CN114880106A

CN114880106A - Calculation power matching method and device and management platform

Info

Publication number: CN114880106A
Application number: CN202111364591.7A
Authority: CN
Inventors: 吴艳光; 张�杰; 高田
Original assignee: Datang Gaohong Zhilian Technology Chongqing Co ltd
Current assignee: Datang Gaohong Zhilian Technology Chongqing Co ltd
Priority date: 2021-11-17
Filing date: 2021-11-17
Publication date: 2022-08-09

Abstract

The invention provides a calculation power matching method, a calculation power matching device and a management platform, wherein the calculation power matching method comprises the following steps: acquiring a maximum calculation power consumption value of each sensor in a plurality of sensors and a minimum residual calculation power value of each MEC equipment in a plurality of multi-access edge calculation MEC equipment; performing calculation force matching between the sensors and the MEC equipment according to the maximum calculation force consumption value of each sensor and the minimum residual calculation force value of each MEC equipment to obtain calculation force matching results; and adjusting the matching relationship between the sensor and the MEC equipment according to the force calculation matching result. According to the technical scheme, the calculation power matching between the sensors and the MEC equipment is carried out according to the maximum calculation power consumption value of each sensor and the minimum residual calculation power value of each MEC equipment, the calculation power matching can be completed, the matching relation between the sensors and the MEC equipment is adjusted according to the calculation power matching result, the load balancing or fault replacing function without services can be achieved, and the problem of insufficient calculation power or waste of calculation power is avoided.

Description

Calculation power matching method and device and management platform

Technical Field

The invention relates to the technical field of communication, in particular to a calculation power matching method, a calculation power matching device and a management platform.

Background

A vehicular wireless communication technology (C-V2X) Vehicle-road cooperative system of a Cellular network mainly comprises a perception subsystem, a cloud control subsystem and a communication subsystem. As shown in fig. 1, the sensing subsystem is mainly composed of a sensor and Multi-Access Edge Computing (MEC). The sensor mainly comprises a camera, a millimeter wave radar, a laser radar and the like, and is responsible for collecting original data of the traffic participants and sending the original data to the MEC. The MEC completes calculation such as target identification, event detection, perception fusion and the like based on the original data, and sends calculation results to a cloud control platform in a cloud control subsystem and a Road Side Unit (RSU) in a communication subsystem. The cloud control subsystem is responsible for functions of service data aggregation, data analysis, intelligent decision, service control, equipment management and the like. The communication subsystem is responsible for completing the functions of broadcasting information such as safety early warning, information service and the like, reporting and issuing service data and the like. The cloud control subsystem further comprises an MEC management platform used for managing the MEC, and the communication subsystem further comprises an On-Board Unit (OBU) used for receiving information and service data such as safety early warning and information service.

In the existing C-V2X vehicle-road cooperative system, an MEC and a sensor are configured in a 1 to N mode. Each MEC is configured and accessed with a plurality of sensors, and the MEC is planned and set in advance based on empirical data of the computing power of the MEC consumed by various sensors. For different intersections, road sections and different time periods, the MEC calculation power consumed by the same type of sensor has larger difference. Based on the static sensor and MEC proportioning relation, the difference can not be flexibly dealt with, and the situation of insufficient calculation force or waste of calculation force is easy to occur.

Disclosure of Invention

The embodiment of the invention provides a calculation power matching method, a calculation power matching device and a management platform, which are used for solving the problems of insufficient calculation power or calculation power waste in the prior art under the static sensor and MEC matching relationship.

In order to solve the above technical problem, an embodiment of the present invention provides the following technical solutions:

the embodiment of the invention provides a calculation force matching method, which comprises the following steps:

acquiring a maximum calculation power consumption value of each sensor in a plurality of sensors and a minimum residual calculation power value of each MEC equipment in a plurality of multi-access edge calculation MEC equipment;

performing calculation force matching between the sensors and the MEC equipment according to the maximum calculation force consumption value of each sensor and the minimum residual calculation force value of each MEC equipment to obtain a calculation force matching result;

and adjusting the matching relation between the sensor and the MEC equipment according to the force calculation matching result.

Optionally, obtaining a maximum computational power consumption value for each sensor of the plurality of sensors comprises:

periodically acquiring the number of targets identified by each sensor sent by the MEC equipment;

and determining the maximum calculated power consumption value of each sensor according to the maximum number of the recognized targets of each sensor in a plurality of periods and the predetermined corresponding relation between the number of the recognized targets of each sensor and the calculated power consumption value of the corresponding sensor.

Optionally, before determining the maximum computational power consumption value of each of the sensors, the method further comprises:

periodically acquiring a computational power consumption value of each sensor;

and establishing a corresponding relation between the quantity of the targets identified by each sensor and the computational power consumption value of the corresponding sensor.

Optionally, establishing a correspondence between the number of targets identified by each sensor and the computational power consumption value of the corresponding sensor includes:

acquiring a calculation power consumption value of a first sensor;

establishing a corresponding relation between the number of the targets identified by the first sensor and the calculated power consumption value of the first sensor according to the number of the targets identified by the first sensor and the calculated power consumption value of the first sensor in a plurality of periods;

determining a calculation power consumption value of a second sensor in the current period according to the corresponding relation between the number of the targets identified by the first sensor and the calculation power consumption value of the first sensor, the number of the targets identified by the first sensor in the current period, the initial calculation power value of first MEC equipment and the first period calculation power value of the first MEC equipment in the current period;

determining the corresponding relation between the number of the targets identified by the second sensor and the calculated power consumption value of the second sensor according to the number of the targets identified by the second sensor and the calculated power consumption value of the second sensor;

the first period calculation force value of the first MEC equipment refers to an MEC total calculation force consumption value when the first MEC equipment respectively processes the sensor data of the first sensor and the sensor data of the second sensor according to a preset target detection algorithm in the current period;

the initial computational force value of the first MEC device indicates a computational force consumption value of the first MEC device in an unloaded state;

the first MEC device processes sensor data of the first sensor and sensor data of the second sensor, respectively;

said first sensor is one of each of said sensors and said second sensor is another of each of said sensors;

the first MEC equipment is one of each MEC equipment.

