Cloud host storage unloading system

Abstract

The invention discloses a cloud host storage unloading system. The cloud host storage unloading system comprises a management center, an OpenStack and a TIPU, the cloud host storage unloading system is carried out according to the following steps: S1, installing and deploying a TIPU internet gateway service and a driver; s2, creating a Manila shared file; s3, encrypting the information of the Mania shared file; s4, calling a TIPU RPC interface to create virtio equipment in the creation process of the cloud host; s5, rendering the cloud host XML in the cloud host creation process; s6, configuring a startup command and injecting secret key data in a cloud host creation process; and S7, completing the creation of the cloud host. The system can ensure that the TIPU senses and directly accesses a specified NFS file directory in real time, and realizes efficient and dynamic unloading of the storage load of the cloud host to the TIPU through seamless interaction with the OpenStack platform, so that the response speed of storage is remarkably improved, the CPU load of the host is reduced, and the resource utilization rate of the whole cloud environment is optimized.

Description

Technical Field

[0001] The present application belongs to the technical field of cloud computing and storage virtualization, and in particular, relates to a cloud host storage offloading system. Background Art

[0002] With the rapid development of cloud computing technology, cloud hosts have become an important part of enterprise information construction. Cloud hosts use virtualization technology to encapsulate computing resources, storage resources, and network resources into an independent virtual environment, providing users with on-demand and elastically scalable computing resources. However, with the continuous growth of business volume and the sharp increase in data volume, traditional cloud host storage methods have gradually exposed problems such as performance bottlenecks and low resource utilization.

[0003] High CPU resource usage: In traditional cloud host architecture, storage operations such as data reading and writing, cache management, etc. are usually handled directly by the CPU. This processing method not only increases the CPU load, but may also cause performance bottlenecks when the CPU is processing other computing-intensive tasks.

[0004] Limited storage performance: Because the CPU needs to handle computing and storage tasks at the same time, storage performance is difficult to fully utilize. Especially when processing a large number of concurrent read and write requests, the storage system's response time and throughput often cannot meet business needs.

[0005] Low resource utilization: Traditional cloud host storage methods usually tightly bind computing and storage resources together, and cannot be flexibly allocated according to actual needs. This leads to low resource utilization and high costs.

[0006] In order to solve the above problems, the industry has proposed storage offloading, which can offload storage-related tasks originally handled by the CPU to the DPU, thereby significantly reducing the CPU load and enabling it to focus more on performing other compute-intensive tasks. Storage offloading can optimize data access paths and reduce delays in data transmission and processing, thereby accelerating the reading and writing speed of data and improving the overall performance of the storage system. Storage offloading can also enhance data security. By building security mechanisms into the DPU, such as data encryption and access control, additional security can be provided during data storage and transmission. The TIPU mentioned in the present invention is a special DPU that supports trusted computing technology.

[0007] However, how to offload the storage of cloud hosts to TIPU flexibly, simply and securely has always been a problem that has plagued the industry. Summary of the invention

[0008] The purpose of this application is to address the problems of low efficiency in storage load management and difficulty in achieving efficient offloading to TIPU during the cloud host creation process in the prior art. This application discloses a cloud host storage offloading system. This system can ensure that TIPU can perceive and directly access the specified NFS file directory in real time, and through seamless interaction with the OpenStack platform, it can achieve efficient and dynamic offloading of cloud host storage load to TIPU, thereby significantly improving the storage response speed, reducing the host CPU load, and optimizing the resource utilization of the overall cloud environment.

[0009] The purpose of this application is achieved through the following technical solutions:

[0010] A cloud host storage offloading system, the cloud host storage offloading system includes a management center, OpenStack and TIPU; the cloud host storage offloading system is performed according to the following steps:

[0011] S1: Gateway service and driver installation and deployment on TIPU;

[0012] S2: Creation of Manila shared files;

[0013] S3: Encrypts Manila shared file information;

[0014] S4: During the cloud host creation process, the TIPU RPC interface is called to create a virtio device.

[0015] S5: Rendering cloud host XML during cloud host creation;

[0016] S6: Configure the boot command and inject key data during the cloud host creation process;

[0017] S7: The cloud host is created.

[0018] According to a preferred implementation, step S1 includes: deploying a gateway service on a TIPU card and deploying a corresponding driver on a computing node of OpenStack.

[0019] According to a preferred embodiment, step S2 includes: the management center uses the API interface of the OpenStack Manila service to create a copy of the private data and shares it with the TIPU on the designated computing node.

[0020] According to a preferred implementation, step S3 includes: the management center encrypts the file information shared by Manila by calling the RPC method of TIPU.

