Edge device disk management method and related device

Abstract

The invention discloses an edge device disk management method and related equipment, and the method comprises the steps: distributing a first disk resource needed by a current service application from a first edge device in a bare disk mapping mode when the current service application is determined to be a bandwidth intensive application based on the network bandwidth demand of at least one dimension of the current service application, a bare disk mapping mode can enable the bandwidth-intensive application to directly read and write a host file system, so that the overhead of a middle level is reduced, and the optimal performance of the bandwidth-intensive application is ensured; and when the current service application is determined to be the non-bandwidth-intensive application, pooling the disk of the second edge device to obtain the storage pool, and allocating the second disk resource required by the current service application, so that the system performance and the resource utilization rate can be improved by pooling the disk of the second edge device. According to the method, the disk resources are finely classified and managed, so that the dynamic and efficient distribution of the disk resources is realized, and the requirements of different business applications on the performance and the disk resources are met.

Description

Technical Field

[0001] The present invention relates to the technical field of disk management, and more particularly to an edge device disk management method and related equipment. Background Art

[0002] In edge computing environments, existing solutions usually use the same edge device disk management solution for the performance and resource requirements of different business applications. However, this approach is difficult to meet the current requirements for edge device disk efficiency, flexibility, and reliability, especially when dealing with bandwidth-intensive applications and non-bandwidth-intensive applications. It cannot ensure the optimal performance of bandwidth-intensive applications and the resource requirements of non-bandwidth-intensive applications, which can easily lead to application performance bottlenecks, low disk resource utilization, and unstable disk resource allocation. Summary of the invention

[0003] In view of this, the present invention discloses an edge device disk management method and related devices to achieve optimal performance of bandwidth-intensive applications and resource requirements of non-bandwidth-intensive applications.

[0004] A disk management method for an edge device, comprising:

[0005] Based on at least one dimension of network bandwidth demand of the current business application, determining whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application;

[0006] If the current business application is the bandwidth-intensive application, a first disk resource required by the current business application is allocated from the first edge device by adopting a bare disk mapping method;

[0007] If the current business application is the non-bandwidth intensive application, the disks of the second edge device are pooled to obtain a storage pool, and the second disk resources required for the current business application are allocated from the storage pool, wherein the first edge device and the second edge device are the same edge device or different edge devices.

[0008] Optionally, if the current business application is the bandwidth-intensive application, allocating the first disk resource required by the current business application from the first edge device in a bare disk mapping manner includes:

[0009] Deploy the current business application to the first edge device;

[0010] Scan all physical disks where non-root directories are located from the first edge device as the first disk resources;

[0011] The first disk resource is allocated to the current business application for use by adopting a raw disk mapping method.

[0012] Optionally, deploying the current service application on the first edge device includes:

[0013] Deploy the current business application to the first edge device using a daemon mechanism;

[0014] Specifically include:

[0015] an edge device having a binding relationship with the current business application as a first edge device;

[0016] When the current business application on the cloud changes, the proxy service on the first edge device is controlled to download the current business application on the cloud to the first edge device and deploy it.

[0017] Optionally, before the edge device having a binding relationship with the current business application is used as the first edge device, the method further includes:

[0018] Selecting a first edge device from all edge devices that meets the network bandwidth requirement of at least one dimension of the current business application;

[0019] A binding relationship between the first edge device and the current business application is established.

[0020] Optionally, if the current business application is the non-bandwidth intensive application, pooling the disks of the second edge device to obtain a storage pool, and allocating the second disk resources required by the current business application from the storage pool includes:

[0021] Determine, by using a cloud scheduling module, from the edge devices that upload resource data, an edge device that meets the requirements of the current business application as the second edge device;

[0022] Based on a disk allocation mechanism, pooling the disks of the second edge device to obtain the storage pool;

[0023] The second disk resource required by the current business application is allocated from the storage pool.

[0024] Optionally, based on the network bandwidth requirement of at least one dimension of the current business application, determining whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application includes:

[0025] Obtaining network bandwidth requirements of at least one dimension of the current business application;

[0026] The network bandwidth requirement of the at least one dimension is matched with the preset bandwidth requirement of the bandwidth-intensive application and the preset bandwidth requirement of the non-bandwidth-intensive application respectively, and it is determined whether the current business application is the bandwidth-intensive application or the non-bandwidth-intensive application.

[0027] Optionally, matching the network bandwidth requirement of the at least one dimension with a preset bandwidth-intensive application bandwidth requirement and a preset non-bandwidth-intensive application bandwidth requirement, respectively, and determining whether the current service application is the bandwidth-intensive application or the non-bandwidth-intensive application, includes:

[0028] In a case where the network bandwidth demand in at least one dimension is a multi-dimensional network bandwidth demand, determining a weighted sum result of the multi-dimensional network bandwidth demand according to a preset weight value corresponding to the network bandwidth demand in each dimension;

[0029] Matching the weighted sum result with the weighted sum threshold in the preset bandwidth-intensive application bandwidth requirement and the weighted sum threshold in the preset non-bandwidth-intensive application bandwidth requirement respectively to obtain a matching result;

[0030] The current service application is determined to be the bandwidth-intensive application or the non-bandwidth-intensive application according to the matching result.