Optionally, determining the computational power consumption value of the second sensor in the current period according to the correspondence between the number of the targets identified by the first sensor and the computational power consumption value of the first sensor, the number of the targets identified by the first sensor in the current period, the initial computational power value of the first MEC equipment, and the first period computational power value of the first MEC equipment in the current period includes:

determining the calculated power consumption value of the first sensor in the current period according to the corresponding relation between the number of the targets identified by the first sensor and the calculated power consumption value of the first sensor and the number of the targets identified by the first sensor in the current period;

and subtracting the calculated force consumption value of the first sensor in the current period from the first period calculated force value of the first MEC equipment in the current period, and subtracting the initial calculated force value of the first MEC equipment to obtain the calculated force consumption value of the second sensor in the current period.

Optionally, obtaining the calculated power consumption value of the first sensor comprises:

periodically obtaining a second periodic force value of the first MEC device while the first MEC device is processing sensor data of the first sensor;

determining a calculation force consumption value of the first sensor in the current period according to a second period calculation force value of the first MEC equipment in the current period and an initial calculation force value of the first MEC equipment;

the second period calculation force value of the first MEC device is an MEC total calculation force consumption value when the first MEC device processes the sensor data of the first sensor according to a preset target detection algorithm in the current period.

Optionally, determining the calculation power consumption value of the first sensor in the current period according to the second period calculation power value of the first MEC device in the current period and the initial calculation power value of the first MEC device, includes:

and subtracting the initial force value of the first MEC equipment from the second period force value of the first MEC equipment in the current period to obtain the force consumption value of the first sensor in the current period.

Optionally, establishing a correspondence between the number of the targets identified by the first sensor and the calculated power consumption value of the first sensor according to the number of the targets identified by the first sensor and the calculated power consumption value of the first sensor in a plurality of cycles includes:

determining the binary group of the number of the targets identified by the first sensor and the calculated force consumption value in different periods according to the number of the targets identified by the first sensor in different periods and the corresponding calculated force consumption value;

mapping the binary group of the quantity of the targets identified by the first sensor and the calculated power consumption value in different periods on a coordinate system to obtain a relation curve of the quantity of the targets identified by the first sensor and the calculated power consumption value;

and performing polynomial fitting on the relation curve to obtain a polynomial function of the corresponding relation between the number of the targets identified by the first sensor and the calculated power consumption value.

Optionally, the obtaining a minimum remaining computation force value of each MEC device of the multiple multi-access edge computing MEC devices includes:

periodically acquiring a residual force value of each MEC device;

and determining the minimum residual force value of each MEC device according to the residual force value of each MEC device.

Optionally, periodically acquiring a remaining computation value of each MEC device, including:

periodically acquiring a periodic force calculation value of each MEC device sent by each MEC device;

and determining the residual force calculation value of each MEC device in the current period according to the period force calculation value of each MEC device and the preset force calculation threshold value of each MEC device.

Optionally, determining a remaining computation force value of each MEC device in a current cycle according to the cycle computation force value of each MEC device and a preset computation force threshold of each MEC device, includes:

and subtracting the period calculation force value of the corresponding MEC equipment in the current period from the preset calculation force threshold value of each MEC equipment to obtain the residual calculation force value of the corresponding MEC equipment in the current period.

Optionally, the calculated force value of the MEC plant is related to the first parameter;

wherein the first parameter comprises at least one of:

network bandwidth;

the CPU processing capacity of the central processing unit;

graphics processor GPU processing capability;

the video memory capacity of the GPU;

memory storage capacity;

hard disk storage capacity;

the calculated force value is one of the following:

calculating a force consumption value;

residual force calculation value;

an initial force calculation value;

and (5) periodically calculating a force value.

Optionally, the unit of measure of the network bandwidth is one of:

percent; kilobits per second kbps; megabits per second Mbps; gigabit per second Gbps;

the measurement unit of the CPU processing capacity of the central processing unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the GPU processing capacity measurement unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the measurement unit of the video memory capacity of the GPU is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the memory storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; a terabyte TB;

the measurement unit of the hard disk storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB; beat byte PB.

An embodiment of the present invention further provides a management platform, which includes a processor, a memory, and a program or an instruction stored on the memory and executable on the processor, and when executed by the processor, the program or the instruction implements the steps of the computational power matching method described in any one of the above.

The embodiment of the present invention further provides a calculation force matching apparatus, including:

the acquisition module is used for acquiring the maximum computational power consumption value of each sensor in the plurality of sensors and the minimum residual computational power value of each MEC equipment in the plurality of multi-access edge computational MEC equipment;

the matching module is used for performing calculation force matching between the sensors and the MEC equipment according to the maximum calculation force consumption value of each sensor and the minimum residual calculation force value of each MEC equipment to obtain a calculation force matching result;

and the adjusting module is used for adjusting the matching relationship between the sensor and the MEC equipment according to the force calculation matching result.

Optionally, the obtaining module includes:

the first acquisition unit is used for periodically acquiring the number of the targets identified by each sensor, which are sent by the MEC equipment;

the first determining unit is used for determining the maximum calculated power consumption value of each sensor according to the maximum number of the identified targets of each sensor in a plurality of periods and the predetermined corresponding relation between the number of the identified targets of each sensor and the calculated power consumption value of the corresponding sensor.

Optionally, the obtaining module further includes:

a second acquisition unit for periodically acquiring the calculated power consumption value of each sensor;

and the corresponding relation establishing unit is used for establishing the corresponding relation between the number of the targets identified by each sensor and the calculation power consumption value of the corresponding sensor.

