CN112003895B - Cloud host evacuation method, device, device and storage medium in OpenStack cloud platform - Google Patents

Abstract

The present application relates to a cloud host evacuation method, device, device and storage medium in an OpenStack cloud platform. The method includes: using the Nova controller service to receive the evacuation request, and determining the cloud host to be evacuated according to the evacuation request; using the Nova controller service and the device management service to obtain device binding information of the cloud host to be evacuated, wherein the device binding information includes PCI devices and / or PCIe device; use the Nova controller service to screen out the cloud host to be evacuated from the cloud hosts to be evacuated according to the device binding information; use the Nova controller service, the cloud host of the bound device and the resource management service to determine whether there is a cloud platform Available computing nodes; if there are available computing nodes, use the Nova controller service, device management service, and Nova compute service to create a new cloud host on the available computing node for the cloud host bound to the device. The solution of the invention realizes the automatic evacuation of the cloud host bound to the device to other computing nodes, improves the robustness of the cloud platform, and saves labor costs.

Description

Cloud host evacuation method, device, device and storage medium in OpenStack cloud platform

technical field

The invention belongs to the field of cloud computing, and in particular relates to a cloud host evacuation method, device, equipment and storage medium in an OpenStack cloud platform.

Background technique

The multi-architecture cloud platform developed by OpenStack includes multiple computing nodes. Each computing node consists of a host and several cloud hosts (virtual machines) running on the host. When the host where the cloud host is located has a hardware failure or power failure When the entire computing node cannot work, the cloud host on the node needs to be evacuated to other available computing nodes.

At present, the Nova evacuate service provided by the OpenStack cloud platform can evacuate cloud hosts that are not bound to host devices, but cannot evacuate cloud hosts with pass-through PCI/PCIe devices. Operators need to manually evacuate and manually clean up the database. Binding relationship data between cloud hosts and PCI/PCIe devices; manual evacuation not only takes a long time, but also has high labor costs.

SUMMARY OF THE INVENTION

In view of this, it is necessary to provide a cloud host evacuation method, apparatus, device and storage medium in the OpenStack cloud platform for the above technical problems.

According to an aspect of the present invention, a cloud host evacuation method in an OpenStack cloud platform is provided, the method comprising the following steps:

Use the Nova controller service to receive the evacuation request, and determine the cloud host to be evacuated according to the evacuation request;

Obtain the device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service, wherein the device binding information includes PCI devices and/or PCIe devices;

Use the Nova controller service to filter out the cloud host to which the device is bound from the cloud hosts to be evacuated according to the device binding information;

Use the Nova controller service, the cloud host and resource management service of the bound device to determine whether there are available computing nodes in the cloud platform;

If there is an available computing node, use the Nova controller service, the device management service, and the Nova compute service to create a new cloud host on the available computing node for the cloud host bound to the device.

In one embodiment, the method further includes:

Use the Nova evacuate service to evacuate the remaining cloud hosts to be evacuated.

In one embodiment, the step of obtaining the device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service includes:

The Nova controller service invokes the device management service to obtain the device binding information of the cloud host to be evacuated;

The device management service returns the device binding information to the Nova controller service.

In one of the embodiments, the step of determining whether there is an available computing node in the cloud platform by using the Nova controller service, the cloud host bound to the device, and the resource management service includes:

The Nova controller service obtains the configuration information of the cloud host to which the device is bound, wherein the configuration information includes CPU information, memory information, and disk information;

Determine the resources required by the new cloud host according to the configuration information and the device binding information of the cloud host to which the device is bound;

The Nova controller service calls the resource management service to query whether there are available computing nodes in the cloud platform that meet the resources required by the new cloud host.

