CN111736915B - Management method, device, equipment and medium for cloud host instance hardware acceleration equipment - Google Patents
Management method, device, equipment and medium for cloud host instance hardware acceleration equipment Download PDFInfo
- Publication number
- CN111736915B CN111736915B CN202010506258.4A CN202010506258A CN111736915B CN 111736915 B CN111736915 B CN 111736915B CN 202010506258 A CN202010506258 A CN 202010506258A CN 111736915 B CN111736915 B CN 111736915B
- Authority
- CN
- China
- Prior art keywords
- hardware acceleration
- host
- cloud
- instance
- resource
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001133 acceleration Effects 0.000 title claims abstract description 324
- 238000007726 management method Methods 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 claims abstract description 71
- 238000013507 mapping Methods 0.000 claims description 32
- 238000004590 computer program Methods 0.000 claims description 11
- 239000002699 waste material Substances 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013403 standard screening design Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Stored Programmes (AREA)
Abstract
The application discloses a management method, a device, equipment and a medium for cloud host instance hardware acceleration equipment, wherein the method comprises the following steps: when a target request of the hardware acceleration equipment is created by the cloud host instance is received, creating a hardware acceleration resource group based on the resource specification of the cloud host instance; selecting an available host from hosts under the control of a cloud platform according to a hardware acceleration resource group; judging whether hardware acceleration equipment on an available host meets a target request or not; and if so, binding the cloud host instance with hardware acceleration equipment of the available host, and starting and operating the cloud host instance. Obviously, because the method manages the hardware acceleration equipment on the cloud host instance by using the cloud platform, compared with the prior art, the method not only avoids the waste of manpower resources, but also simplifies the management process of the hardware acceleration equipment, thereby obviously improving the convenience in the management process of the hardware acceleration equipment on the cloud host instance.
Description
Technical Field
The invention relates to the technical field of servers, in particular to a management method, a management device, management equipment and a management medium for cloud host instance hardware acceleration equipment.
Background
In the actual operation process, hardware acceleration equipment is sometimes required to be added to a cloud host instance to further optimize the cloud host instance, in this case, resource information of relevant hardware acceleration equipment is firstly required to be inquired on a target host, then, the hardware acceleration equipment meeting the resource specification of the cloud host instance is selected from the resource information of the hardware acceleration equipment, finally, the selected hardware acceleration equipment is manually bound with the cloud host instance, and the cloud host instance is started and operated. Obviously, such an operation method not only needs to consume a large amount of human resources, but also causes a complex and tedious management process of the hardware acceleration device on the cloud host instance. At present, no effective solution exists for the technical problem.
Therefore, how to improve convenience in the management process of the hardware acceleration device on the cloud host instance is a technical problem to be solved urgently by technical personnel in the field.
Disclosure of Invention
In view of this, an object of the present invention is to provide a method, an apparatus, a device and a medium for managing a hardware acceleration device of a cloud host instance, so as to improve convenience in the process of managing the hardware acceleration device on the cloud host instance. The specific scheme is as follows:
a management method of a cloud host instance hardware acceleration device is applied to a cloud platform and comprises the following steps:
when a target request of a cloud host instance for creating hardware acceleration equipment is received, creating a hardware acceleration resource group based on the resource specification of the cloud host instance; the hardware acceleration resource group is a set of hardware acceleration devices with the same attribute information in all hosts under the management and control of the cloud platform, and the hardware acceleration resource group records the attribute information of the cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration devices on the target host and the mapping relationship between the target host and the hardware acceleration devices on the target host;
selecting an available host from hosts under the management and control of the cloud platform according to the hardware acceleration resource group;
determining whether a hardware acceleration device on the available host satisfies the target request;
and if so, binding the cloud host instance with the hardware acceleration equipment of the available host, and starting and operating the cloud host instance.
Preferably, the selecting an available host from hosts under the management and control of the cloud platform according to the hardware acceleration resource group includes:
selecting available target hardware acceleration resources by using the names of the hardware acceleration devices recorded by the hardware acceleration resource group;
and selecting the available host from the hosts under the management and control of the cloud platform based on the target mapping relationship between the target hardware acceleration resource and the host corresponding to the target hardware acceleration resource.
Preferably, the binding the cloud host instance with the hardware acceleration device of the available host includes:
creating a mapping relationship between the cloud host instance and the hardware acceleration device of the available host to bind the cloud host instance and the hardware acceleration device of the available host.
Preferably, after the process of determining whether the hardware acceleration device on the available host satisfies the target request, the method further includes:
and if not, deleting the mapping relation among the cloud host instance, the available host and the hardware acceleration equipment corresponding to the available host from the hardware acceleration resource group.
Preferably, after the process of deleting the mapping relationship between the cloud host instance, the available host, and the hardware acceleration device corresponding to the available host from the hardware acceleration resource group, the method further includes:
and executing the step of selecting available hosts from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group again.