Optionally, the correspondence relationship establishing unit is specifically configured to:

acquiring a calculation power consumption value of a first sensor;

the first MEC equipment is one of each MEC equipment.

periodically acquiring a residual force value of each MEC device;

Optionally, the obtaining module includes:

a third obtaining unit, configured to periodically obtain a periodic force calculation value of each MEC device sent by each MEC device;

and the second determining unit is used for determining the residual calculation force value of each MEC device in the current period according to the period calculation force value of each MEC device and the preset calculation force threshold value of each MEC device.

Optionally, the second determining unit is specifically configured to:

wherein the first parameter comprises at least one of:

network bandwidth;

the CPU processing capacity of the central processing unit;

graphics processor GPU processing power;

the video memory capacity of the GPU;

memory storage capacity;

hard disk storage capacity;

the calculated force value is one of the following:

calculating a force consumption value;

residual force calculation value;

an initial force calculation value;

and periodically calculating a force value.

Optionally, the unit of measure of the network bandwidth is one of:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the GPU processing capacity measurement unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the measurement unit of the video memory capacity of the GPU is one of the following units:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the memory storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the hard disk storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; a terabyte TB; beat byte PB.

Embodiments of the present invention also provide a readable storage medium, on which a program or instructions are stored, which when executed by a processor implement the steps of the computational power matching method as described in any one of the above.

The beneficial effects of the invention are:

according to the scheme, the maximum computational power consumption value of each sensor in a plurality of sensors and the minimum residual computational force value of each MEC device in a plurality of multi-access edge computing MEC devices are obtained, the computational power matching between the sensors and the MEC devices is carried out according to the maximum computational power consumption value of each sensor and the minimum residual computational force value of each MEC device, the computational power matching result is obtained, the computational power matching can be completed, the matching relation between the sensors and the MEC devices is adjusted according to the computational power matching result, the load balancing or fault replacing function without services can be achieved, and the problems of insufficient computational power or computational power waste are avoided.

Drawings

FIG. 1 is a schematic structural diagram of a C-V2X vehicle-road coordination system according to an embodiment of the present invention;

FIG. 2 is a flow chart of a computational power matching method provided by an embodiment of the invention;

FIG. 3 is a second flowchart of the computational power matching method according to the embodiment of the present invention;

fig. 4 is a schematic structural diagram of an MEC management platform according to an embodiment of the present invention;

fig. 5 is a schematic diagram of a computational force matching apparatus provided in an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.

First, some concepts related to the embodiments of the present invention will be explained.

In the C-V2X vehicle-road cooperative system, by a method for dynamically evaluating the processing capability of an MEC and a sensor (including Central Processing Unit (CPU), Graphics Processing Unit (GPU), GPU display processing capability, memory processing capability and other indexes, hereinafter referred to as computing power), the service load capability of the MEC and the computing power demand of the sensor are evaluated more accurately, and a basis is provided for service strategies such as service load balancing, service fault taking over, service computing power prediction and the like.

The invention provides a calculation power matching method, a calculation power matching device and a management platform, aiming at the problems of insufficient calculation power or calculation power waste in the static sensor and MEC proportioning relation in the prior art.

As shown in fig. 2, an embodiment of the present invention provides a calculation force matching method, including:

step 201: acquiring a maximum calculation power consumption value of each sensor in a plurality of sensors and a minimum residual calculation power value of each MEC equipment in a plurality of multi-access edge calculation MEC equipment;

it should be noted that the calculation force matching method provided by the embodiment of the present invention is executed by an MEC management platform in the C-V2X vehicle-road coordination system.

The plurality of sensors in the embodiment of the present invention means two or more sensors.

In the embodiment of the invention, a maximum calculation power consumption value list of all the sensors is formed according to the obtained maximum calculation power consumption value of each sensor in the plurality of sensors, and the maximum calculation power consumption value list comprises items such as an Identity Document (ID) of each sensor, an MEC (Identity document) to which each sensor belongs, a calculation power consumption value of each sensor and the like.

And forming a minimum residual force value list of all MEC equipment according to the obtained minimum residual force value of each MEC equipment in the MEC equipment, wherein the minimum residual force value list comprises the MEC ID, the MEC residual force value and other items of each MEC equipment.

Step 202: and performing calculation force matching between the sensors and the MEC equipment according to the maximum calculation force consumption value of each sensor and the minimum residual calculation force value of each MEC equipment to obtain a calculation force matching result.

That is, the MEC management platform may complete computation matching according to the sensor granularity according to the maximum computation value consumption list and the minimum residual computation value list, so as to provide a basis for subsequently adjusting the matching relationship between the sensor and the MEC device.

Step 203: and adjusting the matching relationship between the sensor and the MEC equipment according to the force calculation matching result.

It should be noted that, according to the calculation matching result, adjustment of the matching relationship between different sensors and different MEC devices can be achieved, and further, functions such as service load balancing or service fault alternation can be achieved. Illustratively, the maximum computation power consumption value list includes sensor IDs and maximum computation power consumption values of 4 sensors (sensor 1, sensor 2, sensor 3, sensor 4, respectively), and the minimum residual computation power value list includes MEC IDs and minimum residual computation power values of 3 MEC devices (MEC 1, MEC 2, and MEC 3), wherein the maximum computation power consumption value of sensor 1 is 5%, the maximum computation power consumption value of sensor 2 is 20%, the maximum computation power consumption value of sensor 3 is 10%, the maximum computation power consumption value of sensor 4 is 40%, the minimum residual computation power value of MEC1 is 55%, the minimum residual computation power value of MEC 2 is 50%, and the minimum residual computation power value of MEC 3 is 75%, and before performing computation power matching, if sensor 1 matches with MEC1, sensor 2 and sensor 3 match with MEC 2, sensor 4 matches with MEC 3, in order to realize business balance, according to the calculation matching result, the matching relationship between the sensor and the MEC equipment is adjusted as follows: sensor 2 is matched with MEC1, sensor 1 and sensor 3 are respectively matched with MEC 2, and sensor 4 is matched with MEC 3; if MEC1 fails, then according to the calculation matching result, sensor 1 matched with MEC1 may be adjusted to be matched with MEC 2, that is, the matching relationship between the sensor and the MEC equipment is adjusted as follows: the sensor 1, the sensor 2 and the sensor 3 are respectively matched with the MEC 2, and the sensor 4 is matched with the MEC 3, so that the service fault alternation function can be realized.

