CN106161538B

CN106161538B - Application platform management system fusing X86 and ARM architecture

Info

Publication number: CN106161538B
Application number: CN201510168380.4A
Authority: CN
Inventors: 张京梅
Original assignee: Beijing Dianzan Technology Co ltd
Current assignee: Beijing Dianzan Technology Co ltd
Priority date: 2015-04-12
Filing date: 2015-04-12
Publication date: 2020-06-12
Anticipated expiration: 2035-04-12
Also published as: CN106161538A

Abstract

The invention discloses an application platform management system fusing an X86 and an ARM architecture, which is characterized in that basic architecture as service is deployed for an X86 architecture server cluster and an ARM architecture server cluster respectively, and the basic architecture as service is used for providing computing resources of an X86 architecture and the ARM architecture and managing and scheduling the resources; abstracting a tool box for carrying out unified operation on the cloud computing infrastructure on two infrastructures, namely service platforms, and carrying out API call on different infrastructure platforms; constructing a platform, namely a service, which can carry out statistical analysis and prediction on the load and the resource occupation condition of the application on a tool box, and dynamically scheduling X86 framework computing resources or ARM framework computing resources according to the statistical analysis and prediction results; the user sets the application based on the label, dynamically selects the strategy according to the label and the application load condition, and selects the computing resources of different architectures according to different strategies.

Description

Application platform management system fusing X86 and ARM architecture

Technical Field

The invention belongs to the field of computing resources in a cloud computing infrastructure platform, and particularly relates to an application platform management system fusing an X86 and an ARM architecture.

Background

The existing cloud computing platform as a service (PaaS) can only support an X86 architecture and does not support a mixed mode of an X86 architecture server and an ARM architecture server, so that different computing resources cannot be dynamically selected according to application requirements and loads, waste of the computing resources is caused, and the cost is high.

Disclosure of Invention

In order to overcome the defects in the prior art, the invention provides an application platform management system integrating an X86 architecture and an ARM architecture, which automatically selects the computing resources of the X86 architecture or the computing resources of the ARM architecture according to the application requirements and load dynamics, and fully exerts the advantages of low power consumption of an ARM architecture server and high performance of a GPU (graphics processing unit), thereby improving the energy consumption ratio of the whole cloud computing center.

In order to achieve the purpose, the technical scheme of the invention is as follows: an application platform management system fusing an X86 architecture and an ARM architecture is characterized in that infrastructure as a service is deployed for an X86 architecture server cluster and an ARM architecture server cluster respectively, and the infrastructure as a service is used for providing computing resources of the X86 architecture and the ARM architecture and managing and scheduling the resources; abstracting a tool box for carrying out unified operation on the cloud computing infrastructure on two infrastructures, namely service platforms, and carrying out API call on different infrastructure platforms; constructing a platform, namely a service, which can carry out statistical analysis and prediction on the load and the resource occupation condition of the application on a tool box, and dynamically scheduling X86 framework computing resources or ARM framework computing resources according to the statistical analysis and prediction results; the user sets the application based on the label, dynamically selects the strategy according to the label and the application load condition, and selects the computing resources of different architectures according to different strategies.

Preferably, the system comprises the following components:

requirement/policy repository: providing an interface for defining application requirements for a user, storing the application requirements, converting the application requirements into a deployment and operation strategy, and providing the deployment and operation strategy for a platform delivery component and an operation control component;

software image library: providing basic software required by an application running platform;

a computing resource pool: providing hardware resources required by an application running platform in a virtualization mode based on Hypervisor;

platform as a service delivery component: according to the deployment or adjustment strategy, the application operating platform resources are adjusted by deploying or adjusting the software and hardware resource pool;

a runtime control component: and dynamically adjusting the computing resource pool according to the operation strategy and the actual operation condition, and dynamically providing an adjustment strategy for the platform, namely the service delivery component.

The invention has the beneficial effects that:

the application platform system supports the mixed mode of the X86 architecture server and the ARM architecture server, can automatically select the computing resources of the X86 architecture or the computing resources of the ARM architecture according to the application requirements and load dynamic, and fully exerts the advantages of low power consumption and high performance of the GPU of the ARM architecture server, thereby improving the energy consumption ratio of the whole cloud computing center.

