Cloud storage processing method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a cloud storage processing method and device, computer equipment, a computer readable storage medium and a computer program product, and relates to the technical field of cloud storage. The cloud disk access delay can be reduced, and the cloud disk storage access performance can be improved. The method comprises the steps of determining a block storage engine node where a cloud disk server is located; the block storage engine node starts a node optimizer; binding a service scheduling unit to a block storage engine node, and sending an address of the block storage engine node to a redirector; according to a login request of a cloud disk client mounted on the service scheduling unit, providing an address to the cloud disk client for login through a redirector, so that the service scheduling unit logs in the block storage engine node to perform input and output operation of the cloud disk service; wherein the node optimizer is used for starting a kernel service of the extended Berkley data packet filter at a block storage engine node, and forwarding an input / output request of an input / output operation at a kernel level.

Description

Technical Field

[0001] The present application relates to the field of cloud storage technology, and in particular to a cloud storage processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Most current cloud storage systems are based on traditional centralized storage architectures. Their default schedulers make the cloud disk network input and output (IO) paths too long, while over-reliance on third-party network plug-ins can easily increase the complexity and maintenance costs of cloud storage systems.

[0003] Therefore, the current cloud storage processing technology still needs to be improved in terms of cloud disk storage access performance. Summary of the invention

[0004] Based on this, it is necessary to provide a cloud storage processing method, apparatus, computer equipment, computer-readable storage medium and computer program product to address the above technical issues.

[0005] In a first aspect, the present application provides a cloud storage processing method, comprising:

[0006] Determine the block storage engine node where the cloud disk server is located; wherein the block storage engine node starts a node optimizer;

[0007] Binding a service scheduling unit to the block storage engine node, and sending the address of the block storage engine node to a redirector;

[0008] According to a login request of the cloud disk client mounted by the service scheduling unit, the address is provided to the cloud disk client through the redirector for login, so that the service scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk service;

[0009] The node optimizer is used to start the kernel service of the extended Berkeley packet filter in the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

[0010] In one of the embodiments, determining the block storage engine node where the cloud disk server is located includes: after monitoring the generated cloud disk description record through the cloud disk copy management service, determining the target block storage engine node according to the load information of each node; sending a volume creation request to the block storage engine interface of the target block storage engine node through the cloud disk copy management service; the block storage engine interface is used to call the remote procedure call interface provided by the target block storage engine node to execute volume creation and setting, and obtain the block storage engine node where the cloud disk server is located.

[0011] In one of the embodiments, after the generated cloud disk description record is monitored by the cloud disk copy management service, before the target block storage engine node is determined according to the load information of each node, it also includes: initiating a disk creation request to the management and control request processing interface through the cluster's persistent storage standard interface; generating the cloud disk description record according to the disk creation request through the management and control request processing interface.

[0012] In one of the embodiments, determining the target block storage engine node according to the load information of each node includes: determining a target number of nodes as the target block storage engine nodes according to the load information of each node; binding the service scheduling unit to the block storage engine node includes: binding the service scheduling unit to one of the block storage engine nodes.

[0013] In one of the embodiments, before determining the block storage engine node where the cloud disk server is located, the method further includes: after the cluster deployment is completed, triggering each of the block storage engine nodes of the cluster to start the node optimizer, so that the node optimizer starts the kernel service of the extended Berkeley packet filter.

[0014] In one of the embodiments, binding the business scheduling unit to the block storage engine node includes: when the cluster creates the business scheduling unit and mounts the persistent volume declaration corresponding to the cloud disk, identifying the business scheduling unit through the scheduling unit scheduler, and binding the business scheduling unit to the block storage engine node where the cloud disk server is located.