As a preferred embodiment, acquiring the maximum calculated power consumption value of each of the plurality of sensors includes:

In the embodiment of the invention, the MEC management platform acquires and records the number of the targets identified by each sensor sent by the MEC equipment in a plurality of periods to form the historical number of the identified targets, selects the maximum number of the targets identified by each sensor based on the historical number of the identified targets corresponding to each sensor in a plurality of periods, and determining the maximum computational power consumption value of each sensor according to the predetermined corresponding relationship between the number of the targets recognized by each sensor and the computational power consumption value of the corresponding sensor, specifically, the corresponding relationship between the number of the targets recognized by each sensor and the computational power consumption value of the corresponding sensor is a polynomial function, and the maximum number of the targets recognized by each sensor is substituted into the corresponding polynomial function, the maximum computational power consumption value of each sensor can be determined, and a maximum computational power consumption value list is formed.

It should be further noted that the number of targets identified by each sensor is the MEC device matched with the sensor, and is obtained and counted according to the sensor data sent by the sensor in the current period according to the preset period T1, and each MEC device reports the counted number of targets identified by each sensor to the MEC management platform according to the period T1.

Further, before determining the maximum computational power consumption value for each of the sensors, the method further comprises:

periodically acquiring a computational power consumption value of each sensor;

That is, the MEC management platform obtains and records the maximum computational power consumption value of each sensor in multiple periods, determines the number of the recognition targets and the computational power consumption value corresponding to the same sensor and in the same period, and establishes the corresponding relationship between the number of the recognition targets of the sensor and the computational power consumption value of the corresponding sensor according to the number of the corresponding recognition targets and the computational power consumption value, wherein the corresponding relationship is a polynomial function: and P ═ f (n), wherein P is the calculated power consumption value of the sensor, and n is the number of the targets identified by the sensor.

Through the polynomial function, when the number of the targets identified by the sensor is an arbitrary value, the computational power consumption value of the corresponding sensor can be calculated.

acquiring a calculation power consumption value of a first sensor;

the first MEC device is one of each of the MEC devices.

The process of establishing the correspondence between the number of objects recognized by each sensor and the calculated power consumption value of the corresponding sensor is described in detail below.

In the case where the first MEC apparatus (EMC 1) processes sensor data transmitted from the first sensor (sensor 1) and sensor data transmitted from the second sensor (sensor 2), respectively, the calculated power consumption value of the sensor 1 is acquired in advance, and according to the above-mentioned step of establishing the correspondent relationship between quantity of targets recognized by every sensor and calculated power consumption value of correspondent sensor a polynomial function between quantity of targets recognized by sensor 1 and calculated power consumption value of sensor 1 is established, namely, the number of the objects recognized by the sensor 1 in a plurality of periods is obtained, the number of the recognized objects and the calculated power consumption value which correspond to the sensor 1 and are in the same period are determined, and a polynomial function between the number of the objects recognized by the sensor 1 and the calculated power consumption value of the corresponding sensor is established according to the number of the plurality of corresponding recognition objects and the calculated power consumption value. A step of obtaining a first periodic force calculation value reported by the EMC 1 according to a period (the first periodic force calculation value is an MEC total force consumption value when the MEC1 respectively processes sensor data sent by the sensor 1 and the sensor 2 according to a preset target detection algorithm in one period), obtaining an initial force calculation value of the MEC1 reported by the MEC1, determining the force calculation consumption value of the sensor 2 in a corresponding period according to a polynomial function between the number of targets identified by the sensor 1 and the force calculation consumption value of the corresponding sensor, the number of targets identified by the sensor 1 in the corresponding period, the initial force calculation value of the MEC1 and the first periodic force calculation value of the MEC1, calculating the force calculation consumption value of the sensor 2 in each period, then according to the obtained number of targets identified by the sensor 2, and according to the step of establishing a corresponding relation between the number of targets identified by each sensor and the force calculation consumption value of the corresponding sensor, the method comprises the steps of establishing a polynomial function between the number of targets identified by the sensor 2 and the calculated power consumption value of the sensor 2, namely acquiring the number of targets identified by the sensor 2 in a plurality of periods, determining the number of the identified targets and the calculated power consumption value of the sensor 2 in the same period, and establishing the polynomial function between the number of the targets identified by the sensor 2 and the calculated power consumption value of the corresponding sensor according to the number of the plurality of corresponding identified targets and the calculated power consumption value. Wherein the initial computation value of MEC1 is the computation power consumption value of MEC1 in the unloaded state, and the initial computation value is the same in each cycle.

The type of device of the sensor 1 and the type of device of the sensor 2 may be the same or different.

Specifically, the process of EMC 1 calculating the force value according to the first cycle reported periodically is as follows: the MEC1 processes sensor data sent by the sensor 1 and the sensor 2 at the same time, the MEC1 reports a first periodic force calculation value according to a period, and the first periodic force calculation value comprises a force calculation consumption value of the sensor 1 and a force calculation consumption value of the sensor 2 in the current period.

By repeating the above steps, the corresponding relation between the number of the targets recognized by each sensor and the calculation power consumption value of the corresponding sensor can be established.

