Cloud data center management systems and electronic devices

Abstract

The present application relates to a cloud data center management system and an electronic device. The management system includes a first management module and a virtual machine monitor, the first management module and the virtual machine monitor being located on a DPU, the first management module being configured to send a first control command to the virtual machine monitor to instruct the virtual machine monitor to perform a first operation on at least one virtual machine, the virtual machine monitor being configured to manage any different types of virtualization technologies as virtual machines, and to complete the first operation by calling a callback function of one specific type of device corresponding to the virtual machine when it receives and executes the first control command.

Description

Cross - reference to related applications

[0001] This application claims the priority of a Chinese patent application with the application number 202211328675.X, filed with the China National Intellectual Property Administration on October 27, 2022, and all of its content is incorporated herein by reference.

Technical Field

[0002] This application relates to the field of cloud data center technology, particularly to cloud data center management systems and electronic devices.

Background Art

[0003] Cloud data centers need to be operated and maintained. However, due to their large scale, the high operation efficiency directly affects the operation cost. In multiple cloud technology application scenarios of cloud data centers, the bare - metal application scenario, virtualization application scenario, and container application scenario are still in a separated state, and it is necessary to develop and maintain one management solution for each, resulting in low flexibility and high complexity in the operation and maintenance of cloud data centers.

Summary of the Invention

Problems to be Solved by the Invention

[0004] Cloud data centers need to be operated and maintained. However, due to their large scale, the high operation efficiency directly affects the operation cost. In multiple cloud technology application scenarios of cloud data centers, the bare - metal application scenario, virtualization application scenario, and container application scenario are still in a separated state, and it is necessary to develop and maintain one management solution for each, resulting in low flexibility and high complexity in the operation and maintenance of cloud data centers.

[0005] This application provides a cloud data center management system and electronic devices that can solve the technical problem of low flexibility and high complexity in the operation and maintenance of cloud data centers, which arises because it is necessary to develop and maintain a separate management system for each of the bare-metal application, virtualization application, and container application scenarios. [Means for solving the problem]

[0006] In a first embodiment, the application provides a cloud data center management system including a first management module and a virtual machine monitor. The first management module and virtual machine monitor are located on the DPU (Data Processing Unit). The first management module is configured to send a first control command to the virtual machine monitor to instruct the virtual machine monitor to perform a first operation on at least one virtual machine. The virtual machine monitor manages different types of virtualization technologies as virtual machines, and when it receives and executes the first control command, it is configured to complete the first operation by calling a callback function of one specific type of device corresponding to the virtual machine.

[0007] In a second aspect, the present invention further provides an electronic device including a host, a DPU, and the cloud data center management system described above. The DPU is located in or inserted into the host, or the host is located in or inserted into the DPU. [Brief explanation of the drawing]

[0008] [Figure 1] This is a schematic diagram of a cloud data center management system according to one embodiment. [Figure 2] This is a schematic diagram of a cloud data center management system according to another embodiment. [Figure 3] This is a schematic diagram of an electronic device according to one embodiment. [Modes for carrying out the invention]

[0009] The present application will be described below with reference to the drawings.

[0010] Figure 1 is a schematic diagram of a cloud data center management system according to one embodiment of the present application. As shown in Figure 1, in the embodiment of the present application, the cloud data center management system 1 includes a first management module 10 and a virtual machine monitor 12. The first management module 10 and the virtual machine monitor 12 are located on a DPU. The DPU may be a system-on-a-chip (SoC) based chip or a smart network interface card based on an FPGA (Field Programmable Gate Array).

[0011] The first management module 10 is configured to send a first control command to the virtual machine monitor 12. The first control command instructs the virtual machine monitor 12 to perform a first operation on at least one virtual machine.

[0012] The virtual machine monitor 12 manages different types of virtualization technologies as virtual machines, and when it receives and executes the first control command, it is configured to complete the first operation by calling a callback function of one specific type of device corresponding to the virtual machine.

[0013] The first operation includes creating or deleting at least the virtual machine's backend device 14. The virtual machine's backend device 14 is located on the DPU side. The backend device 14 is a PCIe (Peripheral Component Interconnect Express) device within the virtual machine.

[0014] The aforementioned different types of virtualization technologies include at least bare metal (BM) technology, virtual machine (VM) technology, and container (Container) technology.

[0015] One specific type of device corresponding to the aforementioned virtual machine is the virtual machine's backend device 14.

[0016] The callback function for a specific type of device corresponding to the virtual machine may be a processing function for that specific type of device. For example, if the specific type of device is an NVMe (Non-Volatile Memory Express) device mounted on the virtual machine, the callback function may be an implementation function (nvme_device_realize) for the NVMe device mounted on the virtual machine.

[0017] In the embodiments of this application, the first management module 10 is configured to receive operation and maintenance commands and, upon receiving them, to transmit a first control command to the virtual machine monitor 12, and to receive further operation and maintenance commands. The operation and maintenance commands may be commands transmitted directly or indirectly from the operators and maintainers of the data center.

[0018] In the embodiments of this application, the first management module 10 and the virtual machine monitor 12 can communicate using the QEMU Machine Protocol (QMP).

[0019] The cloud data center management system according to the embodiments of the present application uniformly manages (e.g., schedules) different types of virtualization technologies as virtual machines through a virtual machine monitor. For example, it schedules any of the bare metal technology, virtual machine technology, or container technology as one virtual machine. That is, by scheduling the overall operation and maintenance scenarios of the bare metal technology, virtual machine technology, and container technology up to the DPU, it is possible to avoid developing and maintaining one management solution for each different type of virtualization technology, improve the flexibility of the operation and maintenance of the cloud data center, reduce the complexity of the operation and maintenance of the data center, and improve the efficiency of operation and maintenance.

