Efficient resource task matching method and system

Abstract

The invention discloses an efficient resource task matching method and system. The method comprises the steps that computing power resources and task requirements are obtained in real time; selecting tide resources from the computing power resources; selecting flexible requirements from the task requirements; estimating the computing power X of the tidal resources and the computing power Y required for processing the flexible demand, and if X is greater than or equal to Y, processing the flexible demand by using the tidal resources; wherein the tidal resources are defined as idle machine resources in a set time period; the flexible demand is defined as follows: the user has no requirement on the processing time of the demand, and the demand can be flexibly processed according to the computing power condition. According to the method and the device, the requirements and the resources are classified, and non-emergency tasks are processed by adopting low-cost tidal resources, so that the technical effect of improving the utilization rate of the machine resources is achieved.

Description

Technical Field

[0001] The present application relates to the field of cloud applications, and in particular to an efficient resource task matching method and system. Background Art

[0002] In the field of cloud applications, such as cloud rendering, the existing task scheduling system can schedule tasks according to preset priorities and resource availability, but lacks consideration of cost efficiency. In certain time periods, resources may be overused, causing user tasks to queue up, and the machine resource cost is high during this period, which increases the processing cost for non-urgent tasks; while in certain time periods, there is idleness or a small number of tasks, resulting in uneven resource utilization, causing non-urgent tasks to occupy resources during peak hours, and failing to maximize the economic benefits of machine resources.

[0003] Therefore, there is an urgent need for efficient resource-task matching methods and systems. Summary of the invention

[0004] The purpose of this application is to provide an efficient resource-task matching method and system to solve the technical problem of inefficient resource utilization in the prior art. By classifying demands and resources and using low-cost tidal resources to process non-urgent tasks, the technical effect of improving the utilization rate of machine resources is achieved.

[0005] In view of the above problems, an embodiment of the present application provides an efficient resource task matching method and system.

[0006] In a first aspect, the present application provides an efficient resource task matching method, the method comprising:

[0007] Obtain computing resources and task requirements in real time;

[0008] Selecting tidal resources from the computing power resources;

[0009] selecting flexible requirements from among the task requirements;

[0010] Estimate the computing power X of the tidal resource and the computing power Y required to process the flexible demand, and if X≥Y, use the tidal resource to process the flexible demand;

[0011] Among them, the tidal resources are defined as: machine resources that are idle within a set time period; the flexible demand is defined as: the user has no requirements on the processing time of the demand, and the demand can be flexibly processed according to the computing power situation.

[0012] Furthermore, the computing power resources also include fixed resources and flexible call resources;

[0013] The fixed resources are self-owned resources; the flexible resources are leased resources;

[0014] The task requirements also include fixed requirements and specific requirements;

[0015] Fixed demand refers to the demand for users to process similar tasks for a long time, and the required computing power is relatively fixed; specific demand refers to the demand that users need to start processing in a timely manner and the task delivery time is tight;

[0016] The tidal resources come from the fixed resources and / or the flexible resources that meet the definition of tidal resources.

[0017] Furthermore, the fixed resources are used to process the fixed requirements; and the flexibly called resources are used to process the specific requirements.

[0018] Further, the processing of the flexible demand by using the tidal resource includes:

[0019] Split flexible demand tasks: Set splitting factors, split flexible demand tasks according to the set splitting factors, and obtain multiple task blocks.

[0020] Further, the using the tidal resource to process the flexible demand also includes:

[0021] Estimate the computing power that may be consumed by processing the split task blocks, arrange the task blocks that can be processed in parallel, and set the computing power required for each task block to A. n , n is a natural number, then A1 to A n The sum is Y.

[0022] Further, the using the tidal resource to process the flexible demand also includes:

[0023] Calculate the computing power of tidal resources: By setting the number of idle machines in a time period, the capacity of each machine, and the idle time of each machine, the total computing power of tidal resources X and the computing power of each divisible resource in tidal resources B are calculated. n , n is a natural number.

[0024] Further, the using the tidal resource to process the flexible demand also includes:

[0025] Resource and demand matching processing: Task blocks are matched according to the tidal resource computing power. The matching method is:

[0026] When the computing power required for task block processing is A n Less than the total computing power of the tidal resource X or the computing power of any sub-resource B n , increase the number of parallel processing of task blocks, and use the computing power of the tidal resource to process the task block;

[0027] If there are any remaining task blocks that have not been processed, they will be marked and processed first after tidal resources are generated in the next period.

[0028] In a second aspect, the present application also provides an efficient cloud desktop creation management system, which is characterized by including:

[0029] A first acquisition module, the first acquisition module is used to acquire computing resources and task requirements in real time;

[0030] A first selection module, the first selection module is used to select tidal resources from the computing power resources;

[0031] A second selection module, the second selection module is used to select flexible requirements from the task requirements;

[0032] A first estimation module, the first estimation module is used to estimate the computing power X of the tidal resource and the computing power Y required to process the flexible demand, and if X≥Y, use the tidal resource to process the flexible demand;

[0033] Among them, the tidal resources are defined as: machine resources that are idle within a set time period; the flexible demand is defined as: the user has no requirements on the processing time of the demand, and the demand can be flexibly processed according to the computing power situation.

