Supercomputing cloud computing platform management method and management platform

A technology of cloud computing platform and management method, applied in the field of supercomputing cloud computing platform management method and management platform, can solve the problems of fast computing speed and large storage capacity, and achieve the effect of data interconnection and resource sharing

Pending Publication Date: 2021-08-13
TSINGHUA UNIV
4 Cites 1 Cited by

AI-Extracted Technical Summary

Problems solved by technology

[0006] For this reason, the first purpose of this application is to propose a super-computing cloud computing platform management method, which solves the problem that traditional data centers only provide cloud computing services, and the computing speed, For technical problems that cannot be undertaken by tasks with large storage capacity, by ...
View more

Abstract

The invention provides a supercomputing cloud computing platformization management method and management platform, and relates to the technical field of cloud computing, and the supercomputing cloud computing platformization management method comprises the steps that a first cloud computing data center receives a data request sent by an external network client; the hyper-cloud platform receives the data request processed by the first cloud computing data center, and the hyper-cloud platform is deployed on a second cloud computing data center; the super-computing center receives the data request processed by the super-cloud platform, processes the data request into a simplified instruction, and calls resources of the super-computing center according to the simplified instruction to perform computing processing; and the third cloud computing data center receives the processing result of the supercomputing center. By the adoption of the scheme, the problems that a traditional data center only provides cloud computing service, and tasks which are high in computing speed and large in storage capacity and needed by the national high-tech field and the top-end technology research cannot be borne are solved, and data interconnection and resource sharing of the super computer and the cloud computing data center are achieved.

Application Domain

Resource allocation

Technology Topic

Calculated dataSupercomputer +7

Image

  • Supercomputing cloud computing platform management method and management platform
  • Supercomputing cloud computing platform management method and management platform
  • Supercomputing cloud computing platform management method and management platform

Examples

  • Experimental program(1)