Further, determining the calculation power consumption value of the second sensor in the current period according to the corresponding relationship between the number of the targets identified by the first sensor and the calculation power consumption value of the first sensor, the number of the targets identified by the first sensor in the current period, the initial calculation power value of the first MEC device, and the first period calculation power value of the first MEC device in the current period includes:

That is, according to the polynomial function between the number of objects recognized by the sensor 1 and the calculated power consumption value of the corresponding sensor, the number of objects recognized by the sensor 1 for the corresponding cycle, the initial calculated power value of the MEC1, and the first cycle calculated power value of the MEC1, the process of determining the calculated power consumption value of the sensor 2 for the corresponding cycle is:

substituting the number of the targets identified by the sensor 1 corresponding to each period into a polynomial function between the number of the targets identified by the sensor 1 and the calculated power consumption value of the sensor 1 to obtain the calculated power consumption value of the sensor 1 corresponding to each period, and according to a formula: the calculated power consumption value of the sensor 2 in the current period is equal to the first period calculated power value of the current period MEC1, the calculated power consumption value of the sensor 1 in the current period, and the initial calculated power value of the MEC1, and the calculated power consumption value of the sensor 2 in each period is calculated.

As a preferred embodiment, the obtaining of the calculated power consumption value of the first sensor comprises:

The process of acquiring the calculated power consumption value of the sensor 1 will be described in detail below.

Under the condition that the MEC1 only processes the sensor data of the sensor 1, receiving a second period calculation force value reported by the MEC1 according to a period (the first period calculation force value is within one period, and the MEC1 respectively processes the sensor data sent by the sensor 1 according to a preset target detection algorithm to obtain an MEC total calculation force consumption value), and determining the calculation force consumption value of the sensor 1 of the corresponding period according to the second period calculation force value of the corresponding period and the initial calculation force value of the MEC 1.

Further, determining the calculation power consumption value of the first sensor in the current period according to the second period calculation power value of the first MEC device in the current period and the initial calculation power value of the first MEC device, including:

That is, according to the second period calculation force value of the corresponding period and the initial calculation force value of the MEC1, the process of determining the calculation force consumption value of the sensor 1 of the corresponding period is as follows:

according to the formula: the calculated power consumption value of the sensor 1 in the current cycle is equal to the second cycle calculated power value of the current cycle MEC1 — the initial calculated power value of the MEC1, and the calculated power consumption value of the sensor 1 in each cycle is calculated.

As a preferred embodiment, establishing a correspondence between the number of objects identified by the first sensor and the calculated power consumption value of the first sensor according to the number of objects identified by the first sensor and the calculated power consumption value of the first sensor in a plurality of cycles includes:

Specifically, the process of the MEC management platform establishing the functional relationship between the number of the targets identified by the sensor 1 and the MEC1 computational power consumption value is as follows:

according to the number of the targets recognized by the sensors 1 in a plurality of corresponding periods and the corresponding calculated power consumption values, forming the number of the targets recognized by the sensors 1 in different periods and the number of the calculated power consumption values, namely the number of the recognition targets and the calculated power consumption value, mapping the number of the two-dimensional groups with different values, namely the number of the recognition targets and the calculated power consumption value, on a cross coordinate, wherein the abscissa is the number of the recognition targets, and the ordinate is the calculated power consumption value, so that a relation curve reflecting the number of the targets recognized by the sensors 1 and the calculated power consumption value can be obtained, and the relation curve is fitted to form a polynomial function.

As a preferred embodiment, the obtaining a minimum remaining computation force value of each MEC device of the multiple multi-access edge computing MEC devices includes:

periodically acquiring a residual force value of each MEC device;

In the embodiment of the invention, the MEC management platform selects the minimum value of the residual force value of each MEC device based on the calculated residual force value of each MEC device in each period, namely the minimum residual force value of each MEC device, and forms a minimum residual force value list.

Further, periodically obtaining a remaining computation value for each of the MEC devices, including:

And the period calculation force value of each MEC device is the total calculation force consumption value of each MEC device when the MEC device processes the sensor data of the corresponding sensor according to a preset target detection algorithm in the current period.

The process of periodically acquiring the remaining computation force value of each MEC apparatus will be described below, taking the first MEC apparatus (MEC 1) as an example.

Under the condition that the MEC1 only processes the sensor data of the sensor 1, the sensor 1 sends the sensor data to the MEC1, the MEC1 processes the sensor data of the sensor 1, a target algorithm is started, the calculation force changes, the MEC1 records a periodic calculation force value (a first periodic calculation force value) according to a period, and the MEC1 reports the periodic calculation force value according to the period; under the condition that the MEC1 simultaneously processes the sensor data of the sensor 1 and the sensor 2, the sensor 1 and the sensor 2 simultaneously send a sensor data value MEC1, the MEC1 simultaneously processes the sensor data of the sensor 1 and the sensor 2, a target algorithm is started, the calculation force changes, the MEC1 records a periodic calculation force value (a second periodic calculation force value) according to a period, and the MEC1 reports the periodic calculation force value according to the period.

Because the computing power value of the MEC device is related to a plurality of computing power indexes such as network bandwidth, Central Processing Unit (CPU) processing capacity, Graphics Processing Unit (GPU) display memory capacity, hard disk memory capacity and the like, the MEC management platform sets a computing power threshold value for the MEC1 based on a system operation strategy, wherein the computing power indexes such as the network bandwidth, the Central Processing Unit (CPU) processing capacity, the Graphics Processing Unit (GPU) display memory capacity, the memory capacity and the hard disk memory capacity are respectively set, the computing power threshold value is an upper limit of computing power utilization rate for ensuring stable operation of the system, exemplarily, the computing power threshold value for setting the CPU processing capacity is 85%, namely the utilization rate of the CPU of the MEC1 cannot exceed 85%, the exceeding part is system redundancy computing power, and cannot be distributed and used by the MEC management platform. And the MEC management platform calculates the residual force value of the MEC1 according to the period force value of the MEC1 and the preset force threshold value of the MEC 1.