In one embodiment, the step of using the Nova controller service, the device management service and the Nova compute service to create a new cloud host on the available computing node for the device-bound cloud host includes:

Use the Nova controller service to call the device management service to bind the relationship between the new cloud host and the device, and use the device management service to send the state of the device binding event to the Nova controller service;

Use the Nova controller service to call the Nova compute service to create a new cloud host, and use the Nova compute service to monitor the status of device binding events;

Use the Nova controller service to send the status of the device binding event to the Nova compute service;

If the Nova compute service detects that the device binding event is successful, the Nova compute service is used to generate an xml file of the new cloud host, and the new cloud host is started.

In one embodiment, the method further includes:

If there is no available computing node, use the Nova controller service to generate cloud host evacuation exception information.

In one embodiment, the PCI device and/or PCIe device includes: GPU, FPGA, NVMe and SSD.

According to another aspect of the present invention, a cloud host evacuation device in an OpenStack cloud platform is provided, the device comprising:

a request receiving module, configured to receive an evacuation request using the Nova controller service, and determine the cloud host to be evacuated according to the evacuation request;

an obtaining module, configured to obtain the device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service, wherein the device binding information includes a PCI device and/or a PCIe device;

A screening module, configured to use the Nova controller service to screen out the cloud host bound to the device from the cloud host to be evacuated according to the device binding information;

The computing node determination module is used to determine whether there is an available computing node in the cloud platform by using the Nova controller service, the cloud host of the bound device and the resource management service;

A new cloud host creation module is used to create a new cloud host on the available computing node for the cloud host bound to the device by using the Nova controller service, the device management service and the Nova compute service when there is an available computing node. Cloud hosting.

According to yet another aspect of the present invention, there is also provided a computer device comprising: at least one processor; and

A memory, where the memory stores a computer program that can be executed on the processor, and when the processor executes the program, the foregoing method for evacuating a cloud host in an OpenStack cloud platform is performed.

According to yet another aspect of the present invention, a computer-readable storage medium is also provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the aforementioned method for evacuation of a cloud host in an OpenStack cloud platform is performed.

The cloud host evacuation method, device, equipment and storage medium in the above-mentioned OpenStack cloud platform utilizes Novacontroller service, device management service, resource management service and Nova controller service to realize the automatic evacuation of the cloud host bound to the device to other computing nodes. The robustness of the multi-architecture cloud platform is improved, and the PCI/PCIe devices on the failed node can be released, so that the PCI/PCIe devices on the server can continue to be used after the server fails, saving labor costs.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can also be obtained according to these drawings without creative efforts.

1 is a schematic flowchart of a method for evacuating a cloud host in an OpenStack cloud platform according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an evacuation process of a cloud host with a bound device in another embodiment of the present invention;

3 is a schematic structural diagram of a cloud host evacuation device in an OpenStack cloud platform in another embodiment of the present invention;

FIG. 4 is an internal structure diagram of a computer device in another embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention will be further described in detail below with reference to the specific embodiments and the accompanying drawings.

It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for the purpose of distinguishing two entities with the same name but not the same or non-identical parameters. It can be seen that "first" and "second" It is only for the convenience of expression and should not be construed as a limitation to the embodiments of the present invention, and subsequent embodiments will not describe them one by one.

In one embodiment, please refer to FIG. 1 , the present invention provides a cloud host evacuation method in an OpenStack cloud platform, and the method specifically includes the following steps:

S100, the Nova controller service is used to receive the evacuation request, and the cloud host to be evacuated is determined according to the evacuation request.

Among them, the Nova controller service is a collection of control plug-ins in the Nova component in OpenStack, which is used to control the cloud host creation, deletion, scheduling, power on and off and other operations on the node in OpenStack; the evacuation request refers to the host on a computing node. When a fault occurs, all cloud hosts on the faulty node are evacuated from requests on other computing nodes, and the request includes the faulty computing node and all cloud hosts under the computing node.

S200, use the Nova controller service and the device management service to obtain device binding information of the cloud host to be evacuated, where the device binding information includes PCI devices and/or PCIe devices.