Example Embodiment

[0034] The embodiments of the present application are described in detail below, and examples of the embodiments are illustrated in the drawings, wherein the same or similar reference numerals are indicated by the same or similar elements or elements having the same or similar functions. The following is exemplary, and is intended to be used to explain the present application without understanding the limitation of the present application.
[0035] The superchable cloud computing platform management method and platform of the present application embodiment will be described below with reference to the drawings.
[0036] figure 1 A flow chart of a super-calculated cloud computing platform management method provided by the present application embodiment.
[0037] like figure 1 As shown, the superior cloud computing platform management method, including the following steps:
[0038] Step 101: The first cloud computing data center receives the data request sent by the external network client;
[0039] Step 102: The super cloud platform receives the data request after the first cloud computing data center, wherein the super cloud platform is deployed on the second cloud computing data center;
[0040] Step 103, the overtinest center receives the data request processed by the super cloud platform and processes it as a simplified instruction, and calculates the resource of the super-intensive center according to the simplified instruction;
[0041] Step 104, the third cloud computing data center receives the processing result of the superior center.
[0042] The ultra-calculated cloud computing platform chemical method of the present application embodiment includes: the first cloud computing data center receives data requests sent by the external network client; the ultrasound platform receives data requests processed by the first cloud computing data center, where The ultraunction platform is deployed on the second cloud computing data center; the overtinest center receives the data request processed by the super cloud platform and process it as a simplified instruction, and calculates according to the resource of the simplified instruction calling the superior center; The three cloud computing data center receives the results of the overtime center. As a result, this application can solve the traditional data center only providing cloud computing services, and the technical issues that are fast, the task that has high storage capacity is high, and the storage capacity is unable to undertake. The general user of the computer faces certain thresholds, and the overtime centers are partially calculated, and the data calculation, storage, and interconnection of ordinary cloud computing data center is ashamed. By using the supercomputer to join the data center management, directly connect the data of the supercomputer After the center is data processing, call the resources of the supercomputer for calculation, and return the results to other data centers, implement data interconnection and resource sharing of supercomputer and cloud computing data centers.
[0043] Further, in the present application embodiment, it is also included before the external network client remotely accesses the resources in the superior center.
[0044] Establish a virtual private network on a public network, and the data request sent by the external network client is transmitted to the first cloud computing data center via encryption mode.
[0045] Further, in the present application embodiment, the overtime centered on the data request after the supernuclear platform is received and processed into simplified instructions, and the resource of the super-decision center according to the simplified command is calculated. Including internal network servers and supercomputers, where
[0046] The internal network server will simplify the data request processing by the ultrasound platform, and submit the simplified instruction to the supercomputer calculation process.
[0047] Any point that can be considered to be a stand-alone data center, the data center can complete the data forwarding operation, one of the data centers deploys "Super Cloud Platform", the ultraunction platform and the over-censative center internal server direct Communication.
[0048] Further, in the present application embodiment, calculation processing is performed in accordance with the resource based on the simplified instruction call, the calculation processing includes calls and monitoring of the job. For call and monitoring of the job, the supercomputer includes job call interfaces and Job monitor interface.
[0049] Further, in the present application embodiment, the job call interface includes data upload, data viewing, job submission, and job termination, where data is uploaded and data interface access resources on the supercomputer; where
[0050] When the data is uploaded, when the upload file size is greater than 100MB, the upload file is uploaded to the store of the supercomputer, and when the data is uploaded or the uploaded file existing in the store, the job is submitted;
[0051] When the upload file size is not more than 100MB, the upload file is uploaded directly to the store of the supercomputer via the browser, and the job submission is performed when the data is uploaded or stored in the store.
[0052] Data uploads and data View Access is the storage resource of the supercomputer.
[0053] Further, in the present application embodiment, the job monitoring interface includes job list display, job status query, job log query, and node resource query, where the job status query is inquiry for a job in the results of the job list. .
[0054] The purpose of the job list show is to show the user of all job information being executed, in the form of a list, including the name of the job, the number of resources occupying, the user, submission time, etc. The use of the job status query is to support the user query the running status of the specific job, return status information, where "Job Status" includes operational status that has been submitted to be allocated, normal operation, job error, and operation operation. The use of job log queries is to support user query running log information output at a specific job runtime. The purpose of the node resource query is to support all or part of the compute node in the user query system, and the node status includes status of occupied, idle, sleep, shutdown, and drop.
[0055] The job monitoring has a total of three interfaces, in which the job status query is a specific item in the job list. It means that the job status query can only query a job in the results of the job list, which does not support the simultaneous query. The status of the job. Job list show and job status query public an interface, only the parameters that are incorporated only.
[0056] figure 2 It is a mainstream diagram of the superchable cloud computing platform management method according to the examples of the present application.
[0057] like figure 2 As shown, the ultra-calculated cloud computing platform management method, when the server resources of the Kwai Lake are required to establish a dedicated network on the public network, and the registered users need to open the user virtual private network. VPN account, through the VPN connection, access the Light Supercomputer of the Shenwei Taihu. User (external network client) sends a RESTFUL request, requesting the layered processing to reach the super-calculated center internal network server (Skip machine), the final internal network server will submit the simplified instruction to the supercomputer process.
[0058] image 3 It is a data flow diagram of an overtecadular cloud computing platform management method according to an embodiment of the present application.
[0059] like image 3 As shown, the superchable cloud computing platform management method, any location that can be and more indirect communication can be considered a separate data center, the data center can complete the data forwarding operation, one of the data center deployment "Super Cloud Platform ", Super cloud platform and overall internal server communicate directly.
[0060] Figure 4 It is a supercomputer abstraction interface diagram of the supercomputer computing platform management method of the present application embodiment.
[0061] like Figure 4 As shown, the superior cloud computing platform management method, for the call of the job, the monitoring, the total abstraction and the seven external interfaces. The job calls a total abstract data upload, data viewing, job submission, and job termination and other four interfaces. Data uploads and data View Access is the storage resource of the supercomputer. Considering that the uploaded files may cause the browser crash or the server unable to receive, define the upload file size cannot exceed 100MB, more than this size is recommended to upload it to the Super Computer storage. Jobs can be submitted only when data is uploaded successfully or in the store. Job monitoring includes job list display, job status query, job log query, and node resource query. The job monitoring has a total of three interfaces, where the job status query is a specific item in the job list, so the two query use only one interface, which is different from the parameters that are incorporated when querying.
