CN117056079A - Evaluation method and device for calculation force demand and readable storage medium - Google Patents
Evaluation method and device for calculation force demand and readable storage medium Download PDFInfo
- Publication number
- CN117056079A CN117056079A CN202311103774.2A CN202311103774A CN117056079A CN 117056079 A CN117056079 A CN 117056079A CN 202311103774 A CN202311103774 A CN 202311103774A CN 117056079 A CN117056079 A CN 117056079A
- Authority
- CN
- China
- Prior art keywords
- demand
- index
- power demand
- calculation
- evaluation value
- 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.)
- Pending
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 137
- 238000011156 evaluation Methods 0.000 title claims abstract description 76
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000013507 mapping Methods 0.000 claims abstract description 31
- 238000013473 artificial intelligence Methods 0.000 claims description 38
- 230000015654 memory Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 description 23
- 238000007689 inspection Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 239000000047 product Substances 0.000 description 4
- 239000011521 glass Substances 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000037221 weight management Effects 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005251 gamma ray Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 239000012466 permeate Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a device for evaluating calculation force demand and a readable storage medium, wherein the method comprises the following steps: creating a mapping of atomic scheme categories for the target business scenario; carrying out service portraits on the atom schemes corresponding to the mapped atom scheme categories; and evaluating the power demand according to indexes of a preset power demand index system according to the service image, and calculating a normalized power demand evaluation value according to index weights. The method, the device and the readable storage medium can solve the problems that in the prior art, calculation force demand estimation is carried out according to the whole industrial user level or park level, the current fine-granularity calculation force resource scheduling and guaranteeing of the industrial field service level is difficult to meet, and the service guaranteeing capability of a calculation network is influenced.
Description
Technical Field
The present invention relates to the technical field of computing power networks, and in particular, to a computing power demand assessment method, a computing power demand assessment device, and a readable storage medium.
Background
With the rapid development of AI (Artificial Intelligence ), efficient computing power becomes a key element for supporting the development of an intelligent society, and starts to permeate through various industries, and a computing power network is gradually accepted by the industry as a novel network architecture and possibly becomes a new direction of network evolution. The computing power network can allocate and flexibly schedule computing resources and storage resources among cloud, edges and ends according to service requirements, so that better service is provided for the service.
Because the computing network service relates to thousands of industries, especially in the industrial field, the industry and the scene are quite different, and the demands of the business scene on computing power are quite different. For example, machine vision business requires computing power resources that meet AI model training; the image/video recognition service needs to have computing power resources meeting the recognition algorithm. However, when providing power service for industrial users, power demand estimation is generally performed according to the whole of the industrial user level or the park level, and power resource demand is estimated in this way, so that it is difficult to satisfy the current fine-granularity power resource scheduling and guaranteeing of the industrial field service level, and the service guaranteeing capability of the computing network is affected.
Disclosure of Invention
The technical problem to be solved by the invention is to provide an evaluation method, an evaluation device and a readable storage medium for computing power requirements aiming at the defects of the prior art, so as to solve the problems that computing power requirements are evaluated according to the whole industrial user level or park level in the prior art, the current fine-granularity computing power resource scheduling and guaranteeing of the industrial field service level is difficult to meet, and the service guaranteeing capability of a computing network is influenced.
In a first aspect, the present invention provides a method of assessing a power demand, the method comprising:
creating a mapping of atomic scheme categories for the target business scenario;
carrying out service portraits on the atom schemes corresponding to the mapped atom scheme categories;
and evaluating the power demand according to indexes of a preset power demand index system according to the service image, and calculating a normalized power demand evaluation value according to index weights.
Further, the creating a mapping of atomic scheme categories for the target service scenario specifically includes:
mapping the target business scenario into one or more atomic scheme categories;
wherein the atomic schema categories include at least one of: the image/video feedback type, the remote control type, the machine vision type, the programmable logic controller PLC control type, the position location type and the Internet of things sensing type.
Further, the business portrayal comprises an algorithm model related to terminal equipment, data processing and application services related to a communication link.