[0015] In a second aspect, the present application also provides a cloud storage processing device, including:

[0016] A node determination module is used to determine the block storage engine node where the cloud disk server is located; wherein the block storage engine node starts a node optimizer;

[0017] A scheduling binding module, used to bind the service scheduling unit to the block storage engine node, and send the address of the block storage engine node to the redirector;

[0018] A login processing module, configured to provide the address to the cloud disk client for login through the redirector according to a login request of the cloud disk client mounted by the service scheduling unit, so that the service scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk service;

[0019] The node optimizer is used to start the kernel service of the extended Berkeley packet filter in the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

[0020] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0021] Determine the block storage engine node where the cloud disk service end is located; wherein the block storage engine node starts a node optimizer; bind a business scheduling unit to the block storage engine node, and send the address of the block storage engine node to a redirector; according to a login request of a cloud disk client mounted by the business scheduling unit, provide the address to the cloud disk client through the redirector for login, so that the business scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk business; wherein the node optimizer is used to start a kernel service of an extended Berkeley packet filter at the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

[0022] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0023] Determine the block storage engine node where the cloud disk service end is located; wherein the block storage engine node starts a node optimizer; bind a business scheduling unit to the block storage engine node, and send the address of the block storage engine node to a redirector; according to a login request of a cloud disk client mounted by the business scheduling unit, provide the address to the cloud disk client through the redirector for login, so that the business scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk business; wherein the node optimizer is used to start a kernel service of an extended Berkeley packet filter at the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

[0024] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0025] Determine the block storage engine node where the cloud disk service end is located; wherein the block storage engine node starts a node optimizer; bind a business scheduling unit to the block storage engine node, and send the address of the block storage engine node to a redirector; according to a login request of a cloud disk client mounted by the business scheduling unit, provide the address to the cloud disk client through the redirector for login, so that the business scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk business; wherein the node optimizer is used to start a kernel service of an extended Berkeley packet filter at the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

[0026] The cloud storage processing method, apparatus, computer equipment, computer-readable storage medium and computer program product described above determine the block storage engine node where the cloud disk server is located; the block storage engine node starts the node optimizer; the business scheduling unit is bound to the block storage engine node, and the address of the block storage engine node is sent to the redirector; according to the login request of the cloud disk client mounted by the business scheduling unit, the address is provided to the cloud disk client through the redirector for login, so that the business scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk business; wherein the node optimizer is used to start the kernel service of the extended Berkeley packet filter at the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level. This solution can schedule the business scheduling unit to the block storage engine node where the cloud disk server is located, and combine it with the kernel service of the extended Berkeley packet filter started by starting the node optimizer at the block storage engine node of the cloud storage system. The requests corresponding to the input and output operations of the cloud disk business from this node and sent to this node are forwarded directly at the kernel level, thereby bypassing the protocol stack to shorten the network input and output path, reduce the cloud disk access latency, and improve the cloud disk storage access performance. In addition, the kernel service of the extended Berkeley packet filter is integrated in the block storage engine node, which can directly optimize the node without installing a third-party network plug-in, reducing the deployment complexity and maintenance cost of the cloud storage system. In addition, because unnecessary protocol stacks can be bypassed, resources are optimized overall, which is suitable for various cloud native scenarios that require high-performance storage solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 An application environment diagram of a cloud storage processing method in an embodiment;

[0029] Figure 2 A schematic diagram of a process flow of a cloud storage processing method in one embodiment;

[0030] Figure 3 A schematic diagram of the kernel services for extending the Berkeley packet filter;

[0031] Figure 4 A schematic diagram of a flow chart of a cloud storage processing method in another embodiment;

[0032] Figure 5 is a structural block diagram of a cloud storage processing device in one embodiment;

[0033] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0035] Introduction to some terms involved in this application:

[0036] K8s (Kubernetes): A distributed architecture solution and cluster management system based on container technology.

[0037] K8s Operator: A Kubernetes extension that helps users manage applications and services in a custom way using Kubernetes' declarative API (application programming interface) style.

[0038] CSI (Container Storage Interface): is a standard interface for providing persistent storage for Kubernetes clusters.

[0039] CRD (Custom Resource Definition): Custom resource definition is one of the Kubernetes extension mechanisms.

[0040] CR (Custom Resource): is a specific instance of a custom resource.

[0041] iSCSI (Internet Small Computer System Interface): Internet Small Computer System Interface is a storage protocol based on TCP / IP network.

[0042] SPDK (Storage Performance Development Kit): Storage performance development kit.