Preferably, after the process of binding the hardware acceleration device of the available host to the cloud host instance, the method further includes:
judging whether the hardware acceleration equipment of the available host and the cloud host instance can be successfully bound within preset time or not;
if yes, determining that the hardware acceleration equipment of the available host is successfully bound with the cloud host instance;
if not, determining that the hardware acceleration equipment of the available host fails to be bound with the cloud host instance.
Preferably, after the process of binding the cloud host instance with the hardware acceleration device of the available host, the method further includes:
and when the life cycle of the cloud host instance is finished, deleting the binding relationship between the cloud host instance and the hardware acceleration equipment of the available host.
Correspondingly, the invention also discloses a management device of the cloud host instance hardware acceleration equipment, which is applied to a cloud platform and comprises the following components:
the resource group creation module is used for creating a hardware acceleration resource group based on the resource specification of the cloud host instance when receiving a target request of the cloud host instance for creating the hardware acceleration equipment; the hardware acceleration resource group is a set of hardware acceleration devices with the same attribute information in all hosts under the management and control of the cloud platform, and the hardware acceleration resource group records the attribute information of the cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration devices on the target host and the mapping relationship between the target host and the hardware acceleration devices on the target host;
the host selection module is used for selecting an available host from hosts under the management and control of the cloud platform according to the hardware acceleration resource group;
the request judging module is used for judging whether the hardware acceleration equipment on the available host meets the target request;
and the instance running module is used for binding the cloud host instance with the hardware acceleration equipment of the available host and running the cloud host instance when the judgment result of the request judgment module is yes.
Correspondingly, the invention also discloses a management device of the cloud host instance hardware acceleration device, which comprises the following steps:
a memory for storing a computer program;
a processor, configured to implement the steps of the management method of the cloud host instance hardware acceleration device as disclosed in the foregoing when executing the computer program.
Correspondingly, the invention also discloses a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the management method of the cloud host instance hardware acceleration device are realized.
Therefore, in the invention, when the cloud platform receives a target request of the cloud host instance for creating the hardware acceleration device, firstly, a hardware acceleration resource group is created according to the resource specification of the cloud host instance, then, an available host is selected from hosts under the management and control of the cloud platform by using the hardware acceleration resource group, whether the hardware acceleration device on the available host meets the target request of the cloud host instance for creating the hardware acceleration device is judged, if yes, the cloud host instance and the hardware acceleration device of the available host are bound, and the cloud host instance is operated. Obviously, in the method for managing the hardware acceleration device of the cloud host example, the cloud platform is used for managing the hardware acceleration device on the cloud host example, so that compared with the prior art, the method for managing the hardware acceleration device of the cloud host example not only avoids waste of manpower resources, but also simplifies the management flow of the hardware acceleration device, thereby remarkably improving convenience in the process of managing the hardware acceleration device on the cloud host example. Correspondingly, the management device, the equipment and the medium for the cloud host instance hardware acceleration equipment provided by the invention also have the beneficial effects.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a management method for a cloud host instance hardware acceleration device according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating an exemplary hardware acceleration resource group according to an embodiment of the present invention;
fig. 3 is a flowchart of another management method for a cloud host instance hardware acceleration device according to an embodiment of the present invention;
fig. 4 is a structural diagram of a management apparatus of a cloud host instance hardware acceleration device according to an embodiment of the present invention;
fig. 5 is a structural diagram of a management device of a cloud host instance hardware acceleration device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart of a management method for a cloud host instance hardware acceleration device according to an embodiment of the present invention, where the management method includes:
step S11: when a target request of the hardware acceleration equipment is created by the cloud host instance is received, creating a hardware acceleration resource group based on the resource specification of the cloud host instance;
the hardware acceleration resource group is a set of hardware acceleration devices with the same attribute information in all hosts under the management and control of the cloud platform, and the hardware acceleration resource group records the attribute information of a cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration devices on the target host and the mapping relation between the target host and the hardware acceleration devices on the target host;
step S12: selecting an available host from hosts under the management and control of the cloud platform according to a hardware acceleration resource group;
step S13: judging whether hardware acceleration equipment on an available host meets a target request;
step S14: and if so, binding the cloud host instance with the hardware acceleration equipment of the available host, and starting and operating the cloud host instance.
In this embodiment, a management method for a cloud host instance hardware acceleration device is provided, and by the management method, convenience in a management process of the hardware acceleration device on the cloud host instance can be improved.