Further, before evaluating the power demand according to the business image and the index according to the preset power demand index system, and calculating the normalized power demand evaluation value according to the index weight, the method further comprises:
the calculation force demand index system is formulated, wherein the calculation force demand index system comprises calculation force demand indexes corresponding to various types of terminal equipment, and the calculation force demand indexes comprise general calculation amount demand indexes, artificial intelligence AI calculation amount demand indexes and storage demand indexes.
Further, before evaluating the power demand according to the business image and the index according to the preset power demand index system, and calculating the normalized power demand evaluation value according to the index weight, the method further comprises:
and configuring weight coefficients corresponding to the calculation force demand indexes.
Further, in the weight coefficients of the image/video backhaul class, the weight coefficient of the general calculation amount demand is larger than the weight coefficient of the storage demand, and the weight coefficient of the storage demand is larger than the weight coefficient of the AI calculation amount demand;
among the weight coefficients of the machine vision class, the weight coefficient of the AI computation demand is larger than the weight coefficient of the general computation demand, and the weight coefficient of the general computation demand is larger than the weight coefficient of the storage demand.
Further, the estimating the power demand according to the business portrait and according to the index of the preset power demand index system, and according to the index weight, calculating the normalized power demand estimating value, specifically including:
according to the service portraits, acquiring a general calculation amount demand index evaluation value, an AI calculation amount demand index evaluation value and a storage demand index evaluation value of each terminal device related to each atomic scheme respectively;
acquiring a weight coefficient corresponding to a general calculation amount demand index, an AI calculation amount demand index and a storage demand index;
and calculating a normalized calculation power demand evaluation value according to the general calculation power demand index evaluation value, the AI calculation power demand index evaluation value, the storage demand index evaluation value and the weight coefficient corresponding to the general calculation power demand index, the AI calculation power demand index and the storage demand index of each terminal device related to each atomic scheme.
In a second aspect, the present invention provides an evaluation device for computing force demand, comprising:
the category mapping module is used for creating mapping of atomic scheme categories for the target service scene;
the business portrayal module is connected with the category mapping module and is used for carrying out business portrayal on the atom scheme corresponding to the mapped atom scheme category;
the computing power demand evaluation module is connected with the business portrait module and is used for evaluating the computing power demand according to the business portrait and the index of a preset computing power demand index system, and calculating a normalized computing power demand evaluation value according to the index weight.
In a third aspect, the present invention provides an assessment device for a demand for computing power, comprising a memory in which a computer program is stored, and a processor arranged to run the computer program to implement the assessment method for a demand for computing power as described in the first aspect above.
In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of assessing computational power demands of the first aspect described above.
The invention provides an evaluation method and device of calculation force demand and a readable storage medium. Firstly, creating mapping of atomic scheme categories for a target service scene; then carrying out service portraits on the atom schemes corresponding to the mapped atom scheme categories; and evaluating the power demand according to the business portraits according to indexes of a preset power demand index system, and calculating a normalized power demand evaluation value according to index weights. The invention is oriented to different business scenes in the vertical industry, and can accurately grasp the calculation force requirements of the different business scenes, thereby being convenient for providing resource scheduling basis for changing the follow-up service of the calculation network into the business scenes and improving the service experience of the calculation network of the industry. The method solves the problems that in the prior art, the calculation force demand is estimated according to the whole industrial user level or park level, the current fine-granularity calculation force resource scheduling and guaranteeing of the industrial field service level is difficult to meet, and the service guaranteeing capability of a calculation network is influenced.
Drawings
FIG. 1 is a flow chart of a method for evaluating a power demand according to embodiment 1 of the present invention;
FIG. 2 is a schematic diagram of an evaluation device for calculating power demand according to embodiment 2 of the present invention;
fig. 3 is a schematic structural diagram of an evaluation device for calculating power demand according to embodiment 3 of the present invention.
Detailed Description
In order to make the technical scheme of the present invention better understood by those skilled in the art, the following detailed description of the embodiments of the present invention will be given with reference to the accompanying drawings.
It is to be understood that the specific embodiments and figures described herein are merely illustrative of the invention, and are not limiting of the invention.
It is to be understood that the various embodiments of the invention and the features of the embodiments may be combined with each other without conflict.