[0031] Optionally, matching the network bandwidth requirement of the at least one dimension with a preset bandwidth-intensive application bandwidth requirement and a preset non-bandwidth-intensive application bandwidth requirement, respectively, and determining whether the current service application is the bandwidth-intensive application or the non-bandwidth-intensive application, includes:

[0032] In a case where the network bandwidth demand in the at least one dimension is a multi-dimensional network bandwidth demand, calculating a first similarity value between the multi-dimensional network bandwidth demand and the preset bandwidth-intensive application bandwidth demand;

[0033] Calculating a second similarity value between the multi-dimensional network bandwidth requirement and the preset non-bandwidth-intensive application bandwidth requirement;

[0034] Based on the size relationship between the first similarity value and the second similarity value, the current business application is judged to be the bandwidth-intensive application or the non-bandwidth-intensive application, wherein the application corresponding to the higher similarity value between the first similarity value and the second similarity value is used as the determination result of the current business application.

[0035] Optionally, the network bandwidth requirement of at least one dimension is any one or more of: data transmission volume, network throughput, data transmission delay requirement, data synchronization frequency, transmitted multimedia content, and transmission file size.

[0036] An edge device disk management device, comprising:

[0037] A determination unit, configured to determine whether the current service application is a bandwidth-intensive application or a non-bandwidth-intensive application based on at least one dimension of network bandwidth demand of the current service application;

[0038] A first allocation unit is configured to allocate a first disk resource required by the current business application from a first edge device by adopting a bare disk mapping method if the current business application is the bandwidth-intensive application;

[0039] The second allocation unit is used to pool the disks of the second edge device to obtain a storage pool if the current business application is the non-bandwidth intensive application, and allocate the second disk resources required by the current business application from the storage pool, wherein the first edge device and the second edge device are the same edge device or different edge devices.

[0040] A computer storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the edge device disk management method described above is implemented.

[0041] An electronic device, comprising: a memory and a processor;

[0042] The memory is used to store at least one instruction;

[0043] The processor is used to execute the at least one instruction to implement the above-mentioned edge device disk management method.

[0044] As can be seen from the above technical solutions, the present invention discloses a disk management method for edge devices and related devices. Based on the network bandwidth requirements of at least one dimension of the current business application, it is judged whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application. If the current business application is a bandwidth-intensive application, a bare disk mapping method is used to allocate the first disk resources required by the current business application from the first edge device. The bare disk mapping can enable the bandwidth-intensive application to directly read and write the host file system, reduce the overhead of the intermediate layer, and ensure the best performance of the bandwidth-intensive application. If the current business application is a non-bandwidth-intensive application, the disk of the second edge device is pooled to obtain a storage pool, and the second disk resources required by the current business application are allocated from the storage pool. By pooling the disk of the second edge device, the system performance and resource utilization can be improved, so that non-bandwidth-intensive applications can be dynamically and flexibly allocated storage resources as needed. The present invention adopts different disk resource allocation strategies for different business application types to carry out refined classification management of disk resources, realizes dynamic and efficient allocation of disk resources, effectively solves the problems of application performance bottlenecks, low disk resource utilization, and unstable disk resource allocation, thereby meeting the performance and disk resource requirements of different business applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the disclosed drawings without paying creative work.

[0046] Figure 1 A flow chart of a disk management method for edge devices disclosed in an embodiment of the present invention;

[0047] Figure 2 A flow chart of a method for allocating first disk resources disclosed in an embodiment of the present invention;

[0048] Figure 3 A flow chart of a method for allocating second disk resources disclosed in an embodiment of the present invention;

[0049] Figure 4 A schematic diagram of the structure of a disk management device for an edge device disclosed in an embodiment of the present invention;

[0050] Figure 5 The present invention is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The embodiment of the present invention discloses a disk management method for edge devices and related devices. Based on the network bandwidth requirements of at least one dimension of the current business application, it is judged whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application. If the current business application is a bandwidth-intensive application, a first disk resource required by the current business application is allocated from the first edge device by a bare disk mapping method. The bare disk mapping can enable the bandwidth-intensive application to directly read and write the host file system, reduce the overhead of the intermediate layer, and ensure the best performance of the bandwidth-intensive application. If the current business application is a non-bandwidth-intensive application, the disk of the second edge device is pooled to obtain a storage pool, and the second disk resource required by the current business application is allocated from the storage pool. By pooling the disk of the second edge device, the system performance and resource utilization can be improved, so that the non-bandwidth-intensive application can dynamically and flexibly allocate storage resources as needed. The present invention adopts different disk resource allocation strategies for different business application types to perform refined classification management of disk resources, realizes dynamic and efficient allocation of disk resources, effectively solves the problems of application performance bottlenecks, low disk resource utilization, and unstable disk resource allocation, thereby meeting the performance and disk resource requirements of different business applications.

[0053] See also Figure 1 , a flow chart of a disk management method for an edge device disclosed in an embodiment of the present invention, the method comprising:

[0054] Step S101: Based on at least one dimension of network bandwidth demand of a current business application, determine whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application.

[0055] Among them, the network bandwidth demand of at least one dimension of the current business application includes but is not limited to any one or more of data transmission volume, network throughput, data transmission delay requirement, data synchronization frequency, transmitted multimedia content, and transmission file size.