When a cloud platform receives a target request of a cloud host instance for creating hardware acceleration equipment, firstly, a hardware acceleration resource group is created based on a resource specification of the cloud host instance, that is, the hardware acceleration resource group is created according to a creation requirement of the cloud host instance, wherein the hardware acceleration resource group is a set of hardware acceleration equipment with the same attribute information in all hosts under the management and control of the cloud platform, and in addition, the attribute information of the cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration equipment on the target host and a mapping relation between the target host and the hardware acceleration equipment on the target host are also recorded in the hardware acceleration resource group. It should be noted that, in this embodiment, the target host refers to any host under the control of the cloud platform.
In actual operation, the name of the hardware acceleration resource group can be named as follows:
{“resource_device_group_0:FPGA1”,“resource_device_group_0:FPGA2”,”
resource_device_id_0::PGPU1”,“resource_device_id_0:PGPU2”},
{“resource_device_group_1:FPGA1”,“resource_device_group_1:FPGA2”,”
resource_device_group_1:PGPU1”,“resource_device_group_1:PGPU2”}
each hardware acceleration resource group is equivalent to a dictionary, for example:
{“resource_device_group_0:FPGA1”：{‘key1’:’value1’,‘key2’:‘value2’}
in addition, in practical applications, the hardware acceleration devices are various, such as: a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), a DPDK (Data Plane Development Kit), a SPDK (Storage Plane Development Kit), and so on, so that the target request of the cloud host instance to create the hardware acceleration device may be a target request of the cloud host instance to create the GPU, or a target request of the cloud host instance to create the FPGA, the DPDK, or the SPDK, and the created hardware acceleration resource group includes a GPU resource group, an FPGA resource group, a DPDK resource group, or an SPDK resource group.
Referring to fig. 2, fig. 2 is a schematic diagram of a hardware acceleration resource set according to an embodiment of the present invention. In the hardware acceleration resource group shown in fig. 2, there are N hardware acceleration resource groups, where each hardware acceleration resource group is a set of different hardware acceleration devices, such as: a set of FPGAs, a set of GPUs, or a set of SSDs. Obviously, the process of creating the hardware acceleration device by the cloud host instance can be more flexible and diverse by the setting mode.
After the cloud platform creates the hardware acceleration resource group based on the resource specification of the cloud host instance, the cloud platform selects the available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group, and it can be understood that the cloud platform can select the available host meeting the creation requirement of the cloud host instance from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group because the attribute information of all the hosts under the management and control of the cloud platform and the attribute information of the corresponding hardware acceleration equipment on each host are recorded in the hardware acceleration resource group.
It can be thought that the available host selected and obtained from the hardware acceleration resource group by the cloud platform may have a situation that does not meet the cloud host instance creation request, so in this embodiment, after the available host is selected and obtained, it is further determined whether the hardware acceleration device on the available host meets the target request of the cloud host instance, if it is determined that the hardware acceleration device on the available host does not meet the target request of the cloud host instance, in this case, it is described that the cloud host instance creation is failed, at this time, the process may be ended, thereby avoiding waste of cloud platform resource overhead; if the hardware acceleration equipment on the available host meets the target request of the cloud host instance, the cloud host instance and the hardware acceleration equipment of the available host are bound, so that the occupation of other cloud host instances on the available host can be avoided, and after the cloud host instance and the hardware acceleration equipment of the available host are bound, the cloud host instance is started and operated, so that the process of adding the hardware acceleration equipment on the cloud host instance is completed.
It should be noted that, in the process of binding the cloud host instance and the hardware acceleration device of the available host, if the cloud platform can successfully receive the return message that the cloud host instance and the hardware acceleration device of the available host are successfully bound, it is indicated that the cloud host instance and the hardware acceleration device of the available host are successfully bound, and in this case, the cloud platform may invoke a related driver to generate an XML file of the cloud host instance to start and operate the cloud host instance.
Obviously, in the management method for the cloud host instance hardware acceleration device provided in this embodiment, because the cloud platform is used to create the cloud host instance and the hardware acceleration device is added to the cloud host instance, in this case, not only can the consumption of manpower resources be avoided, but also the convenience in the management process of the hardware acceleration device on the cloud host instance can be improved. In addition, the cloud platform is used for managing the hardware acceleration equipment on the cloud host instance, and the usability and convenience of the cloud host instance in the actual use process can be improved.
It should be noted that, in this embodiment, if there are only 1 hardware acceleration resource group, and there are M hardware acceleration devices in the hardware acceleration resource group, then there are M hardware acceleration devices that the cloud host instance can bind to; if all hardware acceleration devices in the created hardware acceleration resource group are the same, the hardware acceleration device which can be bound by the cloud host instance is the hardware acceleration device
Wherein n is greater than 1; if n hardware acceleration resource groups are created, wherein the hardware acceleration devices in each hardware acceleration resource group are the same, and when M hardware acceleration devices in each hardware acceleration resource group are created, the number of the hardware acceleration devices that can be bound by the cloud host instance is n × M.