[0034] In a third aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described in any one of the first aspects is implemented.

[0035] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0036] The method proposed in this application flexibly utilizes tidal resources and different resource characteristics of different businesses. It flexibly adjusts the task execution time according to business needs and resource costs, improves the utilization efficiency of rendering resources, and reduces resource usage costs.

[0037] To solve the above problems, a smooth scheduling method for tidal resources and fluctuating tasks is proposed. By classifying the demand and resources, low-cost tidal resources are used to process non-urgent tasks, thus improving the utilization rate of machine resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0039] Figure 1 A schematic diagram of the matching principle of resources and tasks in an embodiment of the present application;

[0040] Figure 2 This is a flowchart of a tidal resource processing task according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] The embodiments of the present application solve the technical problem in the prior art that efficiency and personalization cannot be taken into account at the same time by providing an efficient resource task matching method and system.

[0042] Below, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described here. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] The overall idea of ​​the technical solution provided by this application is as follows:

[0044] First, the computing resources and task requirements are acquired in real time; then, tidal resources are selected from the computing resources; flexible requirements are selected from the task requirements; and the computing power X of the tidal resources and the computing power Y required to process the flexible requirements are estimated. If X ≥ Y, the tidal resources are used to process the flexible requirements. That is, the idle machine resources (tidal resources) within the set time period are used to process the user's requirements that have no requirements for processing time and can be flexibly processed according to the computing power situation.

[0045] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0046] Embodiment 1

[0047] like Figure 1 Figure 2 As shown, an embodiment of the present application provides an efficient resource task matching method, the method comprising:

[0048] S1, obtain computing resources and task requirements in real time;

[0049] S2. Select tidal resources from the computing power resources; computing power resources include fixed resources, flexible call resources, and tidal resources;

[0050] Fixed resources are self-owned resources, which are the resources used by the operator to build its own computer room. They provide quantitative resources for long-term cooperative users to process tasks.

[0051] Flexible resource allocation refers to leased resources. Leased resources are obtained through cooperation with suppliers, such as Huawei Cloud, Alibaba Cloud, and Volcano Cloud. For temporary tasks with tight delivery schedules, these resources are allocated first for flexible processing.

[0052] Tidal resources are defined as machine resources that are idle within a set time period, such as during the night time period, when the number of task submissions is small and resources may be idle; tidal resources come from fixed resources (owned resources) and or flexible call resources (rented resources) that meet the definition of tidal resources. Because it has the characteristic of significantly lower cost price than other time periods, this application can greatly improve the flexibility and efficiency of resource utilization by identifying it and matching it with tasks that are sufficient in time and can be processed in stages. In an embodiment of the present application, tidal resources are the application party of the present application method, that is, the operator itself and the supplier of the leased resources. Within a specific set time period, the current state is an idle machine, and the supplier can schedule it for use by the operator at a low price. It is a resource that first meets the two conditions of "set time" and "idle". Tidal resources can be resources provided by any supplier as long as they meet the conditions, which can reduce the waste of resources. For example, if an agreement is made with a supplier that all machines that are idle during the set time period from 1 a.m. to 6 a.m. can be scheduled as tidal resources. Machines that were idle during the most recent period from 1 a.m. to 6 a.m. are tidal resources that can be scheduled immediately. However, these machines may not be idle during the next set time period, and therefore may not necessarily be tidal resources. Tidal resources are dynamic and change in real time, so that resources can be used more accurately.

[0053] S3. Select flexible requirements from the task requirements; task requirements include fixed requirements, specific requirements and flexible requirements;

[0054] Fixed demand refers to the task demand where users process similar tasks for a long time and the required computing power is relatively fixed;

[0055] Specific requirements refer to requirements that users need to start processing in a timely manner and the task delivery time is tight. The required computing power is relatively fixed. For example, an online video streaming service may need to continuously process and transmit video content, which requires the system to provide stable computing resources to encode and transmit video streams.

[0056] Flexible demand refers to the demand that users have no requirements on the processing time of the demand and can be flexibly processed according to the computing power situation; for example, machine model training, software system testing, etc.

[0057] Use fixed resources to handle fixed demands; use flexible resources to handle specific demands; use tidal resources to handle flexible demands.

[0058] S4, estimate the computing power X of the tidal resource and the computing power Y required to process the flexible demand. If X≥Y, use the tidal resource to process the flexible demand. In step S2, the source of tidal resources is introduced. It can be seen that each supplier and operator can provide tidal resources by themselves, and the amount every day (or every set time) is not necessarily the same. The machine that is currently a tidal resource may not be idle in the next set period and will not belong to the tidal resource. Therefore, it is necessary to first determine the computing power of the current tidal resource, and then match the task accordingly.

