environmental state is that the storage capacity of the remote high-performance storage device where the medium-frequency mirror layer is currently located is greater than or equal to a third preset threshold and the usage frequency of the medium-frequency mirror layer is less than the second preset threshold, perform a mirror layer downgrade migration operation; Wherein, performing the mirror layer downgrade migration operation includes: updating the index of the medium-frequency mirror layer to point the storage path of the medium-frequency mirror layer to the remote large-capacity storage device; asynchronously deleting the original mirror layer file of the medium-frequency mirror layer in the remote high-performance storage device.

5. The method according to claim 4, wherein When there are multiple intermediate-frequency mirror layers during the execution of the mirror layer demotion and migration operation, asynchronously deleting the original mirror layer files of the intermediate-frequency mirror layers in the remote high-performance storage device includes: asynchronously and parallelly deleting the original mirror layer files corresponding to each of the multiple intermediate-frequency mirror layers in the remote high-performance storage device.

6. The method according to claim 1, characterized in that, When the type of the mirror layer is the high-frequency mirror layer, step S2 includes: Step S23: When the environmental state is that the storage capacity of the local high-performance storage device where the high-frequency mirror layer is currently located is greater than or equal to the fourth preset threshold, and the usage frequency of the high-frequency mirror layer is less than the first preset threshold, perform the mirror layer demotion and migration operation; Among them, performing the mirror layer demotion and migration operation includes: updating the index of the high-frequency mirror layer to point the storage path of the high-frequency mirror layer to the remote high-performance storage device; asynchronously deleting the original mirror layer file of the high-frequency mirror layer in the local high-performance storage device.

7. The method according to claim 6, wherein When there are multiple high-frequency mirror layers during the execution of the mirror layer demotion and migration operation, asynchronously deleting the original mirror layer files of the high-frequency mirror layers in the local high-performance storage device includes: asynchronously and parallelly deleting the original mirror layer files corresponding to each of the multiple high-frequency mirror layers in the local high-performance storage device.

8. The method according to claim 1, characterized in that, When the type of the mirror layer is the low-frequency mirror layer, step S2 includes: Step S24: When the environmental state is that the network is idle and the usage frequency of the low-frequency mirror layer increases to the second preset threshold, perform the mirror layer upgrade and migration operation; Among them, performing the mirror layer upgrade and migration operation includes: migrating the low-frequency mirror layer from the remote large-capacity storage device of the intelligent computing center to the remote high-performance storage device of the intelligent computing center, and retaining the original mirror layer file of the low-frequency mirror layer in the remote large-capacity storage device; updating the index of the low-frequency mirror layer to point the storage path of the low-frequency mirror layer to the remote high-performance storage device.

9. An intelligent computing center mirror hierarchical caching device for inclusive computing power, characterized in that, The device includes: A determination module, configured to execute step S1: determine multiple mirror layers divided based on the container image, and respectively determine the usage frequency of each mirror layer; Among them, each time a mirror layer is accessed within a preset unit time, the corresponding usage frequency is increased by one; the type of the mirror layer is any one of the following: high-frequency mirror layer, intermediate-frequency mirror layer, and low-frequency mirror layer; the usage frequency of the high-frequency mirror layer is greater than or equal to the first preset threshold; the usage frequency of the intermediate-frequency mirror layer is greater than or equal to the second preset threshold but less than the first preset threshold; the usage frequency of the low-frequency mirror layer is less than the second preset threshold; the high-frequency mirror layer is stored on the local high-performance storage device of the intelligent computing center; the intermediate-frequency mirror layer is stored on the remote high-performance storage device of the intelligent computing center; the low-frequency mirror layer is stored on the remote large-capacity storage device of the intelligent computing center; An execution module, configured to execute step S2: monitor in real time the environmental status of each of the mirror layers, and determine whether to perform a mirror layer migration operation based on the environmental status, the type of the mirror layer, and the usage frequency, where the mirror layer migration operation includes: a mirror layer upgrade migration operation or a mirror layer downgrade migration operation.

10. A server, characterized in that, Comprising: A processor, a memory, and a program stored on the memory and executable on the processor, where when the program is executed by the processor, the steps of an intelligent computing center mirror layer caching method for inclusive computing power as described in any one of claims 1-8 are implemented.

11. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of an intelligent computing center mirror layer caching method for inclusive computing power as described in any one of claims 1-8 are implemented.

12. A computer program product, characterized in that, Including computer instructions, where when the computer instructions are executed by a processor, the steps of an intelligent computing center mirror layer caching method for inclusive computing power as described in any one of claims 1-8 are implemented.

Images