It is to be understood that only the portions relevant to the present invention are shown in the drawings for convenience of description, and the portions irrelevant to the present invention are not shown in the drawings.
It should be understood that each unit and module in the embodiments of the present invention may correspond to only one physical structure, may be formed by a plurality of physical structures, or may be integrated into one physical structure.
It will be appreciated that the terms "first," "second," and the like in embodiments of the present invention are used to distinguish between different objects or to distinguish between different processes on the same object, and are not used to describe a particular order of objects.
It will be appreciated that, without conflict, the functions and steps noted in the flowcharts and block diagrams of the present invention may occur out of the order noted in the figures.
It is to be understood that the flowcharts and block diagrams of the present invention illustrate the architecture, functionality, and operation of possible implementations of systems, apparatuses, devices, methods according to various embodiments of the present invention. Where each block in the flowchart or block diagrams may represent a unit, module, segment, code, or the like, which comprises executable instructions for implementing the specified functions. Moreover, each block or combination of blocks in the block diagrams and flowchart illustrations can be implemented by hardware-based systems that perform the specified functions, or by combinations of hardware and computer instructions.
It should be understood that the units and modules related in the embodiments of the present invention may be implemented by software, or may be implemented by hardware, for example, the units and modules may be located in a processor.
Example 1:
the present embodiment provides a method for evaluating a power demand, as shown in fig. 1, the method includes:
step S101: a map of atomic schema categories is created for the target business scenario.
In this embodiment, the target service scenario is any one of all service scenarios in the industrial field. The target service scene can be equipment point inspection, product quality inspection and the like in combination with specific industries and services.
Optionally, the creating a mapping of atomic scheme categories for the target service scenario specifically includes:
mapping the target business scenario into one or more atomic scheme categories;
wherein the atomic schema categories include at least one of: an image/video backhaul class, a remote control class, a machine vision class, a PLC (Programmable Logic Controller ) control class, a position location class, and an internet of things sensing class.
In this embodiment, a mapping of atomic scheme categories is created for a target business scenario, where the target business scenario may correspond to one or more atomic scheme categories. Atomic scheme categories can be divided into: the image/video feedback type, the remote control type, the machine vision type, the PLC control type, the position locating type, the Internet of things sensing type and the like. Category mapping such as { device click }: image/video backhaul class, thing allies oneself with and senses class }; { product quality inspection: image/video backhaul class, machine vision class, each atom plan class corresponds to a respective atom plan.
Step S102: and carrying out business portraits on the atom schemes corresponding to the mapped atom scheme categories.
In this embodiment, one atomic scheme corresponds to one service image, where the service image includes terminal devices related to a communication link, algorithm models related to data processing and application services (such as a data preprocessing algorithm, a detection algorithm, a positioning algorithm, an identification algorithm, and the like), and may also include the number of devices, a communication period, and the like. One atomic scheme may correspond to a plurality of terminal devices, such as an AGV (Automated Guided Vehicle ) cloud scheduling, and an image/video backhaul scheme may correspond to a plurality of cameras and camera terminals.
Step S103: and evaluating the power demand according to indexes of a preset power demand index system according to the service image, and calculating a normalized power demand evaluation value according to index weights.
In this embodiment, the power demand index system includes power demand indexes corresponding to each type of terminal device, where the power demand indexes include a general calculation amount demand index, an AI calculation amount demand index, and a storage demand index. The category of terminal devices may include, for example, 2D cameras/cameras, 3D cameras, smart capture cards, AR (Augmented Reality), virtual Reality)/VR (Virtual Reality) glasses, PLC/sensors, and the like.
Optionally, before the estimating the power demand according to the business portrait and the index according to the preset power demand index system, and calculating the normalized power demand estimated value according to the index weight, the method further includes:
and formulating the calculation force demand index system.
In this embodiment, a corresponding calculation power demand index system is formulated according to the device information. The calculation power demand index system comprises a general calculation power demand index, an AI calculation power demand index and a storage demand index of each terminal device.
Optionally, before the estimating the power demand according to the business portrait and the index according to the preset power demand index system, and calculating the normalized power demand estimated value according to the index weight, the method further includes:
and configuring weight coefficients corresponding to the calculation force demand indexes.