[0056] Specifically, (1) bandwidth-intensive applications refer to applications that require large amounts of data transmission and usually have high demands on network bandwidth. Such applications require a stable and high-speed network connection during operation to ensure their performance and user experience. Bandwidth-intensive applications can meet any one or more of the following conditions:

[0057] High data transfer: Applications transfer large amounts of data (such as video streaming, online gaming, etc.).

[0058] High network throughput: The application has high requirements for network throughput.

[0059] Low data transmission latency requirement: Applications have strict requirements on data transmission latency (such as video conferencing and online games).

[0060] High data synchronization frequency: The application requires frequent and large-scale data synchronization.

[0061] Multimedia content transmission: Applications transmit large amounts of audio and video data (such as live broadcasts and video on demand).

[0062] Large file transfer: The application needs to transfer large files (such as online storage services) during runtime.

[0063] (2) Non-bandwidth-intensive applications refer to applications that transmit a small amount of data (such as text processing and database queries). Non-bandwidth-intensive applications can meet any one or more of the following conditions:

[0064] Low network throughput: The application does not require high network throughput.

[0065] High tolerance for network delay: The application has a high tolerance for data transmission delay.

[0066] Periodic or intermittent data transmission: Applications transmit small amounts of data periodically or intermittently.

[0067] Simple file transfer: The application mainly transfers small files (such as email clients and simple file downloads).

[0068] Step S102: If the current service application is the bandwidth-intensive application, a first disk resource required by the current service application is allocated from a first edge device by adopting a bare disk mapping method.

[0069] Raw disk mapping, also known as Raw Disk Mapping (RDM) or bare metal mapping, is a technology that maps physical storage devices directly to virtual machines. Raw disk mapping refers to mapping physical storage devices (such as hard disks, SSDs, etc.) directly to virtual machines to form a virtual storage device. This technology improves storage performance and access speed by bypassing the operating system kernel to directly access the hardware. Raw disk mapping is usually used in scenarios that require high-performance storage, such as database servers, high-performance computing clusters, etc. In these scenarios, the requirements for storage performance and access speed are high, and the raw disk mapping method can provide direct and efficient storage access.

[0070] This application adopts the raw disk mapping method to enable the current business application to directly read and write the host file system, thereby reducing the overhead of the intermediate layer and improving the disk I / O efficiency.

[0071] Edge devices are devices that provide entry points into an enterprise or service provider core network.

[0072] Bandwidth-intensive applications such as video surveillance uploads and data collection require extremely high disk I / O performance, that is, bandwidth-intensive applications have high requirements for disk performance and reliability. Based on this, this application uses a bare disk mapping method to directly map to the first disk resource of each edge device when the current business application is a bandwidth-intensive application to ensure the best performance of bandwidth-intensive applications. The bare disk mapping method maps the physical storage device directly to the virtual machine to form a virtual storage device. By bypassing the operating system kernel to directly access the hardware, the current business application of the bandwidth-intensive application type can directly read and write the host file system, thereby reducing the overhead of the intermediate layer and improving the disk I / O efficiency.

[0073] In actual applications, the raw disk mapping method may be the HostPath method.

[0074] HostPath is a type of storage volume in Kubernetes (K8s) that allows files or directories in the file system to be mounted directly into the Pod, allowing the Pod to directly access and operate the file system on the host.

[0075] The working principle of allocating the first disk resource required by the current business application from the first edge device using the HostPath method is as follows: scan all physical disks where non-root directories in the first edge device are located as the first disk resources. After the current business application is deployed to the first edge device, the edge proxy service will mount the scanned first disk resources to the Pod specified by the current business application to realize disk resource allocation.

[0076] Step S103: If the current business application is the non-bandwidth intensive application, pool the disks of the second edge device to obtain a storage pool, and allocate the second disk resources required by the current business application from the storage pool.

[0077] The first edge device and the second edge device may be the same edge device, or the first edge device and the second edge device may be different edge devices, depending on actual needs.

[0078] The principle of disk pooling is to use virtualization technology to integrate multiple physical disks into a unified storage pool to achieve unified management and efficient use of storage resources. Through resource integration, logical abstraction, unified management and efficient use, disk pooling can improve storage utilization, reduce management complexity, enhance data security and improve performance.

[0079] The role of disk pooling is mainly reflected in improving system performance and resource utilization. Disk pooling technology allows an available resource to be obtained from the resource pool for use, and then returned to the pool for other tasks to continue using after use, thereby reducing resource waste and achieving resource reuse.

[0080] Storage pool is a technology that combines multiple storage devices (such as hard disks, solid-state disks, etc.) to form a logical unified storage space. It provides a unified management interface for storage resources, facilitating operations such as adding, deleting, and expanding storage resources.

[0081] The principle of allocating the second disk resource required by the current business application from the storage pool is: divide the physical storage device into several storage resources, each of which has independent storage capacity and access characteristics. These storage resources can be customized according to different requirements, such as setting different parameters such as capacity, performance, and reliability. When a business application needs disk resources, it will put forward corresponding storage requirements, including the required storage capacity, performance, etc. The storage pool management system will find suitable storage resources in the storage pool for allocation based on these requirements.