[0059] Specifically, they include:

[0060] S41: Split tasks for flexible requirements: Set splitting factors, which can be split according to factors such as time, elements, layers, etc., and split the flexible requirement tasks according to the set splitting factors to obtain multiple task blocks.

[0061] In this step, the demand task splitting can adopt the neural network learning method to train and iterate the model; by learning the processing order of historical tasks, analyzing the characteristics of different elements and the relationship between them, and inferring the dependency between elements according to the task processing methods and effects, and then building a decoupling model, and then after repeated training and iteration, the optimal task splitting model is obtained to split the flexible demand, thereby obtaining multiple task blocks. The advantage of such scientific splitting is that it avoids the incoherence, missing, and inability to coordinate tasks caused by arbitrary task splitting, and is not conducive to resource matching.

[0062] S42: Estimate the computing power that may be consumed by processing the split task blocks, and arrange the task blocks that can be processed in parallel, and set the computing power required for each task block to A n , n is a natural number, then A1 to A n The sum is Y.

[0063] S43: Calculate the computing power of tidal resources: Calculate the total computing power X of tidal resources and the computing power B of each divisible resource in tidal resources by setting the number of idle machines in the time period, the capacity of each machine, and the idle time of each machine. n , n is a natural number.

[0064] S44: Resource and demand matching processing: Task block matching is performed according to the tidal resource computing power. The matching method is:

[0065] When the computing power required for task block processing is A n Less than the total computing power of the tidal resource X or the computing power of any sub-resource B n , increase the number of parallel processing of task blocks, and use the computing power of the tidal resource to process the task block; the best case is that each resource block can process exactly one task block.

[0066] If there are any remaining task blocks that have not been processed, they will be marked and processed first after tidal resources are generated in the next period.

[0067] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the same technology of the present invention, the present invention is also intended to include these modifications and variations.

Claims

1. An efficient resource task matching method, characterized in that: The method comprises: Obtain computing resources and task requirements in real time; Selecting tidal resources from the computing power resources; selecting flexible requirements from among the task requirements; Estimate the computing power X of the tidal resource and the computing power Y required to process the flexible demand, and if X≥Y, use the tidal resource to process the flexible demand; Among them, the tidal resources are defined as: machine resources that are idle within a set time period; the flexible demand is defined as: the user has no requirements on the processing time of the demand, and the demand can be flexibly processed according to the computing power situation.

2. The efficient resource task matching method according to claim 1, characterized in that: The computing resources also include fixed resources and flexible call resources; The fixed resources are self-owned resources; the flexible resources are leased resources; The task requirements also include fixed requirements and specific requirements; Fixed demand refers to the demand for users to process similar tasks for a long time, and the required computing power is relatively fixed; specific demand refers to the demand that users need to start processing in a timely manner and the task delivery time is tight; The tidal resources come from the fixed resources and / or the flexible resources that meet the definition of tidal resources.

3. The efficient resource task matching method according to claim 1, characterized in that: The fixed resources are used to process the fixed requirements; and the flexible resources are used to process the specific requirements.

4. The efficient resource task matching method according to claim 1, characterized in that: The using the tidal resource to process the flexible demand comprises: Split flexible demand tasks: Set splitting factors, split flexible demand tasks according to the set splitting factors, and obtain multiple task blocks.

5. The efficient resource task matching method according to claim 4 is characterized in that: The using the tidal resource to process the flexible demand further comprises: Estimate the computing power that may be consumed by processing the split task blocks, arrange the task blocks that can be processed in parallel, and set the computing power required for each task block to A. n , n is a natural number, then A1 to A n The sum is Y.

6. The efficient resource task matching method according to claim 5, characterized in that: The using the tidal resource to process the flexible demand further comprises: Calculate the computing power of tidal resources: By setting the number of idle machines in a time period, the capacity of each machine, and the idle time of each machine, the total computing power of tidal resources X and the computing power of each divisible resource in the tidal resources B are calculated. n , n is a natural number.

7. The efficient resource task matching method according to claim 6, characterized in that: The using the tidal resource to process the flexible demand further comprises: Resource and demand matching processing: Task blocks are matched according to the tidal resource computing power. The matching method is: When the computing power required for task block processing is A n Less than the total computing power of Tide resources X or the computing power of any sub-resource B n , increase the number of parallel processing of task blocks, and use the computing power of the tidal resource to process the task block; If there are any remaining task blocks that have not been processed, they will be marked and processed first after tidal resources are generated in the next period.

8. An efficient cloud desktop creation and management system, characterized by include: A first acquisition module, the first acquisition module is used to acquire computing resources and task requirements in real time; A first selection module, the first selection module is used to select tidal resources from the computing power resources; A second selection module, the second selection module is used to select flexible requirements from the task requirements; A first estimation module, the first estimation module is used to estimate the computing power X of the tidal resource and the computing power Y required to process the flexible demand, and if X≥Y, use the tidal resource to process the flexible demand; Among them, the tidal resources are defined as: machine resources that are idle within a set time period; the flexible demand is defined as: the user has no requirements on the processing time of the demand, and the demand can be flexibly processed according to the computing power situation.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images