As can be seen, in this embodiment, when the cloud platform receives a target request for creating a hardware acceleration device from a cloud host instance, first, a hardware acceleration resource group is created according to a resource specification of the cloud host instance, then, an available host is selected from hosts under control of the cloud platform by using the hardware acceleration resource group, and it is determined whether a hardware acceleration device on the available host meets the target request for creating the hardware acceleration device from the cloud host instance, and if so, the cloud host instance and the hardware acceleration device of the available host are bound, and the cloud host instance is run. Obviously, in the method for managing the hardware acceleration device of the cloud host example, the cloud platform is used for managing the hardware acceleration device on the cloud host example, so that compared with the prior art, the method for managing the hardware acceleration device of the cloud host example not only avoids waste of manpower resources, but also simplifies the management flow of the hardware acceleration device, thereby remarkably improving convenience in the process of managing the hardware acceleration device on the cloud host example.
Based on the above embodiments, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the above steps: the process of selecting the available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group comprises the following steps:
selecting available target hardware acceleration resources by using the names of the hardware acceleration devices recorded by the hardware acceleration resource group;
and selecting an available host from the hosts under the control of the cloud platform based on the target hardware acceleration resource and the target mapping relation between the target hardware acceleration resource and the host corresponding to the target hardware acceleration resource.
It can be understood that, because the attribute information of the hardware acceleration devices of all hosts under the management and control of the cloud platform is stored and recorded in the hardware acceleration resource group, and when the cloud platform receives a target request for creating the hardware acceleration devices by a cloud host instance, the target request necessarily contains a hardware name of the hardware acceleration device that the cloud host instance wants to create, in the process of selecting an available host, the cloud platform can select a target hardware acceleration resource whose resource in the hardware acceleration resource group is not occupied according to the name of the hardware acceleration device recorded in the hardware acceleration resource group, that is, select an available target hardware acceleration resource from the hardware acceleration resource group; when the target hardware acceleration resource is selected, an available host is selected from the hosts under the management and control of the cloud platform based on a target mapping relation between the target hardware acceleration resource and the host corresponding to the target hardware acceleration resource.
Obviously, the technical scheme provided by the embodiment can ensure the credibility when the available host is selected.
Based on the above embodiment, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the above steps: the process of binding a cloud host instance with a hardware acceleration device of an available host includes:
and creating a mapping relation between the cloud host instance and the hardware acceleration equipment of the available host to bind the cloud host instance and the hardware acceleration equipment of the available host.
In actual operation, in the process of binding the cloud host instance and the hardware acceleration device of the available host, the cloud host instance and the hardware acceleration device of the available host may be bound by creating a mapping relationship between the cloud host instance and the hardware acceleration device of the available host. It can be appreciated that, when a mapping relationship between the cloud host instance and the hardware acceleration device of the available host is created, and the cloud host instance and the hardware acceleration device of the available host are bound, the cloud host instance can use the hardware acceleration device on the available host to enable the cloud host instance to perform better operation.
Therefore, by the technical scheme provided by the embodiment, the overall reliability in the binding process of the cloud host instance and the hardware acceleration device of the available host can be further ensured.
Based on the foregoing embodiments, the present embodiment further describes and optimizes a technical solution, please refer to fig. 3, and fig. 3 is a flowchart of another management method for a cloud host instance hardware acceleration device according to an embodiment of the present invention. As a preferred embodiment, the step S13: after the process of determining whether the hardware acceleration device on the available host satisfies the target request, the method further includes:
step S15: and if not, deleting the mapping relation among the cloud host instance, the available host and the hardware acceleration equipment corresponding to the available host from the hardware acceleration resource group.
It can be understood that, if it is determined that the hardware acceleration device on the available host does not satisfy the target request of the cloud host instance, it indicates that the available host does not meet the requirement for creating the cloud host instance, at this time, to avoid that the cloud host instance may also search for the available host in the process of creating the cloud host instance and waste the resource overhead of the cloud platform, the cloud platform deletes the mapping relationship between the cloud host instance, the available host and the hardware acceleration device corresponding to the available host from the hardware acceleration resource group, so that the cloud host instance does not find the available host from the host under the control of the cloud platform in the subsequent process of creating the hardware acceleration device, and thus the resource utilization rate of the cloud platform can be relatively improved.
In a preferred embodiment, the step S15: after the process of deleting the mapping relationship among the cloud host instance, the available host and the hardware acceleration device corresponding to the available host from the hardware acceleration resource group, the method further comprises the following steps:
and the step of selecting the available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group is executed again.
In this embodiment, after the hardware acceleration device of the available host does not satisfy the cloud host instance target request and the mapping relationship between the cloud host instance, the available host, and the hardware acceleration device corresponding to the available host is deleted from the hardware acceleration resource group, the step of selecting the available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group is further performed, and thus the success probability of selecting the available host from all hosts under the management and control of the cloud platform by the cloud platform is further improved.