And repeating the steps, and acquiring the residual force value of each MEC device according to the period.

Further, determining a remaining computation force value of each MEC device in a current cycle according to a cycle computation force value of each MEC device and a preset computation force threshold of each MEC device, including:

That is, the specific process of obtaining the remaining computation force value of the MEC1 by the MEC management platform according to the periodic computation force value of the MEC1 and the preset computation force threshold of the MEC1 by periodic computation is as follows:

according to the formula: the remaining computation force value of the current cycle MEC1 is equal to the computation force threshold of MEC1 — the cycle computation force value of the current cycle MEC1, and the remaining computation force value of MEC1 for each cycle is calculated.

wherein the first parameter comprises at least one of:

network bandwidth;

the CPU processing capacity of the central processing unit;

graphics processor GPU processing power;

the video memory capacity of the GPU;

memory storage capacity;

hard disk storage capacity;

the calculated force value is one of the following:

calculating a force consumption value;

residual force calculation value;

an initial force calculation value;

and (5) periodically calculating a force value.

It should be noted that the MEC comprehensive computation capability is related to parameters such as network bandwidth, CPU processing capability of the central processing unit, GPU processing capability of the graphics processor, video memory capacity, memory storage capacity, and hard disk storage capacity of the GPU of the graphics processor. Wherein, the comprehensive force calculation value comprises a force calculation consumption value, a residual force calculation value, an initial force calculation value and a periodic force calculation value.

Optionally, the unit of measure of the network bandwidth is one of:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the GPU processing capacity measurement unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the memory storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the hard disk storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB; beat byte PB.

Illustratively, the initial computation force value of each MEC is an MEC initial computation force value table generated by the MEC management platform collecting the initial computation force values reported by each MEC. The MEC initial calculation value table comprises indexes such as MEC ID, CPU occupancy rate, GPU video memory occupancy rate and memory occupancy rate. The MEC initial force value table is shown in Table 1 below.

TABLE 1 MEC initial calculation force value Table

It should be noted that, in table 1 above, the measurement units of CPU occupancy, GPU video memory occupancy, and memory occupancy are all percentages, and may also be corresponding to other measurement units.

The maximum computation value consumption list also comprises indexes such as sensor ID, the affiliated MEC ID, CPU occupancy rate, GPU video memory occupancy rate, memory occupancy rate and the like. The maximum computational power consumption values are tabulated below in table 2.

TABLE 2 maximum calculation power consumption value List

It should be noted that, the measurement units of the CPU consumption value, the GPU video memory consumption value, and the memory consumption value in the above table 2 are all percentages, and may also be corresponding to other measurement units.

The minimum residual computing force value list also comprises indexes such as MEC ID, CPU occupancy rate, GPU video memory occupancy rate, memory occupancy rate and the like. The list of minimum remaining force values is shown in table 3 below.

TABLE 3 minimum remaining force value List

It should be noted that, the measurement units of the CPU remaining computation force value, the GPU video memory remaining computation force value, and the memory remaining computation force value in the above table 3 are all percentages, and may also be corresponding to other measurement units.

The calculation force matching method provided by the embodiment of the invention is specifically described below with reference to fig. 3.