[0082] Non-bandwidth-intensive applications do not require high disk I / O performance, but require flexible and dynamic resource configuration, such as log storage, small databases, etc. Therefore, non-bandwidth-intensive applications have high requirements for the adjustment function of disk usage to achieve dynamic and flexible storage resource allocation and resource release. Based on this, the application pools the disks of the second edge device to obtain a storage pool, and allocates the second disk resources required by the current business application from the storage pool.

[0083] In summary, the present invention discloses a disk management method for edge devices. Based on the network bandwidth requirements of at least one dimension of the current business application, the current business application is judged to be a bandwidth-intensive application or a non-bandwidth-intensive application. If the current business application is a bandwidth-intensive application, a first disk resource required for the current business application is allocated from the first edge device by a bare disk mapping method. The bare disk mapping can enable the bandwidth-intensive application to directly read and write the host file system, reduce the overhead of the intermediate layer, and ensure the optimal performance of the bandwidth-intensive application. If the current business application is a non-bandwidth-intensive application, the disk of the second edge device is pooled to obtain a storage pool, and the second disk resource required for the current business application is allocated from the storage pool. By pooling the disk of the second edge device, the system performance and resource utilization can be improved, so that non-bandwidth-intensive applications can be dynamically and flexibly allocated storage resources as needed. The present invention adopts different disk resource allocation strategies for different business application types to carry out refined classification management of disk resources, realizes dynamic and efficient allocation of disk resources, effectively solves the problems of application performance bottlenecks, low disk resource utilization, and unstable disk resource allocation, thereby meeting the performance and disk resource requirements of different business applications.

[0084] In one embodiment, see Figure 2 , a flow chart of a method for allocating a first disk resource disclosed in an embodiment of the present invention, the method comprising:

[0085] Step S201: deploy the current business application to the first edge device.

[0086] In actual applications, the daemon mechanism can be used to deploy the current business application to the first edge device. The deployment process may specifically include:

[0087] An edge device that has a binding relationship with the current business application is used as a first edge device;

[0088] When the current business application on the cloud changes, the proxy service on the first edge device is controlled to download the current business application on the cloud to the first edge device and deploy it.

[0089] In actual applications, an edge device that meets the network bandwidth requirements of the current business application can be bound to the current business application and used as the first edge device.

[0090] Changes to current business applications on the cloud include, but are not limited to, changes to the versions of current business applications.

[0091] Preferably, the daemon mechanism may be the DaemonSet mechanism of Kubernetes.

[0092] Kubernetes is an open source system for automating the deployment, expansion and management of containerized applications. It provides a lightweight and portable way to manage containerized applications and services, making application deployment and management simpler, more efficient and reliable.

[0093] In Kubernetes, DaemonSet is a special workload controller that ensures that a copy of a Pod runs on every node (or certain specific nodes) in the cluster.

[0094] "Pod" is a very important concept in Kubernetes (K8s). It is the smallest and deployable computing unit running in a Kubernetes cluster. A Pod consists of a group of containers (usually closely related and sharing some resources) that are deployed on a single node and managed as a whole.

[0095] The principle of deploying the current business application to the first edge device using the DaemonSet mechanism of Kubernetes is as follows: The binding between the first edge device and the current business application is achieved through labels. First, label the first edge device, and then set the corresponding selector when deploying the current business application. The agent service on the first edge device will periodically check whether the current business application on the cloud has changed to determine whether the first edge device needs to deploy the current business application. If a change in the current business application is detected, the agent service on the first edge device will download the current business application on the cloud to the first edge device and deploy it.

[0096] Step S202: Scan all physical disks where non-root directories are located from the first edge device as first disk resources.

[0097] Non-root directories refer to all directories in the file system except the root directory ( / ). In Linux or Unix-like operating systems, the file system is organized in a tree structure, with the root directory ( / ) at the top level and all other directories and files under the root directory. Non-root directories are organized according to specific levels and rules to facilitate user access and management of files and directories.

[0098] In this embodiment, all physical disks where non-root directories are located are used as the first disk resources mapped to bandwidth-intensive applications.

[0099] Step S203: using a bare disk mapping method, allocating the first disk resource to the current business application.

[0100] In summary, the present invention deploys the current business application of the bandwidth-intensive application type on the first edge device, scans all physical disks where non-root directories are located from the first edge device as the first disk resource, and uses the bare disk mapping method to allocate the first disk resource to the current business application. Because the bare disk mapping method allows the bandwidth-intensive application to directly read and write the host file system, the overhead of the intermediate layer is reduced, ensuring the best performance of the bandwidth-intensive application.

[0101] In one embodiment, before an edge device having a binding relationship with a current business application is used as a first edge device, the following may also be included:

[0102] A first edge device that meets at least one dimension of network bandwidth requirements of a current business application is selected from all edge devices, and a binding relationship between the first edge device and the current business application is established.

[0103] In one embodiment, see Figure 3 , a flow chart of a method for allocating second disk resources disclosed in an embodiment of the present invention, the method comprising:

[0104] Step S301: Use the cloud scheduling module to determine an edge device that meets the requirements of the current business application from the edge devices that upload resource data as the second edge device.

[0105] In actual applications, resource data such as CPU (Central Processing Unit), memory, disk, etc. are regularly collected on each edge device, and the collected resource data is uploaded to the cloud scheduling module so that the cloud scheduling module can perform scheduling, resource planning and strategy optimization according to the resource data of each edge device, thereby improving the overall intelligence and agility of the system.