Based on the above embodiments, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the above steps: after the process of binding the hardware acceleration device of the available host with the cloud host instance, the method further includes:
judging whether hardware acceleration equipment of an available host and a cloud host instance can be successfully bound within preset time;
if yes, determining that the hardware acceleration equipment of the available host is successfully bound with the cloud host instance;
if not, determining that the binding between the hardware acceleration equipment of the available host and the cloud host instance fails.
Specifically, in this embodiment, it is determined whether the hardware acceleration device of the available host can successfully establish the binding relationship with the cloud host instance by determining whether the hardware acceleration device of the available host and the cloud host instance can be successfully bound within the preset time, that is, if the hardware acceleration device of the available host and the cloud host instance can be successfully bound within the preset time, it indicates that the hardware acceleration device of the available host has successfully established the binding relationship with the cloud host instance, and if the hardware acceleration device of the available host and the cloud host instance cannot establish the binding relationship within the preset time, it indicates that the hardware acceleration device of the available host and the cloud host instance are unsuccessfully bound.
Therefore, by the technical scheme provided by the embodiment, the accuracy and the reliability of the cloud platform in the management process of the cloud host instance hardware acceleration equipment can be further ensured.
Based on the above embodiments, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the above steps: after the process of binding the cloud host instance with the hardware acceleration device of the available host, the method further includes:
and when the life cycle of the cloud host instance is finished, deleting the binding relationship between the cloud host instance and the hardware acceleration equipment of the available host.
In this embodiment, in order to avoid that the cloud host instance still occupies the hardware acceleration device on the available host at the end of the life cycle of the cloud host instance, the binding relationship between the cloud host instance and the hardware acceleration device of the available host is deleted at the end of the life cycle of the cloud host instance. It can be thought that, after the binding relationship between the cloud host instance and the hardware acceleration device of the available host is deleted, the hardware acceleration device on the available host can be restored to an available state, so that the resource utilization rate of the hardware acceleration device by the cloud platform can be further improved.
Referring to fig. 4, fig. 4 is a structural diagram of a management apparatus of a cloud host instance hardware acceleration device according to an embodiment of the present invention, where the management apparatus includes:
the resource group creating module 21 is configured to, when a target request for creating the hardware acceleration device by the cloud host instance is received, create a hardware acceleration resource group based on the resource specification of the cloud host instance; the hardware acceleration resource group is a set of hardware acceleration equipment with the same attribute information in all hosts under the management and control of the cloud platform, and the hardware acceleration resource group records the attribute information of a cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration equipment on the target host and the mapping relation between the target host and the hardware acceleration equipment on the target host;
the host selecting module 22 is used for selecting an available host from hosts under the management and control of the cloud platform according to the hardware acceleration resource group;
a request judging module 23, configured to judge whether a hardware acceleration device on an available host meets a target request;
and the instance running module 24 is configured to, when the determination result of the request determining module is yes, bind the cloud host instance with the hardware acceleration device of the available host, and run the cloud host instance.
Preferably, the host selection module 22 includes:
the resource selection unit is used for selecting available target hardware acceleration resources by using the names of the hardware acceleration equipment recorded by the hardware acceleration resource group;
and the host selecting unit is used for selecting an available host from the hosts under the management and control of the cloud platform based on the target mapping relationship between the target hardware acceleration resource and the host corresponding to the target hardware acceleration resource.
Preferably, the instance execution module 24 includes:
and the mapping creating unit is used for creating a mapping relation between the cloud host instance and the hardware acceleration equipment of the available host so as to bind the cloud host instance and the hardware acceleration equipment of the available host.
Preferably, the method further comprises the following steps:
and a mapping deletion module, configured to delete, when the determination result of the request determination module 23 is negative, the mapping relationship between the cloud host instance, the available host, and the hardware acceleration device corresponding to the available host from the hardware acceleration resource group.
Preferably, the method further comprises the following steps:
and the host reselection module is used for performing the step of selecting the available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group after deleting the mapping relation among the cloud host instance, the available host and the hardware acceleration equipment corresponding to the available host from the hardware acceleration resource group.
Preferably, the method further comprises the following steps:
the binding judgment module is used for judging whether the hardware acceleration equipment of the available host and the cloud host instance can be successfully bound within preset time after the process of binding the hardware acceleration equipment of the available host and the cloud host instance;
the first judgment module is used for judging that the hardware acceleration equipment of the available host is successfully bound with the cloud host instance when the judgment result of the binding judgment module is yes;
and the second judging module is used for judging that the hardware acceleration equipment of the available host fails to be bound with the cloud host instance when the judgment result of the binding judging module is negative.
Preferably, the method further comprises the following steps:
and the binding removing module is used for deleting the binding relationship between the cloud host instance and the hardware acceleration equipment of the available host when the life cycle of the cloud host instance is finished after the process of binding the cloud host instance and the hardware acceleration equipment of the available host.