Among them, device management service refers to the PCI device management system in OpenStack, which provides PCI/PCIe device list, attribute, binding, and unbinding operation interface services; device binding information refers to the binding information of the cloud host to the host device, Preferably, the PCI device and/or the PCIe device includes: GPU, FPGA, NVMe and SSD.

S300, use the Nova controller service to screen out the cloud hosts bound to the device from the cloud hosts to be evacuated according to the device binding information.

S400: Determine whether there is an available computing node in the cloud platform by using the Nova controller service, the cloud host bound to the device, and the resource management service.

Among them, the resource management service indicates the use, scheduling, and management of resources such as CPU, Memory, Disk, PCI/PCIe devices in OpenStack; the available computing node refers to a computing node in the multi-architecture cloud platform, which can be used to create Cloud host, and provide corresponding PCI devices and/or PCIe devices.

S500, if there is an available computing node, use the Nova controller service, the device management service, and the Novacompute service to create a new cloud host on the available computing node for the cloud host bound to the device. Among them, Nova compute refers to the computing node service that starts the cloud host in OpenStack.

The above-mentioned method for evacuation of cloud hosts in the OpenStack cloud platform utilizes Nova controller services, device management services, resource management services and Nova controller services to automatically evacuate cloud hosts bound to devices to other computing nodes, thereby improving the multi-architecture cloud The robustness of the platform and the ability to release the PCI/PCIe devices on the failed node so that the PCI/PCIe devices on the server can continue to be used after the server fails to recover, saving labor costs.

Preferably, the method of the present invention further comprises the following steps:

S600, use the Nova evacuate service to evacuate the remaining cloud hosts to be evacuated.

Preferably, the method further includes: S700, if there is no available computing node, use the Nova controller service to generate cloud host evacuation exception information.

The above-mentioned method for evacuation of cloud hosts in the OpenStack cloud platform, by combining the Nova evacuate service provided by the OpenStack cloud platform to evacuate cloud hosts that are not bound to devices, thereby realizing the evacuation of all cloud hosts on the faulty computing node; The automatically evacuated cloud host automatically crawls out abnormal information, and records the log content of the cloud host evacuation, which is convenient for subsequent operation and maintenance personnel to overhaul and maintain.

Referring to FIG. 2, in another embodiment, the above step S200 specifically includes the following sub-steps:

S210, the Nova controller service invokes the device management service to obtain device binding information of the cloud host to be evacuated.

S220, the device management service returns the device binding information to the Nova controller service.

In yet another embodiment, the above step S400 specifically includes the following sub-steps:

S410, the Nova controller service obtains configuration information of the cloud host bound to the device, where the configuration information includes CPU information, memory information, and disk information.

S420: Determine the resources required by the new cloud host according to the configuration information and the device binding information of the cloud host to which the device is bound.

S430, the Nova controller service invokes the resource management service to query whether there are available computing nodes in the cloud platform that meet the resources required by the new cloud host.

In yet another embodiment, the above step S500 specifically includes the following sub-steps:

S510, use the Nova controller service to call the device management service to bind the relationship between the new cloud host and the device, and use the device management service to send the status of the device binding event to the Nova controller service.

S520, use the Nova controller service to call the Nova compute service to create a new cloud host, and use the Nova compute service to monitor the status of device binding events;

S530, use the Nova controller service to send the status of the device binding event to the Nova compute service;

S540, if the Nova compute service monitors the successful device binding event, the Nova compute service is used to generate an xml file of the new cloud host, and the new cloud host is started.