In this embodiment, a weight coefficient of a general calculation amount demand index, a weight coefficient of an AI calculation amount demand index, and a weight coefficient of a storage demand index are respectively configured; different weight coefficient combinations are adopted for different atomic schemes and terminal equipment.
Optionally, in the weight coefficients of the image/video backhaul class, the weight coefficient of the general computation demand is greater than the weight coefficient of the storage demand, and the weight coefficient of the storage demand is greater than the weight coefficient of the AI computation demand;
among the weight coefficients of the machine vision class, the weight coefficient of the AI computation demand is larger than the weight coefficient of the general computation demand, and the weight coefficient of the general computation demand is larger than the weight coefficient of the storage demand.
In this embodiment, since the image/video backhaul class has a high general calculation amount and a high storage requirement, the weight coefficient is as follows: general calculation amount requirement > storage requirement; because the machine vision class has higher AI calculation amount, general calculation amount and storage requirement, the weight coefficient is as follows: AI computation amount demand > general computation amount demand > storage demand; for the categories such as remote control, PLC control, position location, and Internet of things sensing, the weight coefficients can be consistent because the requirements on calculation amount and storage are smaller.
Optionally, the estimating the power demand according to the business portrait and according to the index of the preset power demand index system, and calculating the normalized power demand estimated value according to the index weight, specifically including:
according to the service portraits, acquiring a general calculation amount demand index evaluation value, an AI calculation amount demand index evaluation value and a storage demand index evaluation value of each terminal device related to each atomic scheme respectively;
acquiring a weight coefficient corresponding to a general calculation amount demand index, an AI calculation amount demand index and a storage demand index;
and calculating a normalized calculation power demand evaluation value according to the general calculation power demand index evaluation value, the AI calculation power demand index evaluation value, the storage demand index evaluation value and the weight coefficient corresponding to the general calculation power demand index, the AI calculation power demand index and the storage demand index of each terminal device related to each atomic scheme.
In this embodiment, the power demand is evaluated according to the power demand index system according to the atomic scheme business portraits, and the normalized power demand evaluation value is calculated according to the index weight:
normalized calculation force demand evaluation value=Σ index weight coefficient index evaluation value
Specifically, for a target service scenario, weight coefficient index evaluation values of all calculation force demand indexes in terminal devices corresponding to all atom schemes are respectively obtained, and then the values obtained by each atom scheme are added to obtain the calculation force demand evaluation value of the target service scenario.
For example, machine vision quality inspection scenes involve machine vision-like atomic schemes, including 1 2D camera, 2 3D cameras. Setting weight coefficients of three indexes, and meeting AI calculation amount requirement > general calculation amount requirement > storage requirement;
AI calculation amount requirement: alpha
General computational demand: beta
Storage requirements: gamma ray
Evaluation of calculation force demand:
AI calculation amount requirement: reqai=f (calculated amount, number of algorithm models)
General computational demand:
2D camera: reqCMP (2D) =f (resolution, frame rate, pixel depth, number of devices)
3D camera: reqCMP (3D) = 2*f (resolution, frame rate, angular resolution, scan frequency, number of devices)
Storage requirements:
2D camera: reqtor (2D) =f (resolution, pixel depth, number of images)
3D camera: reqtor (3D) = 2*f (point cloud file size, number of point clouds)
Normalized computing power demand evaluation value calculation:
REQ＝α*ReqAI
+β*(ReqCMP(2D)+ReqCMP(3D))
+γ*(ReqSTOR(2D)+ReqSTOR(3D))
it should be noted that, only if the calculation power requirements of different service scenarios are precisely mastered, the corresponding calculation power resources can be matched and scheduled according to the calculation power requirements of the industrial field service, so that better calculation power service guarantee is provided for the different service scenarios. The invention can evaluate the calculation power requirements of different service scenes, thereby providing resource scheduling basis for changing the follow-up service of the calculation network into the service scene and improving the calculation network service experience of the industry.