Step 1: MEC1 collects various computational force values that reflect its own computational power. Step 2: and the MEC1 reports the collected self-calculated force value. And 3, step 3: and generating an MEC initial calculation force set according to the self calculation force value collected by each MEC. The MEC initial force value set comprises: the calculation force value set comprises a plurality of calculation force value indexes such as CPU occupancy rate, GPU video memory occupancy rate and memory occupancy rate, and an MEC initial calculation force value set, and is used for reflecting the calculation force load condition of the current EMC. And 4, step 4: the sensor 1 transmits raw data (sensor data) to the MEC 1. And 5: the MEC1 processes the original data sent by the sensor 1, starts a target detection algorithm, and changes the computational power. The MEC1 records its own set of calculated forces (first periodic calculated force values) in period T1. Step 6: and the MEC1 reports the self-computing power set recorded by the MEC1 to the MEC management platform according to a period T1. And 7: the MEC1 counts the number of the targets identified by the sensor 1 in the current period according to the period T1; the more the number of the identified targets is, the more the calculation power consumed by the algorithm is, and the higher the value of each index item of the calculation power set of the current MEC is. And 8: the MEC1 reports the number of the targets identified by the sensor 1 to the MEC management platform according to a period T1. And step 9: the MEC management platform calculates a set of computing power consumption data values (hereinafter referred to as computing power consumption set) of the sensor 1 by cycles, that is, calculates a computing power demand set of the sensor 1. The calculation power consumption value set of the sensor comprises calculation power consumption values of multiple dimensions such as CPU occupancy rate, GPU video memory occupancy rate, memory occupancy rate and the like. Each type of computational power consumption value needs to be computed separately. And respectively obtaining a CPU (Central processing Unit) computational power consumption value, a GPU (graphics processing Unit) computational power consumption value, a GPU display computational power consumption value and a memory computational power consumption value according to the following computational formulas, and finally obtaining a computational power consumption set of the sensor 1. The current period sensor calculated force consumption value is the first period calculated force value of the current period, namely the initial calculated force value of MEC 1. Step 10: on the MEC management platform, according to the calculation power consumption set of the sensor 1 calculated in step 9 and the number of the targets identified by the sensor 1 corresponding to the calculation power consumption period (reported in step 8), a binary group < the number of the identified targets, the consumption calculation power value > is formed. Mapping the binary group of different values, namely the number of the recognition targets and the consumption calculation force value, in a cross coordinate system, wherein the horizontal axis represents the number of the recognition targets, and the vertical axis represents the consumption calculation force value, so that a curve reflecting the consumption calculation force value (defined as P) and the number of the recognition targets (defined as n) can be obtained. A polynomial is fitted to the curve to form a P ═ f (n) polynomial function. The consumption calculation force value of the corresponding sensor 1 can be calculated by the polynomial function when the number of the recognition targets is any value. Step 11: the sensor 1 and the sensor 2 simultaneously send original data (sensor data) to the MEC1, at this time, the MEC1 calculates a force value (a second period force value) according to a period reported by the period, the second period force value simultaneously comprises a force calculation consumption value of the sensor 1 and a force calculation consumption value of the sensor 2, and the MEC management platform completes calculation of the force calculation consumption value of the sensor 2 and calculation of a force calculation consumption set according to the period. The calculation formula is as follows: the calculated power consumption value of the sensor 2 in the current cycle is equal to the second cycle calculated power value of the current cycle-the calculated power consumption value of the sensor 1 in the current cycle-the initial calculated power value of the MEC 1. The calculation power consumption value of the sensor 1 in the current period is obtained by the MEC management platform according to the number of the identified targets of the sensor 1 in the current period and the polynomial function of P ═ f (n). Step 12: the MEC management platform sets a computing power threshold value for the MEC1 based on a system operation strategy, wherein indexes such as CPU occupancy rate, GPU video memory occupancy rate and memory occupancy rate are set respectively. The calculation force threshold value is a set calculation force utilization rate upper limit for ensuring the stable operation of the system. The MEC management platform calculates a set of MEC remaining computation force values (MEC remaining computation force set) by cycles. The MEC residual computational power set comprises a plurality of computational power value indexes such as CPU occupancy rate, GPU video memory occupancy rate, memory occupancy rate and the like. And each calculation power index needs to be calculated respectively, and finally the MEC residual calculation power set is obtained. And the MEC residual force value is equal to the MEC force threshold value-the period force value of the current period. Step 13: adding the MEC 2 into the MEC management platform, and repeating the steps 4 to 12 by the MEC management platform to complete the calculation of each sensor calculation power consumption set and the calculation of the residual calculation power set of the MEC 2. Step 14: and the MEC management platform selects the maximum number of the recognition targets of each sensor based on the historical data of the number of the recognition targets of the sensors, calculates and selects the maximum computational power consumption value of each sensor by utilizing a P ═ f (n) polynomial function, and generates a maximum computational power consumption value list of all the sensors. Step 15: and the MEC management platform selects the minimum value of the residual computing power of each MEC based on the MEC residual computing power value calculated in each period as the minimum residual computing power value of the MEC, and finally generates a minimum residual computing power value list of all MECs. Step 16: and the MEC management platform completes calculation force matching according to the sensor granularity according to the maximum calculation force consumption value of each sensor and the minimum residual calculation force value of each MEC, and adjusts the matching relation between the sensors and the MECs. And realizing the functions of service load balancing, service fault succession and the like according to different adjustment strategies.

According to the calculation force matching method provided by the embodiment of the invention, the configuration relation between the sensor and the MEC can be more finely adjusted by quantitatively calculating the calculation force consumption value of the sensor and the residual calculation force value of the MEC, so that the service load balance and the service fault smooth succession are realized, and the calculation force prediction can be provided for the new online service.

As shown in fig. 4, an embodiment of the present invention further provides a management platform, which is an MEC management platform, and includes a processor 400, a transceiver 410, a memory 420, and a program stored in the memory 420 and executable on the processor 400; the transceiver 410 is connected to the processor 400 and the memory 420 through a bus interface, wherein the processor 400 performs the following processes for reading the program in the memory:

and adjusting the matching relationship between the sensor and the MEC equipment according to the force calculation matching result.

Optionally, the processor 400 is specifically configured to:

Optionally, the processor 400 is specifically further configured to:

periodically acquiring a computational power consumption value of each sensor;

Optionally, the processor 400 is specifically configured to:

acquiring a calculation power consumption value of a first sensor;

the first MEC device is one of each of the MEC devices.

Optionally, the processor 400 is specifically configured to:

periodically acquiring a residual force value of each MEC device;

Optionally, the processor 400 is specifically configured to:

wherein the first parameter comprises at least one of:

network bandwidth;

the CPU processing capacity of the central processing unit;

graphics processor GPU processing power;

the video memory capacity of the GPU;

memory storage capacity;

hard disk storage capacity;

the calculated force value is one of the following:

calculating a force consumption value;

residual force calculation value;

an initial force calculation value;

and (5) periodically calculating a force value.

Optionally, the unit of measure of the network bandwidth is one of:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the GPU processing capacity measurement unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the memory storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the hard disk storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB; beat byte PB.

Where in fig. 4, the bus architecture may include any number of interconnected buses and bridges, with various circuits of one or more processors, represented by processor 400, and memory, represented by memory 420, being linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 410 may be a number of elements including a transmitter and a receiver that provide a means for communicating with various other apparatus over transmission media including wireless channels, wired channels, fiber optic cables, and the like. For different user devices, the user interface 430 may also be an interface capable of interfacing with a desired device externally, including but not limited to a keypad, display, speaker, microphone, joystick, etc.

The processor 400 may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or a Complex Programmable Logic Device (CPLD), and may also have a multi-core architecture.

As shown in fig. 5, an embodiment of the present invention further provides a computational power matching apparatus, including:

an obtaining module 501, configured to obtain a maximum computation power consumption value of each sensor in the multiple sensors, and a minimum remaining computation power value of each MEC device in the multiple multi-access edge computation MEC devices;

a matching module 502, configured to perform computation matching between the sensors and the MEC equipment according to the maximum computation consumption value of each sensor and the minimum remaining computation value of each MEC equipment, so as to obtain a computation matching result;

and an adjusting module 503, configured to adjust a matching relationship between the sensor and the MEC device according to the force calculation matching result.