The management device for the cloud host instance hardware acceleration equipment provided by the embodiment of the invention has the beneficial effects of the management method for the cloud host instance hardware acceleration equipment disclosed in the foregoing.
Referring to fig. 5, fig. 5 is a structural diagram of a management device of a cloud host instance hardware acceleration device according to an embodiment of the present invention, where the management device includes:
a memory 31 for storing a computer program;
the processor 32 is configured to implement the steps of the management method of the cloud host instance hardware acceleration device as disclosed in the foregoing when executing the computer program.
The management device of the cloud host instance hardware acceleration device provided by the embodiment of the invention has the beneficial effects of the management method of the cloud host instance hardware acceleration device disclosed in the foregoing.
Correspondingly, the embodiment of the invention also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the management method for the cloud host instance hardware acceleration device are realized.
The computer-readable storage medium provided by the embodiment of the invention has the beneficial effects of the management method of the cloud host instance hardware acceleration equipment disclosed in the foregoing.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The method, the device, the equipment and the medium for managing the cloud host instance hardware acceleration equipment provided by the invention are introduced in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (8)
1. A management method of a cloud host instance hardware acceleration device is applied to a cloud platform and comprises the following steps:
when a target request of a cloud host instance for creating hardware acceleration equipment is received, creating a hardware acceleration resource group based on the resource specification of the cloud host instance; the hardware acceleration resource group is a set of hardware acceleration devices with the same attribute information in all hosts under the management and control of the cloud platform, and the hardware acceleration resource group records the attribute information of the cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration devices on the target host and the mapping relationship between the target host and the hardware acceleration devices on the target host;
the name of the hardware acceleration resource group is named as the following form:
{“resource_device_group_0: FPGA1”, “ resource_device_group_0:FPGA2”,”
resource_device_id_0::PGPU1”, “ resource_device_id_0:PGPU2”},
{“resource_device_group_1:FPGA1”, “ resource_device_group_1:FPGA2”,”
resource_device_group_1:PGPU1”, “ resource_device_group_1:PGPU2”}
wherein each of the set of hardware acceleration resources corresponds to a dictionary,
{“resource_device_group_0:FPGA1”：{‘key1’:’value1’, ‘key2’: ‘value2’}；
the target request for creating the hardware acceleration device by the cloud host instance may be a target request for creating a GPU by the cloud host instance, or may also be a target request for creating an FPGA, a DPDK, or an SPDK by the cloud host instance, so that the hardware acceleration resource group created by the cloud host instance includes a GPU resource group, an FPGA resource group, a DPDK resource group, or an SPDK resource group;
selecting an available host from hosts under the management and control of the cloud platform according to the hardware acceleration resource group;
determining whether a hardware acceleration device on the available host satisfies the target request;
if yes, binding the cloud host instance with hardware acceleration equipment of the available host, and starting and operating the cloud host instance;
the process of selecting an available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group includes:
selecting available target hardware acceleration resources by using the names of the hardware acceleration devices recorded by the hardware acceleration resource group;
selecting the available host from the hosts under the management and control of the cloud platform based on a target mapping relation between the target hardware acceleration resource and the host corresponding to the target hardware acceleration resource;
the process of binding the cloud host instance with the hardware acceleration device of the available host includes:
creating a mapping relationship between the cloud host instance and the hardware acceleration device of the available host to bind the cloud host instance and the hardware acceleration device of the available host.
2. The method of claim 1, wherein the determining whether the hardware acceleration device on the available host satisfies the target request further comprises:
and if not, deleting the mapping relation among the cloud host instance, the available host and the hardware acceleration equipment corresponding to the available host from the hardware acceleration resource group.
3. The method for managing according to claim 2, wherein after the process of deleting the mapping relationship between the cloud host instance, the available host, and the hardware acceleration device corresponding to the available host from the hardware acceleration resource group, further comprising:
and executing the step of selecting available hosts from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group again.
4. The management method according to claim 1, wherein after the binding the hardware acceleration device of the available host with the cloud host instance, the method further comprises:
judging whether the hardware acceleration equipment of the available host and the cloud host instance can be successfully bound within preset time;
if yes, determining that the hardware acceleration equipment of the available host is successfully bound with the cloud host instance;
if not, determining that the hardware acceleration equipment of the available host fails to be bound with the cloud host instance.
5. The management method according to any one of claims 1 to 4, wherein after the binding the cloud host instance with the hardware acceleration device of the available host, further comprising:
and when the life cycle of the cloud host instance is finished, deleting the binding relationship between the cloud host instance and the hardware acceleration equipment of the available host.