In a specific embodiment, the method for evaluating the demand for computing power is applied to a system for evaluating the demand for computing power, the system comprising: the system comprises a mapping unit, a business portrayal unit, a device management unit, an algorithm model management unit, an index system management unit, an algorithm management unit, an index weight management unit and an algorithm power demand assessment unit, wherein the following parts are described:
1. mapping unit: creating a mapping of atomic scheme categories for each business scenario, wherein one business scenario in the mapping can correspond to one or more atomic scheme categories;
(1) The service scene can be combined with specific industries and services, such as equipment point inspection, product quality inspection and the like;
(2) Atomic scheme categories can be divided into: the image/video feedback class, the remote control class, the machine vision class, the PLC control class, the position location class, the Internet of things sensing class and the like;
(3) Mapping examples: { device spot check: image/video backhaul class, thing allies oneself with and senses class }; { product quality inspection: image/video backhaul class, machine vision class }.
2. Business portrayal unit: and carrying out service portraits on the related atomic schemes according to service scenes. The image information includes: selecting or custom-adding new equipment and equipment quantity from the existing equipment types and models; a communication period; algorithm models (e.g., data preprocessing algorithms, detection algorithms, positioning algorithms, identification algorithms, etc.).
3. Device management unit: managing the types of the common terminal equipment and the parameter information related to the atomic scheme;
4. algorithm model management unit: managing common algorithm model information, including: algorithm model name, parameter (params), calculation (flow);
5. index system management unit: setting a calculation force demand index system, which comprises the following demand indexes: general calculation amount demand, AI calculation amount demand, storage demand;
6. algorithm management unit: according to the equipment information, each index evaluation function of the calculation force requirement is formulated
(1) General computational demand:
(a) 2D camera/camera: f (resolution, frame rate, pixel depth, number of devices)
(b) 3D camera: f (resolution, frame rate, angular resolution, scanning frequency, number of devices)
(c) An intelligent acquisition card: f (Point frequency, bus transmission rate, number of devices)
(d) AR/VR glasses: f (delay, pixel, number of devices)
(e) PLC/sensor: f (periodic data amount, sampling frequency, number of devices)
(2) AI calculation amount requirement: f (calculated amount, number of algorithm models)
(3) Storage requirements:
(a) 2D camera: f (resolution, pixel depth, number of images)
(b) A camera head: f (code rate, video duration)
(c) 3D camera: f (Point cloud File size, point cloud quantity)
(d) An intelligent acquisition card: f (dot frequency, quantity)
(e) AR/VR glasses: f (code rate, duration)
(f) PLC/sensor: f (period data amount, duration, quantity)
7. Index weight management unit: and configuring each index weight coefficient of the computing power demand according to the atomic scheme type and the related equipment category.
(1) The image/video feedback class has higher general calculation amount and storage requirement, and weight coefficient: general calculation amount requirement > storage requirement;
(2) The machine vision class has higher AI calculation amount, general calculation amount and storage requirement, and weight coefficient: AI computation amount demand > general computation amount demand > storage demand;
(3) The remote control type, the PLC control type, the position locating type, the Internet of things sensing type and the like have smaller calculation amount and storage requirements, and the weight coefficients can be consistent.
8. A calculation force demand evaluation unit: according to the atomic scheme image information, the power demand is evaluated according to a power demand index system, and a normalized power demand evaluation value is calculated according to index weight.
Normalized calculation force demand evaluation value=Σ index weight coefficient index evaluation value
The evaluation method of the calculation force demand may include the steps of:
1. creating a mapping of atomic schema categories for each business scenario;
2. carrying out business image according to an atomic scheme, wherein the business image comprises information such as terminal equipment related to a communication link, main algorithms related to data processing and application service and the like;
3. formulating a calculation force demand index system comprising specific indexes and index weights;
4. according to the atomic scheme image information, the power demand is evaluated according to a power demand index system, and a normalized power demand evaluation value is calculated according to index weight.
The method and the system are oriented to different business scenes in the vertical industry, and the computing power requirements of the different business scenes are evaluated, so that a resource scheduling basis is provided for changing the follow-up service of the computing network into the business scene, and the computing network service experience of the industry is improved.