[0043] eBPF (extended Berkeley Packet Filter): Extended Berkeley Packet Filter, a program that runs on the Linux kernel to efficiently process network packets.

[0044] The cloud storage processing method provided in the embodiment of the present application can be applied to a server. The server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The server can communicate with the terminal through a network. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc.

[0045] The cloud storage processing method provided in the embodiment of the present application can be based on the following Figure 1 The application environment implementation shown can be used as a system architecture topology for implementing a cloud storage processing method.

[0046] Among them, the main metadata can include:

[0047] XBSDisk: Cloud disk description record, which can include some basic metadata of the cloud disk.

[0048] XBSDiskReplicas: Cloud disk replica information, which can describe which block storage engine XBSTarget nodes the replicas of this cloud disk are allocated to.

[0049] Among them, the main components may include:

[0050] XBSMangerAPI: Management and control request processing interface, which is responsible for processing management and control requests, including operations such as cloud disk creation, deletion, modification, query, expansion, cloning and backup. It can update cloud disk metadata to the cloud disk description record XBSDisk, a custom resource definition CRD.

[0051] XBSRedirector: Redirector that can provide input and output (IO) redirection function. When the input and output request of the cloud disk client fails, it can be redirected to the available block storage engine XBSTarget copy.

[0052] Keepalived: Keep alive, provide high availability services, and ensure the high availability of the management and control request processing interface XBSMangerAPI and the redirector XBSRedirector.

[0053] XBSReplica: Cloud disk replica management, responsible for the management of cloud disk replicas. It can select the optimal block storage engine XBSTarget node according to the load of the storage node, and create or modify the cloud disk replica information XBSDiskReplicas record.

[0054] XBStorAPI: Block storage engine interface, which can receive actual business requests from the management and control request processing interface XBSMangerAPI and convert them into remote procedure call RPC operations for the block storage engine XBSTarget.

[0055] XBSTarget: A block storage engine developed based on SPDK (High-Performance Storage Development Kit), which can receive remote procedure call (RPC) control requests from the block storage engine interface XBStorAPI, and provide input and output (IO) services of actual data volumes through the iSCSI (Internet Small Computer System Interface) protocol.

[0056] XBSNodeOptimizer: Node optimizer, which uses eBPF (Extended Berkeley Packet Filter) technology to optimize input and output (IO) requests directly at the kernel layer, bypassing the protocol stack and shortening the network input and output (IO) path.

[0057] XBSScheduler: A custom scheduling unit Pod scheduler that can prioritize the scheduling of application scheduling unit Pods to the nodes where the cloud disk copies are located, and notify the redirector XBSRedirector to adjust the input and output (IO) paths, and cooperate with the node optimizer XBSNodeOptimizer to greatly improve storage performance.

[0058] like Figure 1 As shown, the above components, including the control request processing interface XBSMangerAPI, redirector XBSRedirector, and keepalived, can be deployed in different containers of the scheduling unit pod named the control node XBSController, and can be hosted using the statefulset of k8s, and can be configured with two copies of primary and backup. The cloud disk copy management XBSReplica and the scheduling unit pod scheduler XBSScheduler can be hosted by an independent deployment (a resource object used to manage the deployment and update of applications), and a replica set can be set. The block storage engine interface XBStorAPI, the block storage engine XBSTarget, and the node optimizer XBSNodeOptimizer can be deployed in different containers of the scheduling unit pod named the data node XBSNode, which is pulled up on each data node.

[0059] The following is based on Figure 1 The system architecture shown is combined with various embodiments and corresponding drawings to illustrate the cloud storage processing method of the present application.

[0060] In an exemplary embodiment, Figure 2 As shown, a cloud storage processing method is provided, combined with Figure 1 , the method may include the following steps:

[0061] Step S201, determine the block storage engine node where the cloud disk server is located.

[0062] In this step, after the cloud disk server is deployed, the block storage engine XBSTarget node where the cloud disk server is located can be determined. The block storage engine XBSTarget node starts the node optimizer XBSNodeOptimizer.

[0063] Step S202: Bind the service scheduling unit to the block storage engine node, and send the address of the block storage engine node to the redirector.