6. The management device of the hardware acceleration equipment of the cloud host instance is applied to a cloud platform and comprises the following components:
the resource group creating module is used for creating a hardware acceleration resource group based on the resource specification of the cloud host instance when a target request of the hardware acceleration equipment created by the cloud host instance is received; the hardware acceleration resource group is a set of hardware acceleration devices with the same attribute information in all hosts under the management and control of the cloud platform, and the hardware acceleration resource group records the attribute information of the cloud host instance, the attribute information of a target host under the management and control of the cloud platform, the attribute information of the hardware acceleration devices on the target host and the mapping relationship between the target host and the hardware acceleration devices on the target host;
the name of the hardware acceleration resource group is named as the following form:
{“resource_device_group_0: FPGA1”, “ resource_device_group_0:FPGA2”,”
resource_device_id_0::PGPU1”, “ resource_device_id_0:PGPU2”},
{“resource_device_group_1:FPGA1”, “ resource_device_group_1:FPGA2”,”
resource_device_group_1:PGPU1”, “ resource_device_group_1:PGPU2”}
wherein each of the sets of hardware acceleration resources corresponds to a dictionary,
{“resource_device_group_0:FPGA1”：{‘key1’:’value1’, ‘key2’: ‘value2’}；
the target request for creating the hardware acceleration device by the cloud host instance may be a target request for creating a GPU by the cloud host instance, or may also be a target request for creating an FPGA, a DPDK, or an SPDK by the cloud host instance, so that the hardware acceleration resource group created by the cloud host instance includes a GPU resource group, an FPGA resource group, a DPDK resource group, or an SPDK resource group;
the host selection module is used for selecting an available host from hosts under the management and control of the cloud platform according to the hardware acceleration resource group;
the request judging module is used for judging whether the hardware acceleration equipment on the available host meets the target request;
the instance running module is used for binding the cloud host instance with the hardware acceleration equipment of the available host and running the cloud host instance when the judgment result of the request judgment module is yes;
the process of selecting an available host from the hosts under the management and control of the cloud platform according to the hardware acceleration resource group includes:
selecting available target hardware acceleration resources by using the names of the hardware acceleration devices recorded by the hardware acceleration resource group;
selecting the available host from the hosts under the management and control of the cloud platform based on a target mapping relation between the target hardware acceleration resource and the host corresponding to the target hardware acceleration resource;
the process of binding the cloud host instance with the hardware acceleration device of the available host includes:
creating a mapping relationship between the cloud host instance and the hardware acceleration device of the available host to bind the cloud host instance and the hardware acceleration device of the available host.
7. A management device of a cloud host instance hardware acceleration device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of a method for managing a cloud host instance hardware acceleration device of any one of claims 1 to 5 when executing said computer program.
8. A computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of a management method for a cloud host instance hardware acceleration device according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010506258.4A CN111736915B (en) | 2020-06-05 | 2020-06-05 | Management method, device, equipment and medium for cloud host instance hardware acceleration equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010506258.4A CN111736915B (en) | 2020-06-05 | 2020-06-05 | Management method, device, equipment and medium for cloud host instance hardware acceleration equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111736915A CN111736915A (en) | 2020-10-02 |
| CN111736915B true CN111736915B (en) | 2022-07-05 |
Family
ID=72648348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010506258.4A Active CN111736915B (en) | 2020-06-05 | 2020-06-05 | Management method, device, equipment and medium for cloud host instance hardware acceleration equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111736915B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113674131A (en) * | 2021-07-21 | 2021-11-19 | 山东海量信息技术研究院 | Hardware accelerator equipment management method and device, electronic equipment and storage medium |
| CN115629876B (en) * | 2022-10-19 | 2023-07-28 | 慧之安信息技术股份有限公司 | Intelligent video processing method and system based on extensible hardware acceleration |
| CN115361283B (en) * | 2022-10-20 | 2023-03-24 | 深圳依时货拉拉科技有限公司 | K8S-based cloud host management method and device and computer equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107239346A (en) * | 2017-06-09 | 2017-10-10 | 郑州云海信息技术有限公司 | A kind of whole machine cabinet computing resource tank node and computing resource pond framework |
| CN107241421A (en) * | 2017-06-21 | 2017-10-10 | 北京云联万维技术有限公司 | A kind of cloud host resource method for obligating and device |