[0062] The second embodiment of this application has proposed a superior cloud computing platform management platform, including: first cloud computing data center, second cloud computing data center, superior center, and third cloud computing data center, in the second cloud computing The data center is deployed with a super cloud platform;
[0063] The first cloud computing data center is used to receive data requests sent by the external network client;
[0064] The super cloud platform is used to receive data requests after the first cloud computing data center;
[0065] The superior center, is used to receive the data request processed by the super cloud platform and process it as a simplified instruction, and calculate the resources according to the resource of the super-intensive center;
[0066] The third cloud computing data center is used to receive processing results from the overtime centers.
[0067] Further, in the present application embodiment, the overtinent cloud computing platform management platform also includes a virtual private network, and the virtual private network is built on the outer network client.
[0068] Further, in the present application embodiment, the superior center includes an internal web server and a supercomputer, and the internal network server will simplify the instructions from the data request processed by the super cloud platform, and submit the simplified instruction to the supercomputer calculation process.
[0069] The ultra-calculated cloud computing platform management platform of the present application embodiment includes: the first cloud computing data center, the second cloud computing data center, the superior center, and the third cloud computing data center, deploying in the second cloud computing data center. There is a super cloud platform; where the first cloud computing data center is used to receive data requests sent by the external network client; the super cloud platform is used to receive data requests after processing the first cloud computing data center; superior center, Used to receive data requests processed by the super cloud platform and process it as a simplified instruction, and calculate the resource for the simplified instruction to call the super-centered resource; the third cloud computing data center, used to receive processing by the overtinent center result. As a result, this application can solve the traditional data center only providing cloud computing services, and the technical issues that are fast, the task that has high storage capacity is high, and the storage capacity is unable to undertake. The general user of the computer faces a certain threshold, and the overall calculation center has some calculation of the resource idle problem. By using the supercomputer to join the data center management, directly connect the data center of the supercomputer for data processing, determine the resources required for the calculation task, select the appropriate Cloud computing or supercapacity resources are completed and return the results to other data centers to implement data interconnection and resource sharing of supercomputers and cloud computing data centers.
[0070] In order to achieve the above embodiment, the present invention also proposes a computer device, including a memory, processor, and a computer program stored on the memory and can run on the processor. When the processor performs a computer program, the above-described application is implemented. Calculate the platform management method and platform.
[0071] In the description of this specification, a description of the reference terms "one embodiment", "some embodiments", "example", "specifically example", or "some example", etc., meant to combine the specific characteristics described in connection with this embodiment. , Structures, materials or features are included in at least one embodiment or example of the present application. In the present specification, schematic representations of the above terms are not necessarily directed to the same embodiments or examples. Moreover, the specific features, structures, materials or features described may be combined in any one or more embodiments or examples. In addition, in the case of non-conflict, those skilled in the art can combine and combine different embodiments or features described in this specification.
[0072] Moreover, the term "first", "second" is used only for the purpose of describing, and cannot be understood as an indication or implicit relative importance or implicitting the number of techniques indicated. Thus, features with "first", "second" may be indicated or implicitly including at least one of this feature. In the description of the present application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise specifically defined.
[0073] Any process or method described in the flowchart or herein can be understood as a module, segment, or part of a code including one or more executable instructions including one or more steps for implementing the steps of custom logic functions or processes. And the range of preferred embodiments of the present application includes additional implementation, in which the function can be performed without pressing or discussed in the order, including the function according to the function, according to the substantially simultaneous manner, which should be performed. It is understood by those skilled in the art of the present invention.
[0074]The logic and / or steps shown in the flowchart or here, for example, may be considered to be a sequencing list for implementing the executable instructions for logic functions, and can be implemented in any computer readable medium. Instructions for instruction execution systems, devices, or devices (such as computer-based systems, systems including processors, or other systems that can take instructions from instruction execution systems, devices or devices, or combined with these instruction execution systems, devices Or use. For this manual, * # * computer readable media * # * can be any of the instructions that can be included, stored, communicate, propagate, or transmit programs for instruction execution systems, devices, or devices, or combined with these instructions to execute system, devices or devices. The device used is used. More specific examples of computer readable media (non-exhausted lists) include the following: electrical connecting portions (electronic devices), portable computer cartridges (magnetic devices), random access memory (RAM), random access memory Read-only memory (ROM), erased editable read only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read only memory (CDROM). In addition, the computer readable medium can even be a paper or other suitable medium that can print the program there, since the editing, interpretation or need, for example, by optically scanning the paper or other medium The process is processed to obtain the program electronically, and then store it in the computer memory.
[0075] It should be understood that the part of the present application can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiment, multiple steps or methods can be implemented in software or firmware stored in the memory and executed by the appropriate instructions. For example, if you use hardware and in another embodiment, it is possible to achieve any of the following techniques well known in the art: having a logic gate circuit for realizing logic functions for data signals. Discrete Logic circuit, a dedicated integrated circuit with a suitable combined logic gate circuit, a programmable gate array (PGA), field programmable gate array (FPGA), and the like.
[0076] One of ordinary skill in the art will appreciate that all or some of the steps that implement the above-described embodiment method are to be completed by the hardware that can be related to instructions, the program can be stored in a computer readable storage medium, which When executed, one of the steps including method embodiments or a combination thereof.
[0077] Further, each functional unit in the various embodiments of the present application can be integrated in a processing module, or each unit is generated separately, or two or more units can be integrated into one module. The above-described integrated modules can be implemented in the form of hardware or in the form of a software function module. The integrated module can be stored in a computer readable storage medium if implemented in the form of a software function module and is used as a separate product.
[0078] The storage medium mentioned above can be a read-only memory, a disk, or a disc or the like. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and cannot be understood as limiting the present application, and those of ordinary skill in the art can be described in the scope of the present application. EXAMPLES Change, modify, replace, and variants.

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products