[0106] Non-bandwidth-intensive applications refer to applications that transmit a small amount of data (such as text processing, database query). Non-bandwidth-intensive applications can meet any one or more of low network throughput, high tolerance for network delay, periodic or intermittent data transmission, and simple file transfer. Based on this, when the current business application is a non-bandwidth-intensive application, the cloud scheduling module can determine the edge device that meets the disk resource demand as the second edge device based on the disk resource demand of the non-bandwidth-intensive application and the resource data of each edge device.

[0107] Step S302: Using a disk allocation mechanism, pool the disks of the second edge device to obtain a storage pool.

[0108] Step S303: Allocate the second disk resource required by the current business application from the storage pool.

[0109] Disk allocation mechanisms include but are not limited to continuous allocation, linked allocation, index allocation, etc. Disk pooling, also known as storage pooling, is a technology that integrates multiple physical disk resources into a logical storage unit (i.e., storage pool). By using the disk allocation mechanism and disk pooling, the system can dynamically allocate and manage storage resources to improve storage utilization and flexibility.

[0110] Preferably, the disk allocation mechanism may be a PVC (Persistent Volume Claim) mechanism.

[0111] PVC (Persistent Volume Claim) is a resource object in Kubernetes that represents a request for persistent storage resources. It is an abstract representation of storage resources by users or developers, used to describe the storage characteristics required by Pods, such as capacity, access mode, etc. PVC enables Pods to request and use persistent storage resources without having to worry about the specific implementation of storage resources. This improves the flexibility and portability of storage resources, allowing Pods to reuse the same storage requests in different environments.

[0112] The principle of this embodiment of the present application using the PVC mechanism to pool the disk of the second edge device to obtain a storage pool is mainly related to the combination of Kubernetes storage management and pooling technology. Through the PVC mechanism, Kubernetes integrates multiple PVs into a unified storage pool and allows users to access resources in the storage pool through PVC. This design improves storage utilization, simplifies storage management, and supports cross-node access to storage resources, which can ensure that non-bandwidth-intensive applications can use disk resources dynamically and flexibly.

[0113] Among them, PV (PersistentVolume) is a specific storage resource abstraction in the Kubernetes cluster, which describes the properties of the storage volume, such as capacity, access mode, storage category, etc. PV is usually created and managed by the cluster administrator, representing a physical or virtual disk that can be used to store data in the cluster.

[0114] In actual applications, when a new edge device is connected to the edge computing environment, the number of physical disks on the edge device must be detected first. If the number of physical disks on the edge device is not less than two, one of the physical disks can be selected and the pooling technology can be used to pool the selected physical disk space to obtain the corresponding storage pool to achieve flexible resource allocation.

[0115] The pooling technology may be: LVM or ZFS.

[0116] LVM (Logical Volume Manager) is a disk logical volume management tool under Linux, which provides a flexible way to manage disk space. LVM allows the file system to be resized dynamically without repartitioning or formatting the disk, so it is very important for environments that need to frequently adjust storage requirements.

[0117] The principle of using LVM to pool the selected physical disk space is as follows: LVM pools the physical disk space by combining multiple physical volumes into one or more volume groups. Among them, the physical volume is created on the physical disk space and represents part or all of the physical disk space. This pooled structure allows administrators to flexibly allocate storage resources according to application requirements and improve the utilization of storage space.

[0118] ZFS (Zettabyte File System), also known as Dynamic File System, is an advanced file system designed to provide almost unlimited storage space. ZFS not only supports high-capacity data storage, but also has many advantages such as automatic verification of data integrity, self-optimization, and easy management of storage pool / volume systems.

[0119] The principle of using ZFS to pool the selected physical disk space is: mainly based on the storage pool concept of ZFS and its related technical characteristics, through disk aggregation, space allocation and management, metadata management, fault tolerance and redundancy and other technical means, to achieve efficient use and flexible management of storage resources.

[0120] In summary, the present invention pools the second edge devices required by the current business applications, and allocates applicable second disk resources to the current business applications of non-bandwidth intensive application type from the storage pool obtained after pooling. Since disk pooling can improve system performance and resource utilization, the current business applications can dynamically and flexibly allocate storage resources as needed.

[0121] In one embodiment, step S101 may specifically include:

[0122] Obtaining the network bandwidth requirement of at least one dimension of the current business application; matching the network bandwidth requirement of at least one dimension with the preset bandwidth-intensive application bandwidth requirement and the preset bandwidth requirement of non-bandwidth-intensive application, respectively, to determine whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application.

[0123] The network bandwidth requirements of at least one dimension of the current business application can be found in Figure 1 The corresponding parts of the illustrated embodiment will not be described in detail here.

[0124] It should be noted that when the network bandwidth demand of at least one dimension of the current business application includes only one network bandwidth demand, the network bandwidth demand of the current business application is directly matched with the preset bandwidth-intensive application bandwidth demand and the preset non-bandwidth-intensive application bandwidth demand for similarity, and based on the matched application type, it can be determined whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application.

[0125] In the case where at least one dimension of network bandwidth demand of the current business application is a multi-dimensional network bandwidth demand, the following two determination methods may be used.