[0020] FIG. 2 is a schematic configuration diagram of a cloud data center management system according to another embodiment of the present application. As shown in FIG. 2, the embodiments of the present application may include the content of the embodiment shown in FIG. 1 of the present application. Based on the cloud data center management system according to the embodiment shown in FIG. 1 of the present application, the cloud data center management system 1 according to the embodiment shown in FIG. 2 of the present application may further include a second management module 16. The second management module 16 is arranged on a host, and the host may be a server or a server cluster.

[0021] When the first management module 10 further receives feedback regarding the completion of the execution of the first control command, it is configured to send a second control command to the second management module 16. The second control command is for instructing the second management module 16 to execute a second operation on the virtual machine.

[0022] The second management module 16 is configured to receive and execute the second control command so as to execute the second operation.

[0023] The second operation includes at least creating or deleting the front - end driver 18 of the virtual machine. The front - end driver 18 of the virtual machine is located on the host side. The front - end driver 18 is a drive program corresponding to the back - end device 14 in the virtual machine. The main role of the front - end driver 18 is to discover the back - end device 14, accept requests from inside the virtual machine, and communicate with the back - end device 14 according to the virtio protocol. In the case of a Linux system, common front - end drivers may include virtio - net, virtio - blk, and virtio - scsi, etc.

[0024] In the embodiment of the present application, the second management module 16 and the first management module 10 can realize communication through a Socket connection based on TCP (Transmission Control Protocol).

[0025] The cloud data center management system according to the embodiment of the present application instructs the virtual machine monitor to uniformly manage (for example, schedule) any different types of virtualization technologies as virtual machines by the first management module located on the DPU side. For example, any of the bare - metal technology, virtual machine technology, or container technology is scheduled as one virtual machine to realize the creation or deletion of the back - end device corresponding to the virtual machine. Then, the first management module instructs the second management module located on the host side to complete the creation or deletion of the front - end driver, so as to schedule the operation and maintenance scenarios of the bare - metal technology, virtual machine technology, and container technology to the DPU, avoid developing and maintaining one management solution for each different type of virtualization technology, improve the flexibility of the operation and maintenance of the cloud data center, reduce the complexity of the operation and maintenance of the data center, and improve the efficiency of operation and maintenance.

[0026] Embodiments of the present application can further provide an electronic device. As shown in Figure 3, the electronic device 100 may include a host 11, a DPU 12, and the cloud data center management system 1 located within the host 11 and the DPU 12. The DPU 12 is located within or inserted into the host 11, or the host 11 is located within or inserted into the DPU 12.

[0027] Details of the electronic devices provided in this application are not described in detail here, as they can be obtained by referring to the contents of the cloud data center management system provided in the above embodiment.

[0028] The electronic device relating to this application, by deploying the cloud data center management system, allows for the unified management (e.g., scheduling) of different types of virtualization technologies as virtual machines via a virtual machine monitor. For example, bare metal technology, virtual machine technology, or container technology can all be scheduled as a single virtual machine. That is, by scheduling the integrated operation and maintenance of bare metal technology, virtual machine technology, and container technology up to the DPU, it is possible to avoid developing and maintaining a separate management plan for each type of virtualization technology. This improves the flexibility of cloud data center operation and maintenance, reduces the complexity of data center operation and maintenance, and improves the efficiency of operation and maintenance.

Claims

1. A cloud data center management system including a first management module and a virtual machine monitor, The first management module and the virtual machine monitor are located on the data processing unit (DPU). The first management module is configured to send a first control command to the virtual machine monitor to instruct the virtual machine monitor to perform a first operation on at least one virtual machine. A cloud data center management system characterized in that the virtual machine monitor manages different types of virtualization technologies as virtual machines, and when it receives and executes the first control command, it is configured to complete the first operation by calling a callback function of one specific type of device corresponding to the virtual machine.

2. It further includes a second management module deployed on the host, The first management module is further configured to send a second control command to the second management module when it receives feedback that the first control command has been completed, the second control command instructing the second management module to perform a second operation on the virtual machine. The management system according to claim 1, wherein the second management module is configured to receive and execute the second control command to perform the second operation.

3. The management system according to claim 1, wherein the first operation includes creating or deleting a backend device for the virtual machine, the backend device for the virtual machine being located on the DPU side and being a certain type of device corresponding to the virtual machine.

4. The management system according to claim 2, wherein the second operation includes creating or deleting a front-end driver for the virtual machine, and the front-end driver for the virtual machine is located on the host side.

5. The management system according to claim 1, characterized in that the different types of virtualization technologies include bare metal technology, virtual machine technology, and container technology.

6. The management system according to claim 1, wherein the first management module is further configured to receive operation and maintenance commands, and transmits the first control command to the virtual machine monitor upon receiving operation and maintenance commands.

7. The management system according to claim 1, characterized in that the first management module and the virtual machine monitor can communicate using the QEMU machine protocol QMP.

8. The management system according to claim 1, characterized in that the second management module and the first management module can communicate with each other by a socket connection based on the TCP transmission control protocol.

9. The management system according to claim 2, characterized in that the DPU is located on the host.

10. It is an electronic device, The system includes a host, a data processing unit (DPU), and a cloud data center management system according to any one of claims 1-9. An electronic device characterized in that the DPU is located in or inserted into the host, or the host is located in or inserted into the DPU.

Images