| WO2018196462A1 (en) * | 2017-04-26 | 2018-11-01 | 中兴通讯股份有限公司 | Resource scheduling apparatus, resource scheduling system and resource scheduling method |
| CN108833592A (en) * | 2018-07-13 | 2018-11-16 | 无锡华云数据技术服务有限公司 | Cloud host schedules device optimization method, device, equipment and storage medium |
| CN109495555A (en) * | 2018-11-01 | 2019-03-19 | 郑州云海信息技术有限公司 | A kind of method and apparatus based on cloud platform creation cloud host |
| CN110764901A (en) * | 2019-09-17 | 2020-02-07 | 阿里巴巴集团控股有限公司 | Data processing method based on GPU (graphics processing Unit) resources, electronic equipment and system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9697006B2 (en) * | 2012-12-19 | 2017-07-04 | Nvidia Corporation | Technique for performing memory access operations via texture hardware |
| US9983938B2 (en) * | 2015-04-17 | 2018-05-29 | Microsoft Technology Licensing, Llc | Locally restoring functionality at acceleration components |
| CN107707515B (en) * | 2017-02-15 | 2018-06-08 | 贵州白山云科技有限公司 | A kind of method and device that Intelligent Hybrid acceleration is carried out to different safety class resource |
-
2020
- 2020-06-05 CN CN202010506258.4A patent/CN111736915B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018196462A1 (en) * | 2017-04-26 | 2018-11-01 | 中兴通讯股份有限公司 | Resource scheduling apparatus, resource scheduling system and resource scheduling method |
| CN107239346A (en) * | 2017-06-09 | 2017-10-10 | 郑州云海信息技术有限公司 | A kind of whole machine cabinet computing resource tank node and computing resource pond framework |
| CN107241421A (en) * | 2017-06-21 | 2017-10-10 | 北京云联万维技术有限公司 | A kind of cloud host resource method for obligating and device |
| CN108833592A (en) * | 2018-07-13 | 2018-11-16 | 无锡华云数据技术服务有限公司 | Cloud host schedules device optimization method, device, equipment and storage medium |
| CN109495555A (en) * | 2018-11-01 | 2019-03-19 | 郑州云海信息技术有限公司 | A kind of method and apparatus based on cloud platform creation cloud host |
| CN110764901A (en) * | 2019-09-17 | 2020-02-07 | 阿里巴巴集团控股有限公司 | Data processing method based on GPU (graphics processing Unit) resources, electronic equipment and system |
Non-Patent Citations (2)
| Title |
|---|
| GPU-Accelerated Computation for Electromagnetic Scattering of a Double-Layer Vegetation Model;Xiang Su ET AL;《 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing ( Volume: 6, Issue: 4, Aug. 2013)》;20130429;全文 * |
| 两段式虚拟网络功能硬件加速资源部署机制;范宏伟等;《计算机应用》;20181231(第009期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111736915A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111736915B (en) | Management method, device, equipment and medium for cloud host instance hardware acceleration equipment | |
| KR101643022B1 (en) | Catalog-based software component management | |
| CN108287708B (en) | Data processing method and device, server and computer readable storage medium | |
| CN106527958B (en) | Method and equipment for writing data and acquiring data for distributed storage system | |
| CN110018998B (en) | File management method and system, electronic equipment and storage medium | |
| EP3260993A1 (en) | File operation method and device | |
| CN114416667B (en) | Method and device for rapidly sharing network disk file, network disk and storage medium | |
| CN110020305B (en) | Webpage loading method, webpage loading device, computer equipment and storage medium | |
| CN113806300B (en) | Data storage method, system, device, equipment and storage medium | |
| CN110975293A (en) | Method, device, server and medium for establishing resource reference relation table | |
| CN112579307A (en) | Physical lock resource allocation detection method and device and electronic equipment | |
| US9298499B2 (en) | Identifier generation using named objects | |
| CN114667506A (en) | Management of multi-physical function non-volatile memory devices | |
| CN115114232A (en) | Method, device and medium for enumerating historical version objects | |
| CN113687779B (en) | Data migration method, device, electronic equipment and readable storage medium | |
| CN112559127B (en) | Virtual machine creation method, device, host and storage medium | |
| JP2018510428A (en) | Code caching system | |
| US20140365430A1 (en) | Information processing apparatus, system, and control method | |
| CN115098297B (en) | Consistent snapshot generation method and system for cloud primary storage data volume | |
| CN115080114B (en) | Application program transplanting processing method, device and medium | |
| CN115150392A (en) | Remote file copying method, system, computing equipment and storage medium | |
| CN114443583A (en) | Method, device and equipment for arranging fragment space and storage medium | |
| CN114070844A (en) | File downloading method and device, computing equipment and storage medium | |
| CN113377724A (en) | Cache space management method, device and storage medium | |
| US11017032B1 (en) | Document recovery utilizing serialized data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231117 Address after: Room 2301, No. 395 Linjiang Avenue, Tianhe District, Guangzhou City, Guangdong Province, 510655 (Location: Self made Unit 01) Patentee after: Guangdong Inspur Intelligent Computing Technology Co.,Ltd. Patentee after: INSPUR ELECTRONIC INFORMATION INDUSTRY Co.,Ltd. Address before: No. 1036, Shandong high tech Zone wave road, Ji'nan, Shandong Patentee before: INSPUR ELECTRONIC INFORMATION INDUSTRY Co.,Ltd. |
|
| TR01 | Transfer of patent right |