[0126] First method:

[0127] Determine a weighted sum of the multi-dimensional network bandwidth requirements according to a preset weight value corresponding to the network bandwidth requirements of each dimension;

[0128] Matching the weighted sum result with the weighted sum threshold in the preset bandwidth-intensive application bandwidth requirement and the weighted sum threshold in the preset non-bandwidth-intensive application bandwidth requirement respectively to obtain a matching result;

[0129] The current service application is determined to be the bandwidth-intensive application or the non-bandwidth-intensive application according to the matching result.

[0130] In this embodiment, the matching result may be characterized as whether the application closest to the current service application is a bandwidth-intensive application or a non-bandwidth-intensive application, as the final matching result.

[0131] In actual applications, different preset weight values ​​can be set for each dimension of network bandwidth demand according to its impact on data transmission. For example, the data transmission volume has a greater impact on the data transmission of the current business application than the data synchronization frequency. The weight value corresponding to the data transmission volume can be set higher than the weight value corresponding to the data synchronization frequency.

[0132] The value of the network bandwidth requirement in each dimension is determined according to actual needs, for example, the value is 1, 2, etc.

[0133] In actual applications, the weighted sum threshold in the bandwidth demand of preset bandwidth-intensive applications and the weighted sum threshold in the bandwidth demand of preset non-bandwidth-intensive applications are determined according to actual needs. Since bandwidth-intensive applications refer to applications that require a large amount of data transmission and usually have high demands on network bandwidth, and non-bandwidth-intensive applications refer to applications that transmit a small amount of data and usually have low demands on network bandwidth, the value of the weighted sum threshold in the bandwidth demand of preset bandwidth-intensive applications can be set higher than the weighted sum threshold in the bandwidth demand of preset non-bandwidth-intensive applications.

[0134] Second method:

[0135] Calculating a first similarity value between the multi-dimensional network bandwidth requirement and the preset bandwidth-intensive application bandwidth requirement;

[0136] Calculating a second similarity value between the multi-dimensional network bandwidth requirement and the preset non-bandwidth-intensive application bandwidth requirement;

[0137] Based on the magnitude relationship between the first similarity value and the second similarity value, it is determined whether the current service application is the bandwidth-intensive application or the non-bandwidth-intensive application.

[0138] The bandwidth requirements of bandwidth-intensive applications can be based on Figure 1 The conditions that bandwidth-intensive applications can meet in the implementation shown in the figure are determined. Similarly, the bandwidth requirements of non-bandwidth-intensive applications can be determined based on Figure 1 The conditions that can be satisfied by non-bandwidth intensive applications in the illustrated implementation are determined.

[0139] In practical applications, the cosine similarity method can be used to calculate the first similarity value between the multi-dimensional network bandwidth demand and the preset bandwidth-intensive application bandwidth demand, and the second similarity value between the multi-dimensional network bandwidth demand and the preset non-bandwidth-intensive application bandwidth demand. In this embodiment, the application corresponding to the higher similarity value between the first similarity value and the second similarity value is used as the determination result of the current business application.

[0140] It should be noted that, when the network bandwidth demand of at least one dimension of the current business application is a multi-dimensional network bandwidth demand, the method for determining whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application includes but is not limited to the two methods shown above. Other methods that can respectively match the multi-dimensional network bandwidth demand of the current business application with the preset bandwidth-intensive application bandwidth demand and the preset non-bandwidth-intensive application bandwidth demand to obtain the determination method of whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application all fall within the scope of protection of this application.

[0141] Corresponding to the above method embodiment, the present invention also discloses an edge device disk management device.

[0142] See also Figure 4 , a schematic diagram of the structure of an edge device disk management device disclosed in an embodiment of the present invention, the device may include:

[0143] The determination unit 401 is configured to determine whether the current service application is a bandwidth-intensive application or a non-bandwidth-intensive application based on at least one dimension of network bandwidth demand of the current service application.

[0144] Among them, at least one dimension of network bandwidth demand is: any one or more of: data transmission volume, network throughput, data transmission delay requirement, data synchronization frequency, transmitted multimedia content, and transmission file size.

[0145] A first allocation unit 402 is configured to allocate a first disk resource required by the current business application from a first edge device by adopting a bare disk mapping method if the current business application is the bandwidth-intensive application;

[0146] The second allocation unit 403 is used to pool the disks of the second edge device to obtain a storage pool if the current business application is the non-bandwidth intensive application, and allocate the second disk resources required by the current business application from the storage pool, wherein the first edge device and the second edge device are the same edge device or different edge devices.

[0147] In summary, the present invention discloses an edge device disk management device, which determines whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application based on the network bandwidth demand of at least one dimension of the current business application. If the current business application is a bandwidth-intensive application, a first disk resource required by the current business application is allocated from the first edge device by a bare disk mapping method. The bare disk mapping can enable the bandwidth-intensive application to directly read and write the host file system, reduce the overhead of the intermediate layer, and ensure the best performance of the bandwidth-intensive application. If the current business application is a non-bandwidth-intensive application, the disk of the second edge device is pooled to obtain a storage pool, and the second disk resource required by the current business application is allocated from the storage pool. By pooling the disk of the second edge device, the system performance and resource utilization can be improved, so that non-bandwidth-intensive applications can be dynamically and flexibly allocated storage resources as needed. The present invention adopts different disk resource allocation strategies for different business application types to perform refined classification management of disk resources, realizes dynamic and efficient allocation of disk resources, effectively solves the problems of application performance bottlenecks, low disk resource utilization, and unstable disk resource allocation, thereby meeting the performance and disk resource requirements of different business applications.