The evaluation method of the calculation force demand provided by the embodiment of the invention comprises the steps of firstly, creating mapping of atomic scheme categories for a target service scene; then carrying out service portraits on the atom schemes corresponding to the mapped atom scheme categories; and evaluating the power demand according to the business portraits according to indexes of a preset power demand index system, and calculating a normalized power demand evaluation value according to index weights. The invention is oriented to different business scenes in the vertical industry, and can accurately grasp the calculation force requirements of the different business scenes, thereby being convenient for providing resource scheduling basis for changing the follow-up service of the calculation network into the business scenes and improving the service experience of the calculation network of the industry. The method solves the problems that in the prior art, the calculation force demand is estimated according to the whole industrial user level or park level, the current fine-granularity calculation force resource scheduling and guaranteeing of the industrial field service level is difficult to meet, and the service guaranteeing capability of a calculation network is influenced.
Example 2:
as shown in fig. 2, the present embodiment provides an evaluation device for calculating a force demand, for executing the above-mentioned evaluation method for calculating a force demand, including:
a category mapping module 11, configured to create a mapping of an atomic scheme category for a target service scenario;
a business portrayal module 12 connected with the category mapping module 11 for carrying out business portrayal on the atom plan corresponding to the mapped atom plan category;
the computing power demand evaluation module 13 is connected with the service portrait module 12 and is used for evaluating the computing power demand according to the predetermined computing power demand index system and index according to the service portrait, and calculating the normalized computing power demand evaluation value according to the index weight.
Optionally, the category mapping module 11 is specifically configured to:
mapping the target business scenario into one or more atomic scheme categories;
wherein the atomic schema categories include at least one of: the image/video feedback type, the remote control type, the machine vision type, the programmable logic controller PLC control type, the position location type and the Internet of things sensing type.
Optionally, the business portraits include terminal equipment involved in the communication link, algorithm models involved in data processing and application services.
Optionally, the apparatus further comprises:
the system comprises an index system making module, a power calculation system setting module and a power calculation module, wherein the power calculation system making module is used for making the power calculation demand index system, the power calculation demand index system comprises power calculation demand indexes corresponding to various types of terminal equipment, and the power calculation demand indexes comprise general calculation amount demand indexes, artificial intelligence AI calculation amount demand indexes and storage demand indexes.
Optionally, the apparatus further comprises:
the weight coefficient configuration module is used for configuring weight coefficients corresponding to the calculation force demand indexes.
Optionally, in the weight coefficients of the image/video backhaul class, the weight coefficient of the general computation demand is greater than the weight coefficient of the storage demand, and the weight coefficient of the storage demand is greater than the weight coefficient of the AI computation demand;
among the weight coefficients of the machine vision class, the weight coefficient of the AI computation demand is larger than the weight coefficient of the general computation demand, and the weight coefficient of the general computation demand is larger than the weight coefficient of the storage demand.
Optionally, the computing power demand assessment module 13 specifically includes:
an index evaluation value calculation unit, configured to obtain, according to the service image, a general calculation amount demand index evaluation value, an AI calculation amount demand index evaluation value, and a storage demand index evaluation value of each terminal device related to each atom plan, respectively;
the weight coefficient acquisition unit is used for acquiring weight coefficients corresponding to the general calculation amount demand index, the AI calculation amount demand index and the storage demand index;
and the normalized calculation power demand unit is used for calculating a normalized calculation power demand evaluation value according to the general calculation power demand index evaluation value, the AI calculation power demand index evaluation value, the storage demand index evaluation value, the general calculation power demand index, the AI calculation power demand index and the weight coefficient corresponding to the storage demand index of each terminal device related to each atomic scheme.
Example 3:
referring to fig. 3, the present embodiment provides an evaluation device of the computational power demand, comprising a memory 21 and a processor 22, the memory 21 storing a computer program, the processor 22 being arranged to run the computer program to perform the evaluation method of the computational power demand in embodiment 1.
The memory 21 is connected to the processor 22, the memory 21 may be a flash memory, a read-only memory, or other memories, and the processor 22 may be a central processing unit or a single chip microcomputer.