In yet another embodiment, in order to facilitate the understanding of the technical solution of the present invention, it is assumed that there are two computing nodes with PCI/PCIe devices in the cloud computing platform as an example, which are denoted as computing node 1 and computing node 2, respectively. Two cloud hosts are running on the cloud host. Cloud host 1 is bound to the GPU of computing node 1, and cloud host 2 is bound to the SSD of computing node 1. If computing node 1 is powered off, but computing node 2 is running normally, the Nova controller receives In the event of node failure of host 1, the evacuation of cloud host 1 and cloud host 2 begins. The specific process is as follows:

Step 1, Nova controller, the service calls the device management service to obtain the PCI/PCIe device information bound to the cloud host 1 and the cloud host 2, and according to the PCI/PCIe device information and the CPU, Memory, Disk in the specifications of the cloud host 1 and the cloud host 1 information, call the resource management system to select computing node 2 as the available computing node;

Step 3, the Nova controller service calls the device management service to bind the relationship between the host and the PCI/PCIe device. Since the PCI device management service binds the PCI/PCIe device for an asynchronous operation, the device management service sends the PCI/PCIe binding event to the Nova controller Serve.

Step 4, the Nova controller service calls the Nova compute service to create a new cloud host 1 and a new cloud host 2;

Step 5: On the one hand, the Nova compute service creates information such as the network card and disk of the new cloud host 1 and new cloud host 2;

Step 6, the Nova controller service sends the PCI/PCIe device binding event status to the Nova compute service;

Step 7: After Nova compute receives the PCI/PCIe device binding event successfully, it generates the xml files of the new cloud host 1 and the new cloud host 2, starts the new cloud host 1 and the cloud host 2, and then connects the cloud hosts on the computing node 1. 1 and the cloud host 2 are evacuated to the computing node 2, and the new cloud host 1 on the computing node 2 is bound to the GPU of the computing node 2, and the new cloud host 2 is bound to the SSD of the computing node 2.

At this time, for computing node 1, its GPU and SSD are released, which avoids the operation and maintenance personnel from cleaning up the binding relationship data between cloud host 1 and cloud host 2 and PCI/PCIe devices in the database. The cloud host 2 can normally run the applications in the cloud host 1 and the cloud host 2, which ensures the normal processing of cloud platform services; when the computing node 1 recovers from a fault, the GPU and SSD can be used again.

In yet another embodiment, please refer to FIG. 3 , the present invention provides a cloud host evacuation device 70 in an OpenStack cloud platform, and the device specifically includes:

The request receiving module 71 is used for receiving the evacuation request by using the Nova controller service, and determining the cloud host to be evacuated according to the evacuation request;

An obtaining module 72, configured to obtain the device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service, wherein the device binding information includes PCI devices and/or PCIe devices;

A screening module 73, configured to use the Nova controller service to screen out the cloud hosts bound to the device from the cloud hosts to be evacuated according to the device binding information;

The computing node determination module 74 is used for determining whether there is an available computing node in the cloud platform by using the Nova controller service, the cloud host of the bound device and the resource management service;

The new cloud host creation module 75 is configured to use the Nova controller service, the device management service and the Nova compute service to create a new cloud host on the available computing node for the cloud host bound to the device when there is an available computing node.

It should be noted that, for the specific limitation of the cloud host evacuation device in the OpenStack cloud platform, reference may be made to the limitation on the cloud host evacuation method in the OpenStack cloud platform above, which will not be repeated here. Each module in the cloud host evacuation device in the above-mentioned OpenStack cloud platform can be implemented in whole or in part by software, hardware and combinations thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

According to another aspect of the present invention, a computer device is provided, and the computer device may be a server. Please refer to FIG. 4 for an internal structure diagram of the computer device. The computer device includes a processor, memory, a network interface, and a database connected by a system bus. 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, an internal memory. The nonvolatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store data. The network 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, the above-described cloud host evacuation method in the OpenStack cloud platform is implemented.

According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the above-described method for evacuation of a cloud host in an OpenStack cloud platform is implemented.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage In the medium, when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other medium used in the various embodiments provided in this application may include non-volatile and/or volatile memory. Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Road (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features It is considered to be the range described in this specification.