[0148] In one embodiment, the first allocation unit 402 includes:

[0149] A deployment subunit, configured to deploy the current service application to the first edge device;

[0150] A scanning subunit, used to scan all physical disks where non-root directories are located from the first edge device as the first disk resources;

[0151] The allocation subunit is used to allocate the first disk resource to the current business application for use by adopting a raw disk mapping method.

[0152] In one embodiment, the deployment subunit may be specifically used for:

[0153] Deploy the current business application to the first edge device using a daemon mechanism;

[0154] In one embodiment, the deployment subunit may be specifically used for:

[0155] an edge device having a binding relationship with the current business application as a first edge device;

[0156] When the current business application on the cloud changes, the proxy service on the first edge device is controlled to download the current business application on the cloud to the first edge device and deploy it.

[0157] In one embodiment, the deployment subunit may be specifically used for:

[0158] Before the edge device having a binding relationship with the current business application is used as the first edge device, a first edge device that satisfies the network bandwidth requirement of at least one dimension of the current business application is selected from all edge devices;

[0159] A binding relationship between the first edge device and the current business application is established.

[0160] In one embodiment, the second allocation unit 403 may be specifically used for:

[0161] Determine, by using a cloud scheduling module, from the edge devices that upload resource data, an edge device that meets the requirements of the current business application as the second edge device;

[0162] Based on a disk allocation mechanism, pooling the disks of the second edge device to obtain the storage pool;

[0163] The second disk resource required by the current business application is allocated from the storage pool.

[0164] In one embodiment, the determining unit 401 may be specifically configured to:

[0165] Obtaining network bandwidth requirements of at least one dimension of the current business application;

[0166] The network bandwidth requirement of the at least one dimension is matched with the preset bandwidth requirement of the bandwidth-intensive application and the preset bandwidth requirement of the non-bandwidth-intensive application respectively, and it is determined whether the current business application is the bandwidth-intensive application or the non-bandwidth-intensive application.

[0167] In one embodiment, the determining unit 401 may be further configured to:

[0168] In a case where the network bandwidth demand in at least one dimension is a multi-dimensional network bandwidth demand, determining a weighted sum result of the multi-dimensional network bandwidth demand according to a preset weight value corresponding to the network bandwidth demand in each dimension;

[0169] Matching the weighted sum result with the weighted sum threshold in the preset bandwidth-intensive application bandwidth requirement and the weighted sum threshold in the preset non-bandwidth-intensive application bandwidth requirement respectively to obtain a matching result;

[0170] The current service application is determined to be the bandwidth-intensive application or the non-bandwidth-intensive application according to the matching result.

[0171] In one embodiment, the determining unit 401 may be further configured to:

[0172] In a case where the network bandwidth demand in the at least one dimension is a multi-dimensional network bandwidth demand, calculating a first similarity value between the multi-dimensional network bandwidth demand and the preset bandwidth-intensive application bandwidth demand;

[0173] Calculating a second similarity value between the multi-dimensional network bandwidth requirement and the preset non-bandwidth-intensive application bandwidth requirement;

[0174] Based on the size relationship between the first similarity value and the second similarity value, the current business application is judged to be the bandwidth-intensive application or the non-bandwidth-intensive application, wherein the application corresponding to the higher similarity value between the first similarity value and the second similarity value is used as the determination result of the current business application.

[0175] It should be noted that, for the specific working principles of each component in the device embodiment, please refer to the corresponding part of the method embodiment, which will not be repeated here.

[0176] Corresponding to the above embodiment, the present invention further discloses a computer storage medium, which stores at least one instruction, and when the at least one instruction is executed by a processor, the steps shown in the embodiment of the edge device disk management method are implemented.

[0177] Computer storage media may be tangible media that may contain or store programs for use with or in conjunction with an instruction execution system, device, or apparatus. Computer storage media may be machine-readable signal media or machine-readable storage media. Computer storage media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0178] Corresponding to the above embodiment, Figure 5 As shown, the present invention also provides an electronic device, which may include: a processor 1 and a memory 2;

[0179] The processor 1 and the memory 2 communicate with each other via a communication bus 3;

[0180] Processor 1, configured to execute at least one instruction;

[0181] Memory 2, used to store at least one instruction;

[0182] The processor 1 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0183] The memory 2 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0184] The processor executes at least one instruction to implement the steps shown in the embodiment of the edge device disk management method.

[0185] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0186] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0187] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A disk management method for an edge device, characterized in that: include: Based on at least one dimension of network bandwidth demand of the current business application, determining whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application; If the current business application is the bandwidth-intensive application, a first disk resource required by the current business application is allocated from the first edge device by adopting a bare disk mapping method; If the current business application is the non-bandwidth intensive application, the disks of the second edge device are pooled to obtain a storage pool, and the second disk resources required for the current business application are allocated from the storage pool, wherein the first edge device and the second edge device are the same edge device or different edge devices.