Example 4:
the present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the evaluation method of the computational power demand in embodiment 1 described above.
Computer-readable storage media include volatile or nonvolatile, removable or non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, computer program modules or other data. Computer-readable storage media includes, but is not limited to, RAM (Random Access Memory ), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory, charged erasable programmable Read-Only Memory), flash Memory or other Memory technology, CD-ROM (Compact Disc Read-Only Memory), digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer.
In summary, the method, the device and the readable storage medium for evaluating the computing power demand provided by the embodiment of the invention firstly create the mapping of the atomic scheme category for the target service scene; then carrying out service portraits on the atom schemes corresponding to the mapped atom scheme categories; and evaluating the power demand according to the business portraits according to indexes of a preset power demand index system, and calculating a normalized power demand evaluation value according to index weights. The invention is oriented to different business scenes in the vertical industry, and can accurately grasp the calculation force requirements of the different business scenes, thereby being convenient for providing resource scheduling basis for changing the follow-up service of the calculation network into the business scenes and improving the service experience of the calculation network of the industry. The method solves the problems that in the prior art, the calculation force demand is estimated according to the whole industrial user level or park level, the current fine-granularity calculation force resource scheduling and guaranteeing of the industrial field service level is difficult to meet, and the service guaranteeing capability of a calculation network is influenced.
It is to be understood that the above embodiments are merely illustrative of the application of the principles of the present invention, but not in limitation thereof. Various modifications and improvements may be made by those skilled in the art without departing from the spirit and substance of the invention, and are also considered to be within the scope of the invention.
Claims (10)
1. A method of assessing a demand for computing force, the method comprising:
creating a mapping of atomic scheme categories for the target business scenario;
carrying out service portraits on the atom schemes corresponding to the mapped atom scheme categories;
and evaluating the power demand according to indexes of a preset power demand index system according to the service image, and calculating a normalized power demand evaluation value according to index weights.
2. The method according to claim 1, wherein creating a mapping of atomic schema categories for a target business scenario specifically comprises:
mapping the target business scenario into one or more atomic scheme categories;
wherein the atomic schema categories include at least one of: the image/video feedback type, the remote control type, the machine vision type, the programmable logic controller PLC control type, the position location type and the Internet of things sensing type.
3. The method of claim 1, wherein the business representation comprises an algorithmic model of terminal devices involved in communication links, data processing, and application services.
4. The method according to claim 1, wherein before evaluating the power demand according to the traffic image according to the index of the power demand index system established in advance and calculating the normalized power demand evaluation value according to the index weight, the method further comprises:
the calculation force demand index system is formulated, wherein the calculation force demand index system comprises calculation force demand indexes corresponding to various types of terminal equipment, and the calculation force demand indexes comprise general calculation amount demand indexes, artificial intelligence AI calculation amount demand indexes and storage demand indexes.
5. The method of claim 4, wherein before evaluating the power demand according to the traffic image according to the index of the power demand index system established in advance and calculating the normalized power demand evaluation value according to the index weight, the method further comprises:
and configuring weight coefficients corresponding to the calculation force demand indexes.
6. The method of claim 5, wherein the weight coefficient of the general computation demand is greater than the weight coefficient of the storage demand, and the weight coefficient of the storage demand is greater than the weight coefficient of the AI computation demand, among the weight coefficients of the image/video backhaul class;
among the weight coefficients of the machine vision class, the weight coefficient of the AI computation demand is larger than the weight coefficient of the general computation demand, and the weight coefficient of the general computation demand is larger than the weight coefficient of the storage demand.
7. The method according to claim 5, wherein said evaluating the power demand according to the power demand index system index set in advance based on the service image, and calculating the normalized power demand evaluation value according to the index weight, specifically comprises:
according to the service portraits, acquiring a general calculation amount demand index evaluation value, an AI calculation amount demand index evaluation value and a storage demand index evaluation value of each terminal device related to each atomic scheme respectively;
acquiring a weight coefficient corresponding to a general calculation amount demand index, an AI calculation amount demand index and a storage demand index;
and calculating a normalized calculation power demand evaluation value according to the general calculation power demand index evaluation value, the AI calculation power demand index evaluation value, the storage demand index evaluation value and the weight coefficient corresponding to the general calculation power demand index, the AI calculation power demand index and the storage demand index of each terminal device related to each atomic scheme.