[0064] In this step, the business scheduling unit pod can be bound to the above-mentioned block storage engine XBSTarget node, and the address of the bound block storage engine XBSTarget node is sent to the redirector XBSRedirector, thereby allowing the business scheduling unit pod to be bound and allocated to the same data node (as the cloud disk server).

[0065] Step S203, according to the login request of the cloud disk client mounted by the business scheduling unit, the redirector provides an address to the cloud disk client for login, so that the business scheduling unit can log in to the block storage engine XBSTarget node to perform input and output operations of the cloud disk business.

[0066] In this step, the cloud disk client mounted on the business scheduling unit pod can log in to the same block storage engine XBSTarget node (as the cloud disk server) through iscsi (Internet Small Computer System Interface) and make the final data input and output (IO) request. Among them, according to the login request of the cloud disk client mounted on the business scheduling unit pod, the address of the above-mentioned bound block storage engine XBSTarget node can be provided by XBSRedirector and returned to the cloud disk client for login, so that the business scheduling unit pod can log in to the block storage engine XBSTarget node to perform input and output (IO) operations of the cloud disk business. Figure 1 and Figure 3As shown, the node optimizer XBSNodeOptimize is started on the block storage engine XBSTarget node. The node optimizer XBSNodeOptimizer is used to start the kernel service of the extended Berkeley packet filter eBPF on the block storage engine XBSTarget node, and forward the input and output requests of the above-mentioned input and output operations at the kernel level. Specifically, the block storage engine XBSTarget input and output io requests from this node and sent to this node can be directly forwarded in the kernel, bypassing the protocol stack. Among them, the node optimizer XBSNodeOptimize uses the extended Berkeley packet filter eBPF program to directly forward the data packet at the socket level inside the kernel. The data packet that has passed the extended Berkeley packet filter eBPF redirect function (the main function is to redirect the data packet to a specific processing path during the data packet processing) can bypass the relevant devices and protocol stack of the local machine and be directly forwarded to the socket socket monitored by the destination process.

[0067] The cloud storage processing method of this embodiment determines the block storage engine node where the cloud disk server is located; the block storage engine node starts the node optimizer; the business scheduling unit is bound to the block storage engine node, and the address of the block storage engine node is sent to the redirector; according to the login request of the cloud disk client mounted by the business scheduling unit, the address is provided to the cloud disk client through the redirector for login, so that the business scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk business; wherein the node optimizer is used to start the kernel service of the extended Berkeley packet filter at the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level. This solution can schedule the business scheduling unit to the block storage engine node where the cloud disk server is located, and combine it with the kernel service of the extended Berkeley packet filter started by starting the node optimizer at the block storage engine node of the cloud storage system. The requests corresponding to the input and output operations of the cloud disk business from this node and sent to this node are forwarded directly at the kernel level, thereby bypassing the protocol stack to shorten the network input and output path, reduce the cloud disk access latency, and improve the cloud disk storage access performance. In addition, the kernel service of the extended Berkeley packet filter is integrated in the block storage engine node, which can directly optimize the node without installing a third-party network plug-in, reducing the deployment complexity and maintenance cost of the cloud storage system. In addition, because unnecessary protocol stacks can be bypassed, resources are optimized overall, which is suitable for various cloud native scenarios that require high-performance storage solutions.

[0068] In an exemplary embodiment, determining the block storage engine XBSTarget node where the cloud disk server is located in step S201 may include:

[0069] After the generated cloud disk description record is monitored by the cloud disk copy management service, the target block storage engine node is determined according to the load information of each node; a volume creation request is sent to the block storage engine interface of the target block storage engine node through the cloud disk copy management service; the block storage engine interface is used to call the remote procedure call interface provided by the target block storage engine node to execute volume creation and setting, and obtain the block storage engine node where the cloud disk server is located.