2. The edge device disk management method according to claim 1, characterized in that: If the current business application is the bandwidth-intensive application, allocating the first disk resource required by the current business application from the first edge device in a bare disk mapping manner includes: Deploy the current business application to the first edge device; Scan all physical disks where non-root directories are located from the first edge device as the first disk resources; The first disk resource is allocated to the current business application for use by adopting a raw disk mapping method.

3. The edge device disk management method according to claim 2, characterized in that: The deploying the current service application on the first edge device includes: Deploy the current business application to the first edge device using a daemon mechanism; Specifically include: an edge device having a binding relationship with the current business application as a first edge device; When the current business application on the cloud changes, the proxy service on the first edge device is controlled to download the current business application on the cloud to the first edge device and deploy it.

4. The edge device disk management method according to claim 3, characterized in that: Before the edge device having a binding relationship with the current business application is used as the first edge device, the method further includes: Selecting a first edge device from all edge devices that meets the network bandwidth requirement of at least one dimension of the current business application; A binding relationship between the first edge device and the current business application is established.

5. The edge device disk management method according to claim 1, characterized in that: If the current business application is the non-bandwidth intensive application, pooling the disks of the second edge device to obtain a storage pool, and allocating the second disk resources required by the current business application from the storage pool, including: Determine, by using a cloud scheduling module, from the edge devices that upload resource data, an edge device that meets the requirements of the current business application as the second edge device; Based on a disk allocation mechanism, pooling the disks of the second edge device to obtain the storage pool; The second disk resource required by the current business application is allocated from the storage pool.

6. The edge device disk management method according to any one of claims 1 to 5, characterized in that: Based on at least one dimension of network bandwidth demand of the current business application, determining whether the current business application is a bandwidth-intensive application or a non-bandwidth-intensive application includes: Obtaining network bandwidth requirements of at least one dimension of the current business application; The network bandwidth requirement of the at least one dimension is matched with the preset bandwidth requirement of the bandwidth-intensive application and the preset bandwidth requirement of the non-bandwidth-intensive application respectively, and it is determined whether the current business application is the bandwidth-intensive application or the non-bandwidth-intensive application.

7. The edge device disk management method according to claim 6, characterized in that: The matching of the network bandwidth requirement of the at least one dimension with the preset bandwidth-intensive application bandwidth requirement and the preset non-bandwidth-intensive application bandwidth requirement, respectively, and determining whether the current service application is the bandwidth-intensive application or the non-bandwidth-intensive application, includes: In a case where the network bandwidth demand in at least one dimension is a multi-dimensional network bandwidth demand, determining a weighted sum result of the multi-dimensional network bandwidth demand according to a preset weight value corresponding to the network bandwidth demand in each dimension; Matching the weighted sum result with the weighted sum threshold in the preset bandwidth-intensive application bandwidth requirement and the weighted sum threshold in the preset non-bandwidth-intensive application bandwidth requirement respectively to obtain a matching result; The current service application is determined to be the bandwidth-intensive application or the non-bandwidth-intensive application according to the matching result.

8. The edge device disk management method according to claim 6, characterized in that: The matching of the network bandwidth requirement of the at least one dimension with the preset bandwidth-intensive application bandwidth requirement and the preset non-bandwidth-intensive application bandwidth requirement, respectively, and determining whether the current service application is the bandwidth-intensive application or the non-bandwidth-intensive application, includes: In a case where the network bandwidth requirement of the at least one dimension is a multi-dimensional network bandwidth requirement, calculating a first similarity value between the multi-dimensional network bandwidth requirement and the preset bandwidth-intensive application bandwidth requirement; Calculating a second similarity value between the multi-dimensional network bandwidth requirement and the preset non-bandwidth-intensive application bandwidth requirement; Based on the size relationship between the first similarity value and the second similarity value, the current business application is judged to be the bandwidth-intensive application or the non-bandwidth-intensive application, wherein the application corresponding to the higher similarity value between the first similarity value and the second similarity value is used as the determination result of the current business application.

9. The edge device disk management method according to claim 6, characterized in that: The network bandwidth requirement in at least one dimension is any one or more of: data transmission volume, network throughput, data transmission delay requirement, data synchronization frequency, transmitted multimedia content, and transmission file size.

10. An edge device disk management device, characterized in that: include: A determination unit, configured to determine whether the current service application is a bandwidth-intensive application or a non-bandwidth-intensive application based on at least one dimension of network bandwidth demand of the current service application; A first allocation unit is configured to allocate a first disk resource required by the current business application from a first edge device by adopting a bare disk mapping method if the current business application is the bandwidth-intensive application; The second allocation unit is used to pool the disks of the second edge device to obtain a storage pool if the current business application is the non-bandwidth intensive application, and allocate the second disk resources required by the current business application from the storage pool, wherein the first edge device and the second edge device are the same edge device or different edge devices.

11. A computer storage medium, characterized in that: The computer storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, the edge device disk management method as described in any one of claims 1 to 9 is implemented.

12. An electronic device, characterized in that: The electronic device comprises: a memory and a processor; The memory is used to store at least one instruction; The processor is used to execute the at least one instruction to implement the edge device disk management method as described in any one of claims 1 to 9.

Images