Drawings

FIG. 1 is a schematic diagram of the overall architecture of the present invention incorporating X86 and ARM;

FIG. 2 is a schematic diagram of a platform as a service architecture according to the present invention;

FIG. 3 is a flow diagram of a fusion X86 and ARM architecture computing resource deployment adjustment component of the present invention;

FIG. 4 is a flow chart of deployment adjustment of computing resources in accordance with the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.

The whole architecture fusing the X86 and the ARM, as shown in figures 1 and 2,

1) respectively deploying infrastructure as a service (IaaS) for the X86 architecture server cluster and the ARM architecture server cluster, and providing computing resources of an X86 architecture and an ARM architecture, and managing and scheduling the resources;

2) abstracting a tool kit (ToolKit) for performing unified operation on the cloud computing infrastructure on two infrastructures, namely service platforms, and taking charge of performing API call on different infrastructure platforms;

3) constructing a platform as a service (PaaS) capable of performing statistical analysis and prediction on the load and resource occupation condition of the application on the tool box in the step 2), and dynamically scheduling X86 framework computing resources or ARM framework computing resources according to the statistical analysis and prediction results;

4) a user can set the application based on a label (Tag), dynamically select strategies according to the label and the application load condition, and select the computing resources of different architectures according to different strategies.

The strategy refers to a scheme set of X86 and ARM dynamic scheduling allocation technology resources for realizing the energy-saving target, and the strategy is stored in a requirement/strategy database.

The fused X86 and ARM architecture computing resource deployment adjustment component and flow diagram are shown in fig. 3.

1) Requirement/policy repository: providing an interface for defining application requirements for a user, storing the application requirements, converting the application requirements into a deployment and operation strategy, and providing the deployment and operation strategy for a platform delivery component and an operation control component;

2) software image library: providing basic software required by an application running platform;

3) a computing resource pool: providing hardware resources required by an application running platform in a virtualization mode based on Hypervisor;

4) platform as a service delivery component: according to the deployment or adjustment strategy, the application operating platform resources are adjusted by deploying or adjusting the software and hardware resource pool;

5) a runtime control component: and dynamically adjusting the computing resource pool according to the operation strategy and the actual operation condition, and dynamically providing an adjustment strategy for the platform, namely the service delivery component.

The flow of performing statistical analysis of resources on an application and performing deployment adjustment of computing resources according to the analysis predictions is shown in fig. 4.

The invention can give full play to the different characteristics of the X86 architecture server and the ARM architecture server to adapt to different application requirements and loads, and give full play to the advantages of low power consumption and high performance of the GPU of the ARM architecture server, thereby improving the energy consumption ratio of the whole cloud computing center.

It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.

Claims

1. An application platform management system fusing an X86 architecture and an ARM architecture is characterized in that infrastructure as a service is deployed for an X86 architecture server cluster and an ARM architecture server cluster respectively, and the infrastructure as a service is used for providing computing resources of the X86 architecture and the ARM architecture and managing and scheduling the resources; abstracting a tool box for carrying out unified operation on the cloud computing infrastructure on two infrastructures, namely service platforms, and carrying out API call on different infrastructure platforms; constructing a platform, namely a service, which can perform statistical analysis and prediction on the load and resource occupation condition of the application on a tool box, comparing the utilization rates of a CPU and a GPU by counting the utilization rates of the CPU and the GPU, expanding the computing resources of an X86 framework and updating an application requirement strategy library when the utilization rate of the CPU is greater than that of the GPU, or expanding the computing resources of an ARM framework and updating the application requirement strategy library; the user sets the application based on the label, dynamically selects the strategy according to the label and the application load condition, and selects the computing resources of different architectures according to different strategies.

2. The converged X86 and ARM architecture application platform management system of claim 1, wherein the system comprises the following components:

application requirement policy base: providing an interface for defining application requirements for a user, storing the application requirements, converting the application requirements into a deployment and operation strategy, and providing the deployment and operation strategy for a platform, namely a service delivery component and an operation control component;

software image library: providing basic software required by an application platform;

a computing resource pool: providing hardware resources required by an application platform in a virtualization mode based on a Hypervisor of a virtual machine;

platform as a service delivery component: according to the deployment and operation strategy, the application platform resources are adjusted by deploying or adjusting the software and hardware resource pool;