The above-mentioned embodiments only represent several embodiments of the present application, and the descriptions thereof are specific and detailed, but should not be construed as a limitation on the scope of the invention patent. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the scope of protection of the patent of the present application shall be subject to the appended claims.

Claims (10)

1. A cloud host evacuation method in an OpenStack cloud platform is characterized by comprising the following steps:

receiving an evacuation request by utilizing a Nova controller service, and determining a cloud host to be evacuated according to the evacuation request;

acquiring device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service, wherein the device binding information comprises PCI (peripheral component interconnect) devices and/or PCIe (peripheral component interconnect express) devices;

screening out cloud hosts bound with equipment from the cloud hosts to be evacuated by utilizing Nova controller service according to the equipment binding information;

determining whether available computing nodes exist in the cloud platform by using Nova controller service, a cloud host of a binding device and resource management service;

if an available computing node exists, creating a new cloud host on the available computing node for the cloud host of the bound device using the Nova controller service, the device management service, and the Nova computer service.

2. The method of claim 1, further comprising:

and evacuating the remaining cloud hosts to be evacuated by utilizing the Nova evacuate service.

3. The method as claimed in claim 1, wherein the step of obtaining the device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service comprises:

the Nova controller service calls the equipment management service to acquire equipment binding information of the cloud host to be evacuated;

the device management service returns the device binding information to the Nova controller service.

4. The method of claim 1, wherein the step of determining whether there are available computing nodes in the cloud platform using the Nova controller service, the cloud host of the bound device, and the resource management service comprises:

the Nova controller service acquires configuration information of a cloud host of the binding device, wherein the configuration information comprises CPU information, memory information and disk information;

determining resources required by a new cloud host according to the configuration information and the equipment binding information of the cloud host of the binding equipment;

the Nova controller service calls the resource management service to inquire whether available computing nodes meeting the resources required by the new cloud host exist in the cloud platform.

5. The method of claim 1, wherein the step of creating a new cloud host on the available compute node for the cloud host of the bound device using the Nova controller service, the device management service, and the Nova computer service comprises:

calling the relationship between the new cloud host and the equipment bound by the equipment management service by using the Nova controller service, and sending the state of an equipment bound event to the Nova controller service by using the equipment management service;

calling Nova computer service by using the Nova controller service to create a new cloud host, and monitoring the state of a device binding event by using the Nova computer service;

sending the state of a device binding event to the Nova computer service by utilizing a Nova controller service;

and if the Nova computer service monitors that the equipment binding event is successful, generating an xml file of the new cloud host by using the Nova computer service, and starting the new cloud host.

6. The method of claim 1, further comprising:

and if no available computing node exists, generating cloud host evacuation exception information by utilizing the Nova controller service.

7. The method of claim 1, wherein the PCI device and/or PCIe device comprises: GPU, FPGA, NVMe, and SSD.

8. A cloud host evacuation device in an OpenStack cloud platform, the device comprising:

the request receiving module is used for receiving an evacuation request by utilizing a Nova controller service and determining a cloud host to be evacuated according to the evacuation request;

an obtaining module, configured to obtain device binding information of the cloud host to be evacuated by using the Nova controller service and the device management service, where the device binding information includes a PCI device and/or a PCIe device;

the screening module is used for screening out the cloud host bound with the equipment from the cloud host to be evacuated according to the equipment binding information by utilizing Nova controller service;

the computing node determining module is used for determining whether available computing nodes exist in the cloud platform by using Nova controller service, a cloud host of the binding equipment and resource management service;

and a new cloud host creation module, configured to, when there is an available computing node, create a new cloud host on the available computing node for the cloud host of the bound device using the Nova controller service, the device management service, and the Nova computer service.

9. A computer device, comprising:

at least one processor; and

a memory storing a computer program operable in the processor, the processor when executing the program performing the method of any of claims 1-7.

10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, is adapted to carry out the method of any one of claims 1 to 7.

Images