8. An evaluation device for calculation force demand, characterized by comprising:
the category mapping module is used for creating mapping of atomic scheme categories for the target service scene;
the business portrayal module is connected with the category mapping module and is used for carrying out business portrayal on the atom scheme corresponding to the mapped atom scheme category;
the computing power demand evaluation module is connected with the business portrait module and is used for evaluating the computing power demand according to the business portrait and the index of a preset computing power demand index system, and calculating a normalized computing power demand evaluation value according to the index weight.
9. An assessment device of a computational power demand, characterized in that it comprises a memory in which a computer program is stored and a processor arranged to run the computer program to implement the assessment method of a computational power demand according to any one of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, implements a method of assessing a demand for computing power according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311103774.2A CN117056079A (en) | 2023-08-29 | 2023-08-29 | Evaluation method and device for calculation force demand and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311103774.2A CN117056079A (en) | 2023-08-29 | 2023-08-29 | Evaluation method and device for calculation force demand and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117056079A true CN117056079A (en) | 2023-11-14 |
Family
ID=88653405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311103774.2A Pending CN117056079A (en) | 2023-08-29 | 2023-08-29 | Evaluation method and device for calculation force demand and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117056079A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117370034A (en) * | 2023-12-07 | 2024-01-09 | 之江实验室 | Evaluation method and device of computing power dispatching system, storage medium and electronic equipment |
-
2023
- 2023-08-29 CN CN202311103774.2A patent/CN117056079A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117370034A (en) * | 2023-12-07 | 2024-01-09 | 之江实验室 | Evaluation method and device of computing power dispatching system, storage medium and electronic equipment |
| CN117370034B (en) * | 2023-12-07 | 2024-02-27 | 之江实验室 | Evaluation method and device of computing power dispatching system, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117056079A (en) | Evaluation method and device for calculation force demand and readable storage medium | |
| CN109300151B (en) | Image processing method and device and electronic equipment | |
| CN112232426B (en) | Training method, device and equipment of target detection model and readable storage medium | |
| CN114764778A (en) | Target detection method, target detection model training method and related equipment | |
| CN113592390B (en) | Storage digital twin method and system based on multi-sensor fusion | |
| CN111898682B (en) | Method and device for correcting new model based on multiple source models and computer equipment | |
| CN112329616B (en) | Target detection method, device, equipment and storage medium | |
| CN112330709A (en) | Foreground image extraction method and device, readable storage medium and terminal equipment | |
| CN112597995B (en) | License plate detection model training method, device, equipment and medium | |
| CN115240168A (en) | Perception result obtaining method and device, computer equipment and storage medium | |
| CN111862040A (en) | Portrait picture quality evaluation method, device, equipment and storage medium | |
| CN109492697B (en) | Picture detection network training method and picture detection network training device | |
| CN112668675B (en) | Image processing method and device, computer equipment and storage medium | |
| CN111985439B (en) | Face detection method, device, equipment and storage medium | |
| CN113808142A (en) | Ground identifier identification method and device and electronic equipment | |
| CN117540822A (en) | Federal type incremental learning method, equipment and storage medium across mobile edge network | |
| CN111428858A (en) | Method and device for determining number of samples, electronic equipment and storage medium | |
| CN114638996A (en) | Model training method, device, equipment and storage medium based on counterstudy | |
| CN111027376A (en) | Method and device for determining event map, electronic equipment and storage medium | |
| CN115827198A (en) | Task allocation method, task allocation system, and computer-readable storage medium | |
| CN111488928B (en) | Method and device for acquiring samples | |
| CN112380313A (en) | Method and device for updating confidence coefficient of high-precision map | |
| CN113807150A (en) | Data processing method, attitude prediction method, data processing device, attitude prediction device, and storage medium | |
| CN116386060B (en) | Automatic water gauge data labeling method, device, equipment and medium | |
| CN110442501A (en) | The method, apparatus and electronic equipment of analog service exception |
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 |