[0021] According to a preferred embodiment, step S4 includes: the management center sets the encrypted shared file information and storage information as metadata of the virtual machine;

[0022] After encapsulating the corresponding metadata parameters, call the OpenStack API interface to create a virtual machine;

[0023] When receiving a virtual machine creation request, the driver on the computing node calls the RPC interface of TIPU and creates the corresponding virtio-fs device on the TIPU card.

[0024] According to a preferred embodiment, step S5 includes: the computing node driver renders an XML configuration file of the virtio-fs device according to the return information of the RPC interface, wherein the XML configuration file includes the hardware and software configuration of the virtual machine.

[0025] According to a preferred embodiment, step S6 includes: using the cloud-init tool, the computing node driver configures the first startup execution command of the virtual machine, the command to be executed each time the machine is turned on, and the key data required to decrypt the private data.

[0026] According to a preferred implementation, step S7 includes: the computing node driver calls the Libvirt interface to start the virtio-fs device, and virtio-fs uses the shared files provided by the Manila service, thereby achieving secure access to and use of private data.

[0027] The aforementioned main scheme of the present application and its further options can be freely combined to form multiple schemes, all of which are schemes that can be adopted and claimed for protection in the present application. After understanding the scheme of the present application, those skilled in the art can understand that there are multiple combinations based on the prior art and common knowledge, all of which are technical schemes to be protected by the present application, and they are not exhaustively listed here.

[0028] Beneficial effects of this application:

[0029] 1. Significantly improve storage response speed: Directly sense and access the NFS file directory through TIPU, reducing the data transmission path between the CPU and the storage device, greatly shortening the storage access delay, and thus significantly improving the response speed of the storage system.

[0030] 2. Reduce the host CPU load: Offloading the storage load originally borne by the CPU to the TIPU for processing effectively reduces the workload of the host CPU, allowing the CPU to focus more on performing other key tasks and improve the performance and efficiency of the overall system.

[0031] 3. Optimize resource utilization: The intelligent scheduling and unloading mechanism of the present invention can dynamically allocate storage resources according to the actual needs of the cloud environment, avoid idleness and waste of resources, and significantly improve the resource utilization of the overall cloud environment.

[0032] 4. Enhanced storage security: By encrypting key storage configuration information and securely transmitting it in the form of metadata during the cloud host creation process, the confidentiality and integrity of data during transmission and storage are ensured, thereby enhancing the security of the storage system.

[0033] 5. Improve operational convenience: The automated configuration and Cloud-init mechanism of the present invention simplifies the process of cloud host creation and storage configuration, reduces manual intervention, reduces operational complexity, and improves the work efficiency of operation and maintenance personnel.

[0034] 6. Promote the sustainable development of cloud computing environment: By improving energy efficiency, reducing energy consumption and reducing resource waste, the present invention helps to promote the green and sustainable development of cloud computing environment, which is in line with the current global general requirements for environmental protection and energy conservation and emission reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of the structure of the cloud host storage offloading system of this application;

[0036] Figure 2 This is a schematic diagram of the workflow of the cloud host storage offloading system of this application. DETAILED DESCRIPTION

[0037] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0038] It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. In addition, the present application should point out that in the present application, if the specific structure, connection relationship, position relationship, power source relationship, etc. are not specifically written, the structure, connection relationship, position relationship, power source relationship, etc. involved in the present application are all known by those skilled in the art based on the existing technology without creative work.

[0039] refer to Figure 1 and Figure 2 As shown, the present application discloses a cloud host storage offloading system, which includes a management center, OpenStack and TIPU.

[0040] The cloud host storage offloading system of this application implements cloud host storage offloading according to the following steps.

[0041] Step S1: Gateway service and driver installation and deployment on TIPU.

[0042] Preferably, step S1 includes: deploying a gateway service on the TIPU card and deploying a corresponding driver on the computing node of OpenStack, and ensuring that all relevant parameters are correctly configured to ensure smooth communication between the service and the driver.

[0043] Step S2: Creation of Manila shared files.

[0044] Preferably, step S2 includes: the management center uses the API interface of the OpenStack Manila service to create a copy of the private data and shares it with the TIPU on the designated computing node.

[0045] Step S3: Encrypt Manila shared file information.

[0046] Preferably, step S3 includes: the management center encrypts the file information shared by Manila by calling the RPC method of TIPU, thereby enhancing the security of data and preventing unauthorized access.

[0047] Step S4: During the cloud host creation process, the TIPU RPC interface is called to create a virtio device.

[0048] Preferably, step S4 includes: the management center sets the encrypted shared file information and storage information as the metadata of the virtual machine; after encapsulating the corresponding metadata parameters, calling the OpenStack API interface to create a virtual machine; when receiving the virtual machine creation request, the driver on the computing node calls the RPC interface of TIPU to create a corresponding virtio-fs device on the TIPU card.