[0070] In this embodiment, the cloud disk copy management XBSReplica service can monitor whether the generated cloud disk description record XBSDisk is generated. After the cloud disk copy management XBSReplica service monitors the generated cloud disk description record XBSDisk, the target block storage engine XBSTarget node (as the storage backend) can be determined according to the load information of each node in the cluster, and one or more target block storage engine XBSTarget nodes can be determined. Then, the cloud disk copy management XBSReplica service can send a volume creation request to the block storage engine interface XBStorAPI of each target block storage engine XBSTarget node, and the block storage engine interface XBStorAPI calls the remote procedure call RPC interface provided by the target block storage engine XBSTarget node to execute volume creation and setting, thereby obtaining the block storage engine XBSTarget node where the cloud disk server is located.

[0071] In an exemplary embodiment, further, after the generated cloud disk description record XBSDisk is monitored by the cloud disk copy management XBSReplica service, before the target block storage engine XBSTarget node is determined according to the load information of each node, it may also include:

[0072] A disk creation request is initiated to the management and control request processing interface through the cluster's persistent storage standard interface; a cloud disk description record is generated according to the disk creation request through the management and control request processing interface.

[0073] In this embodiment, a disk creation request can be initiated to the management and control request processing interface XBSMangerAPI through the persistent storage standard interface CSI of the K8s cluster. After the management and control request processing interface XBSMangerAPI receives the disk creation request from the management and control request processing interface XBSMangerAPI, it can generate a cloud disk description record XBSDisk according to the disk creation request.

[0074] In an exemplary embodiment, further, the above-mentioned determining the target block storage engine node according to the load information of each node may include: determining a target number of nodes as the target block storage engine nodes according to the load information of each node; binding the business scheduling unit to the block storage engine node in step S202 may include: binding the business scheduling unit to one of the block storage engine nodes.

[0075] In this embodiment, two (target number) nodes can be selected as target block storage engine XBSTarget nodes according to the load information of each node in the K8s cluster. In the binding stage, the service scheduling unit pod can be bound to one of the block storage engine XBSTarget nodes.

[0076] In an exemplary embodiment, before determining the block storage engine node where the cloud disk server is located in step S201, the method of the present application may further include:

[0077] After the cluster is deployed, each block storage engine node of the cluster is triggered to start the node optimizer, so that the node optimizer starts the kernel service of the extended Berkeley packet filter.

[0078] In this embodiment, after the storage cluster is deployed, each block storage engine XBSTarget node of the cluster can be triggered to start the node optimizer XBSNodeOptimizer, and the node optimizer XBSNodeOptimizer will start the kernel service of the extended Berkeley packet filter eBPF on the respective block storage engine XBSTarget nodes.

[0079] In an exemplary embodiment, binding the service scheduling unit to the block storage engine node in step S202 may include:

[0080] When the cluster creates a business scheduling unit and mounts the persistent volume declaration corresponding to the cloud disk, the business scheduling unit is identified by the scheduling unit scheduler and bound to the block storage engine node where the cloud disk server is located.

[0081] In this embodiment, when the K8s cluster creates a business scheduling unit pod and mounts the persistent volume declaration pvc corresponding to the cloud disk, the corresponding business scheduling unit pod can be identified by the scheduling unit scheduler XBSScheduler, and then the business scheduling unit pod can be bound to the block storage engine XBSTarget node where the corresponding cloud disk server is located. If there are two block storage engine XBSTarget nodes, it can be bound to one of them.

[0082] Combination Figure 1 and Figure 4The overall process of the cloud storage processing method of this application is described as follows:

[0083] After the storage cluster is deployed, the system starts the node optimizer XBSNodeOptimizer on the block storage engine XBSTarget node. The service starts the kernel service of the extended Berkeley packet filter eBPF on the block storage engine XBSTarget node, and forwards the block storage engine XBSTarget input and output IO requests from and to this node directly in the kernel, bypassing the protocol stack.

[0084] The K8s cluster initiates a disk creation operation through the persistent storage standard interface CSI. After the management request processing interface XBSMangerAPI receives the request from the persistent storage standard interface CSI, it generates a cloud disk description record XBSDisk.

[0085] After the cloud disk copy management XBSReplica service monitors the newly generated cloud disk description record XBSDisk, it selects two optimal block storage engine XBSTarget nodes as the storage backend according to the storage node load, and sends the volume creation request to the block storage engine interface XBStorAPI of the corresponding node.