According to the embodiment of the invention, the maximum computational power consumption value of each sensor in a plurality of sensors and the minimum residual computational power value of each MEC device in a plurality of multi-access edge computing MEC devices are obtained, the computational power matching between the sensors and the MEC devices is carried out according to the maximum computational power consumption value of each sensor and the minimum residual computational power value of each MEC device, the computational power matching result is obtained, the computational power matching can be completed, and the matching relationship between the sensors and the MEC devices is adjusted according to the computational power matching result, so that the load balancing or fault replacing function without services can be realized, and the problems of insufficient computational power or computational power waste are further avoided.

Optionally, the obtaining module 501 includes:

Optionally, the obtaining module 501 further includes:

acquiring a calculation power consumption value of a first sensor;

determining the corresponding relation between the quantity of the targets identified by the second sensor and the calculated power consumption value of the second sensor according to the quantity of the targets identified by the second sensor and the calculated power consumption value of the second sensor;

said first sensor is one of each of said sensors and said second sensor is the other of each of said sensors;

the first MEC device is one of each of the MEC devices.

periodically acquiring a residual force value of each MEC device;

Optionally, the obtaining module 501 includes:

Optionally, the second determining unit is specifically configured to:

wherein the first parameter comprises at least one of:

network bandwidth;

the CPU processing capacity of the central processing unit;

graphics processor GPU processing power;

the video memory capacity of the GPU;

memory storage capacity;

hard disk storage capacity;

the calculated force value is one of the following:

calculating a force consumption value;

residual force calculation value;

an initial force calculation value;

and (5) periodically calculating a force value.

Optionally, the unit of measure of the network bandwidth is one of:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the GPU processing capacity measurement unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

percent; kilobytes KB; megabyte MB; gigabyte GB; a terabyte TB;

the measurement unit of the memory storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the hard disk storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB; beat byte PB.

It should be noted that the computation power matching apparatus provided in the embodiment of the present invention is an apparatus capable of executing the computation power matching method, and all embodiments of the computation power matching method described above are applicable to the apparatus and can achieve the same or similar technical effects.

The readable storage medium may be any available medium or data storage device that can be accessed by a processor, including but not limited to magnetic memory (e.g., floppy disks, hard disks, magnetic tape, magneto-optical disks (MOs), etc.), optical memory (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor memory (e.g., ROMs, EPROMs, EEPROMs, nonvolatile memory (NAND FLASH), Solid State Disks (SSDs)), etc.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These processor-executable instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While the preferred embodiments of the present invention have been described, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the following claims.

Claims

1. A method of computational force matching, comprising:

2. The algorithm matching method of claim 1, wherein obtaining a maximum algorithm consumption value for each of a plurality of sensors comprises:

3. The algorithm force matching method of claim 2, wherein prior to determining a maximum algorithm force consumption value for each of the sensors, the method further comprises:

periodically acquiring a computational power consumption value of each sensor;

4. The calculation power matching method according to claim 3, wherein establishing a correspondence between the number of objects recognized by each of the sensors and the calculation power consumption value of the corresponding sensor comprises:

acquiring a calculation power consumption value of a first sensor;

the first MEC device is one of each of the MEC devices.

5. The calculation power matching method according to claim 4, wherein determining the calculation power consumption value of the second sensor in the current period according to the correspondence between the number of the targets identified by the first sensor and the calculation power consumption value of the first sensor, the number of the targets identified by the first sensor in the current period, the initial calculation power value of the first MEC equipment, and the first period calculation power value of the first MEC equipment in the current period comprises:

6. The computing power matching method of claim 4, wherein obtaining the computing power consumption value of the first sensor comprises:

7. The algorithm matching method of claim 6, wherein determining the algorithm consumption value of the first sensor during the current cycle based on the second cycle algorithm value of the first MEC device and the initial algorithm value of the first MEC device during the current cycle comprises:

8. The calculation power matching method according to claim 4, wherein establishing a correspondence between the number of the objects identified by the first sensor and the calculation power consumption value of the first sensor according to the number of the objects identified by the first sensor and the calculation power consumption value of the first sensor in a plurality of cycles comprises:

9. The computation force matching method of claim 1, wherein obtaining a minimum remaining computation force value for each of a plurality of multi-access edge computing MEC devices comprises:

periodically acquiring a residual force value of each MEC device;

10. The algorithm force matching method of claim 9, wherein periodically obtaining the remaining algorithm force value of each MEC device comprises:

11. The computation force matching method according to claim 10, wherein determining a remaining computation force value of each MEC device in a current cycle according to a cycle computation force value of each MEC device and a preset computation force threshold of each MEC device comprises:

12. The computing force matching method of claim 1, wherein the computing force value of the MEC apparatus is related to a first parameter;

wherein the first parameter comprises at least one of:

network bandwidth;

the CPU processing capacity of the central processing unit;

graphics processor GPU processing power;

the video memory capacity of the GPU;

memory storage capacity;

hard disk storage capacity;

the calculated force value is one of the following:

calculating a force consumption value;

residual force calculation value;

an initial force calculation value;

and (5) periodically calculating a force value.

13. The computationally intensive matching method of claim 12, wherein the network bandwidth is measured in one of:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

the GPU processing capacity measurement unit is one of the following units:

percent; hz; kilohertz, kHz; MHz; gigahertz Ghz;

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the memory storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB;

the measurement unit of the hard disk storage capacity is one of the following:

percent; kilobyte KB; megabyte MB; gigabyte GB; terabyte TB; beat byte PB.

14. A management platform comprising a processor, a memory, and a program or instructions stored on the memory and executable on the processor, which when executed by the processor, implement the steps of the algorithm matching method of any of claims 1 to 13.

15. An arithmetic force matching device, comprising:

16. A readable storage medium, on which a program or instructions are stored, which when executed by a processor implement the steps of the algorithm matching method according to any of claims 1 to 13.