[0049] In this application, the encrypted storage configuration information is encapsulated as metadata and parsed and used during the cloud host creation process. This method ensures the security and flexibility of the storage configuration information and provides a basis for subsequent storage resource scheduling and unloading.

[0050] Furthermore, by calling the RPC interface of the TIPU gateway, the virtio device is dynamically created and seamlessly integrated with the cloud host. This process realizes the flexible configuration and rapid response of storage resources, meeting the demand for high-performance storage in the cloud computing environment.

[0051] Step S5: Rendering the cloud host XML during the cloud host creation process.

[0052] Preferably, step S5 includes: the computing node driver renders an XML configuration file of the virtio-fs device according to the return information of the RPC interface, wherein the XML configuration file defines the hardware and software configuration of the virtual machine to ensure its correct operation.

[0053] Specifically, Nova-compute Driver uses the virtio device information obtained from the TIPU gateway to automatically update and render the cloud host's XML configuration file to ensure that the cloud host can recognize and load these newly added storage devices when it starts. In addition, the virtofs device must be explicitly mounted on the compute node to provide the cloud host with a direct access channel to the shared storage.

[0054] Step S6: Configure the boot command and inject key data during the cloud host creation process.

[0055] Preferably, step S6 includes: using the cloud-init tool, the computing node driver configures the first startup execution command of the virtual machine, the command to be executed each time the machine is turned on, and the key data required to decrypt the private data.

[0056] Specifically, in order to ensure that the cloud host can use the newly added virtio devices smoothly, this application introduces the Cloud-init mechanism. Before the cloud host starts, Cloud-init automatically injects the necessary boot commands and key data into the cloud host. These commands and data will guide the cloud host on how to correctly configure and use virtio devices, thereby achieving a seamless storage offloading and access experience. The introduction of the Cloud-init mechanism simplifies the configuration process of the cloud host and improves the automation of storage offloading and access.

[0057] Step S7: The cloud host is created.

[0058] Preferably, step S7 includes: the computing node driver calls the Libvirt interface to start the virtio-fs device, and the virtio-fs uses the shared files provided by the Manila service, thereby achieving secure access to and use of the private data.

[0059] This system can ensure that TIPU can perceive and directly access the specified NFS file directory in real time. Through seamless interaction with the OpenStack platform, it can realize efficient and dynamic offloading of cloud host storage load to TIPU, thereby significantly improving the storage response speed, reducing the host CPU load, and optimizing the resource utilization of the overall cloud environment.

[0060] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A cloud host storage offloading system, characterized in that: The cloud host storage offloading system includes a management center, OpenStack and TIPU; the cloud host storage offloading system is performed according to the following steps: S1: Gateway service and driver installation and deployment on TIPU; S2: Creation of Manila shared files; S3: Encrypts Manila shared file information; S4: During the cloud host creation process, the TIPU RPC interface is called to create a virtio device; S5: Rendering cloud host XML during cloud host creation; S6: Configure the boot command and inject key data during the cloud host creation process; S7: The cloud host is created.

2. The cloud host storage offloading system according to claim 1, characterized in that: Step S1 includes: deploying a gateway service on the TIPU card and deploying a corresponding driver on the computing node of OpenStack.

3. The cloud host storage offloading system according to claim 1 or 2, characterized in that: Step S2 includes: the management center uses the API interface of the OpenStack Manila service to create a copy of the private data and shares it with the TIPU on the designated computing node.

4. The cloud host storage offloading system according to claim 3, characterized in that: Step S3 includes: the management center encrypts the file information shared by Manila by calling the RPC method of TIPU.

5. The cloud host storage offloading system according to claim 4, characterized in that: Step S4 includes: the management center sets the encrypted shared file information and storage information as metadata of the virtual machine; After encapsulating the corresponding metadata parameters, call the OpenStack API interface to create a virtual machine; When receiving a virtual machine creation request, the driver on the computing node calls the RPC interface of TIPU and creates the corresponding virtio-fs device on the TIPU card.

6. The cloud host storage offloading system according to claim 5, characterized in that: Step S5 includes: the computing node driver renders an XML configuration file of the virtio-fs device according to the return information of the RPC interface, wherein the XML configuration file includes the hardware and software configuration of the virtual machine.

7. The cloud host storage offloading system according to claim 6, characterized in that: Step S6 includes: using the cloud-init tool, the computing node driver configures the first startup execution command of the virtual machine, the command to be executed each time the machine is turned on, and the key data required to decrypt the private data.

8. The cloud host storage offloading system according to claim 7, characterized in that: Step S7 includes: the computing node driver calls the Libvirt interface to start the virtio-fs device, Virtio-fs uses the shared files provided by the Manila service to achieve secure access and use of private data.

Images