[0086] The block storage engine interface XBStorAPI calls the remote procedure call RPC interface provided by the block storage engine XBSTarget to complete the creation and setting of the volume.

[0087] When the K8s cluster creates a business scheduling unit pod and mounts the persistent volume declaration pvc corresponding to the cloud disk, the scheduling unit scheduler XBSScheduler identifies the corresponding business scheduling unit pod, binds the business scheduling unit pod to one of the nodes where the block storage engine XBSTarget of the corresponding cloud disk backend is located, and sends the block storage engine XBSTarget address corresponding to the bound node to the redirector XBSRedirector, so that the business scheduling unit pod is bound and assigned to the same data node;

[0088] The cloud disk client mounted on the business scheduling unit pod logs in to the block storage engine XBSTarget on the same node through the Internet Small Computer System Interface (iscsi) and makes the final data input and output io request.

[0089] Since the node optimizer XBSNodeOptimizer optimizes message forwarding at the kernel level by extending the Berkeley Packet Filter eBPF, shortening the cloud disk input and output IO path, the optimization goal is achieved in the entire process.

[0090] In this embodiment, in each node where the storage engine is located, linux eBPF can be used to forward the cloud disk network input and output io requests from this node and sent to this node directly at the kernel level, bypassing the protocol stack to shorten the network input and output io; secondly, based on the K8s extended scheduler capability, a custom scheduler can be implemented, and the scheduling unit pod can be scheduled to the node where the cloud disk server is located as much as possible through a custom strategy, combined with the extended Berkeley packet filter eBPF technology optimization, so as to achieve the improvement of cloud disk input and output io performance. At the same time, because unnecessary protocol stacks are bypassed, the overall resource utilization is also optimized, which is suitable for various cloud native scenarios that require high-performance storage solutions. Among them, the storage system component node optimizer XBSNodeOptimizer integrates the kernel network optimization services related to the extended Berkeley packet filter eBPF, which can directly optimize the node without installing a third-party network plug-in, reducing the system deployment complexity and maintenance costs. Among them, the application is scheduled to the node where the storage volume copy is located by the custom scheduler first. The storage system can schedule the business scheduling unit pod according to the location of the cloud disk server node copy. Combined with the node optimization service, it can shorten the network input and output I / O path of the cloud disk, reduce the cloud disk access delay, and improve storage access performance. Among them, the optimized input and output I / O path is opened through the redirection service, and the scheduler is linked with the redirection service. The scheduler writes the address of the target block storage engine XBSTarget node that can achieve input and output I / O path optimization to the redirection service. When the persistent storage standard interface CSI actually mounts the cloud disk, it obtains the target address from the redirection service to mount the cloud disk copy of this node.

[0091] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0092] Based on the same inventive concept, the embodiment of the present application also provides a cloud storage processing device for implementing the cloud storage processing method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more cloud storage processing device embodiments provided below can refer to the limitations of the cloud storage processing method above, and will not be repeated here.

[0093] In an exemplary embodiment, Figure 5 As shown, a cloud storage processing device is provided, and the device 500 may include:

[0094] The node determination module 501 is used to determine the block storage engine node where the cloud disk server is located; wherein the block storage engine node starts the node optimizer;

[0095] A scheduling binding module 502, used to bind the service scheduling unit to the block storage engine node, and send the address of the block storage engine node to the redirector;

[0096] A login processing module 503 is used to provide the address to the cloud disk client for login through the redirector according to the login request of the cloud disk client mounted by the business scheduling unit, so that the business scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk business;

[0097] The node optimizer is used to start the kernel service of the extended Berkeley packet filter in the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

[0098] In an exemplary embodiment, the node determination module 501 is used to determine the target block storage engine node according to the load information of each node after monitoring the generated cloud disk description record through the cloud disk copy management service; send a volume creation request to the block storage engine interface of the target block storage engine node through the cloud disk copy management service; the block storage engine interface is used to call the remote procedure call interface provided by the target block storage engine node to execute volume creation and setting, and obtain the block storage engine node where the cloud disk server is located.

[0099] In an exemplary embodiment, the node determination module 501 is also used to initiate a disk creation request to the management and control request processing interface through the cluster's persistent storage standard interface; and generate the cloud disk description record according to the disk creation request through the management and control request processing interface.

[0100] In an exemplary embodiment, the node determination module 501 is used to determine a target number of nodes as the target block storage engine nodes according to the load information of each node; the scheduling binding module 502 is used to bind the business scheduling unit to one of the block storage engine nodes.

[0101] In an exemplary embodiment, the device 500 may also include: a service triggering module, which is used to trigger each block storage engine node of the cluster to start the node optimizer after the cluster deployment is completed, so that the node optimizer starts the kernel service of the extended Berkeley packet filter.

[0102] In an exemplary embodiment, the scheduling binding module 502 is used to identify the business scheduling unit through the scheduling unit scheduler and bind the business scheduling unit to the block storage engine node where the cloud disk server is located when the cluster creates the business scheduling unit and mounts the persistent volume declaration corresponding to the cloud disk.

[0103] Each module in the above cloud storage processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0104] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device can be used to store data such as input and output requests. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a cloud storage processing method is implemented.

[0105] Those skilled in the art will understand that Figure 6The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0106] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0107] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0108] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0109] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0110] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0111] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A cloud storage processing method, characterized in that: The method comprises: Determine the block storage engine node where the cloud disk server is located; wherein the block storage engine node starts a node optimizer; Binding a service scheduling unit to the block storage engine node, and sending the address of the block storage engine node to a redirector; According to a login request of the cloud disk client mounted by the service scheduling unit, the address is provided to the cloud disk client through the redirector for login, so that the service scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk service; The node optimizer is used to start the kernel service of the extended Berkeley packet filter in the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

2. The method according to claim 1, characterized in that Determining the block storage engine node where the cloud disk server is located includes: After the generated cloud disk description record is monitored by the cloud disk copy management service, the target block storage engine node is determined according to the load information of each node; A volume creation request is sent to the block storage engine interface of the target block storage engine node through the cloud disk copy management service; the block storage engine interface is used to call the remote procedure call interface provided by the target block storage engine node to execute volume creation and setting, and obtain the block storage engine node where the cloud disk server is located.

3. The method according to claim 2, characterized in that After the generated cloud disk description record is monitored by the cloud disk copy management service, before the target block storage engine node is determined according to the load information of each node, the method further includes: Initiate a disk creation request to the management request processing interface through the cluster's persistent storage standard interface; The cloud disk description record is generated according to the disk creation request through the management and control request processing interface.

4. The method according to claim 2, characterized in that: The step of determining the target block storage engine node according to the load information of each node includes: According to the load information of each node, a target number of nodes are determined as the target block storage engine nodes; The step of binding the service scheduling unit to the block storage engine node includes: Bind the service scheduling unit to one of the block storage engine nodes.

5. The method according to claim 1, characterized in that: Before determining the block storage engine node where the cloud disk server is located, the method further includes: After the cluster deployment is completed, each block storage engine node of the cluster is triggered to start the node optimizer, so that the node optimizer starts the kernel service of the extended Berkeley packet filter.

6. The method according to any one of claims 1 to 5, characterized in that The step of binding the service scheduling unit to the block storage engine node includes: When the cluster creates the business scheduling unit and mounts the persistent volume declaration corresponding to the cloud disk, the business scheduling unit is identified by the scheduling unit scheduler, and the business scheduling unit is bound to the block storage engine node where the cloud disk server is located.

7. A cloud storage processing device, characterized in that: The device comprises: A node determination module is used to determine the block storage engine node where the cloud disk server is located; wherein the block storage engine node starts a node optimizer; A scheduling binding module, used to bind the service scheduling unit to the block storage engine node, and send the address of the block storage engine node to the redirector; A login processing module, configured to provide the address to the cloud disk client for login through the redirector according to a login request of the cloud disk client mounted by the service scheduling unit, so that the service scheduling unit can log in to the block storage engine node to perform input and output operations of the cloud disk service; The node optimizer is used to start the kernel service of the extended Berkeley packet filter in the block storage engine node, and forward the input and output requests of the input and output operations at the kernel level.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images