阅读说明：本技术 基于边缘网关数据质量评价的计算能力迁移方法及系统 (Computing power migration method and system based on edge gateway data quality evaluation ) 是由 仲刚 何斌 李波 于 2019-07-24 设计创作，主要内容包括：本发明公开了一种基于边缘网关数据质量评价的计算能力迁移方法及系统,解决了数据应用效率不高的问题,其技术方案要点是给边缘网关配置局部计算能力服务,边缘网关采集接入设备的数据并上送给边缘计算平台,按照评价指标对边缘网关和边缘计算平台接收到的数据进行质量评价,得到边缘网关的评价结果DQ(gw)和边缘计算平台的评价结果DQ(pf)。当DQ(gw)/DQ(pf)大于第一阈值时,边缘计算平台发送启动指令给边缘网关,边缘网关则启动其局部计算能力服务,边缘计算平台停止相关的局部计算能力；当DQ(gw)/DQ(pf)小于第二阈值时,边缘计算平台发送停止指令给边缘网关,边缘网关则停止其局部计算能力服务,边缘计算平台启动相关的局部计算能力,最终实现数据应用效率的提高。(the invention discloses a computing power migration method and a computing power migration system based on edge gateway data quality evaluation, which solve the problem of low data application efficiency. When DQ (gw)/DQ (pf) is greater than a first threshold, the edge computing platform sends a start instruction to the edge gateway, the edge gateway starts local computing capability service, and the edge computing platform stops related local computing capability; when DQ (gw)/DQ (pf) is smaller than a second threshold, the edge computing platform sends a stop instruction to the edge gateway, the edge gateway stops its local computing capability service, the edge computing platform starts the related local computing capability, and finally the data application efficiency is improved.)

1. A computing power migration method based on edge gateway data quality evaluation is characterized by comprising the following steps:

the method comprises the steps that an edge gateway collects data of access equipment in an autonomous region, evaluates the quality of the data according to evaluation indexes to obtain an evaluation result DQ (gw), and configures local computing power service;

the data and the DQ (gw) are sent to an edge computing platform, the edge computing platform evaluates the quality of the data according to the evaluation index to obtain an evaluation result DQ (pf), and the edge computing platform judges whether DQ (gw)/DQ (pf) reaches the standard or not;

if the DQ (gw)/DQ (pf) is greater than a first threshold, the edge computing platform sends a start instruction to the edge gateway, the edge gateway starts the local computing capability service, and the edge computing platform stops the related local computing capability;

If the DQ (gw)/DQ (pf) is less than a second threshold, the edge computing platform starts up the associated local computing power or

The edge computing platform sends a stopping instruction to the edge gateway, the edge gateway stops the local computing capacity service, and meanwhile, the edge computing platform starts the related local computing capacity;

Wherein DQ (pf) is more than 0 and less than or equal to DQ (gw) and less than or equal to 1.

2. The method for computing power migration based on edge gateway data quality assessment according to claim 1, wherein the first threshold is 1.2 and the second threshold is 1.05.

3. the computing power migration method based on edge gateway data quality evaluation according to claim 1 or 2, wherein the evaluation index at least includes data access condition, data completeness, data timeliness and data consistency, wherein the data access condition at least includes:

Real-time on-line rate of access device, FI₁＝A₁/B₁，A₁for the number of data values of the access device in the acquisition cycle, B₁the number of data values of the equipment to be accessed in the acquisition period;

historical online rate of access device, FI₂＝A₂/B₂，A₂Number of data values of access devices in a defined time or frequency period, B₂the number of data values of the equipment to be accessed in a specified time or frequency period;

The completeness of the data at least comprises:

completeness of data batch FI₃＝A₃/B₃，A₃Collecting the number of data items in a batch record for a single access device, B₃Acquiring the number of data items to be acquired in the batch record for a single access device;

completeness of data record FI₄＝A₄/B₄，A₄The number of complete records collected in a given time or frequency period, B₄The number of the complete records which should be collected in a specified time or frequency period;

Ratio of effective recording, FI₅＝A₅/B₅，A₅Number of valid records collected within a specified time or frequency period, B₅The number of records actually acquired in a specified time or frequency period;

The data timeliness include at least:

data update timeliness ratio of FI₆＝A₆/B₆，A₆The number of access equipment data values acquired in time in a specified time or frequency period, B₆the number of the access equipment data values which are required to be acquired in a specified time or frequency period;

The data consistency includes at least:

data managementNew conformity ratio, FI₇＝A₇/B₇，A₇average period of access device data values acquired in real time within a defined time or frequency period, B₇Is the collection period;

Data value validity ratio of FI₈＝A₈/B₈，A₈the number of effective data values of the access equipment, B, collected in time in a specified time or frequency period₈the number of access device data values to be acquired in a predetermined time or frequency period.

4. The computing power migration method based on edge gateway data quality evaluation according to claim 3, wherein if the evaluation result is DQ, thenwherein FI_i＝A_i/B_i，W_iIs a weight parameter andn∈[8,+∞]And i is a positive integer.

5. A computing power migration system based on edge gateway data quality evaluation is characterized by comprising:

at least one access device;

the edge gateway is used for acquiring data of access equipment in the autonomous region, evaluating the data quality according to evaluation indexes to obtain an evaluation result DQ (gw), uploading the data and the DQ (gw) to an edge computing platform, and configuring local computing capability service;

The edge computing platform evaluates the data quality according to the evaluation index to obtain an evaluation result DQ (pf);

The edge computing platform comprises a judging unit, and the judging unit judges whether DQ (gw)/DQ (pf) reaches the standard:

if the DQ (gw)/DQ (pf) is greater than a first threshold, the edge computing platform sends a start instruction to the edge gateway, the edge gateway starts the local computing capability service, and the edge computing platform stops the related local computing capability;

If the DQ (gw)/DQ (pf) is less than a second threshold, the edge computing platform starts up the associated local computing power or

the edge computing platform sends a stopping instruction to the edge gateway, the edge gateway stops the local computing capacity service, and meanwhile, the edge computing platform starts the related local computing capacity;

wherein DQ (pf) is more than 0 and less than or equal to DQ (gw) and less than or equal to 1.

6. The edge gateway data quality assessment based computing power migration system according to claim 5, wherein said first threshold value is 1.2 and said second threshold value is 1.05.

7. The edge gateway data quality evaluation based computing power migration system according to claim 5 or 6, wherein the evaluation index at least comprises data access condition, data completeness, data timeliness and data consistency, wherein the data access condition at least comprises:

real-time on-line rate of access device, FI₁＝A₁/B₁，A₁For the number of data values of the access device in the acquisition cycle, B₁the number of data values of the equipment to be accessed in the acquisition period;

Historical online rate of access device, FI₂＝A₂/B₂，A₂Number of data values of access devices in a defined time or frequency period, B₂the number of data values of the equipment to be accessed in a specified time or frequency period;

the completeness of the data at least comprises:

completeness of data batch FI₃＝A₃/B₃，A₃collecting the number of data items in a batch record for a single access device, B₃Is a singleThe access equipment acquires the number of data items to be acquired in the batch record;

Completeness of data record FI₄＝A₄/B₄，A₄The number of complete records collected in a given time or frequency period, B₄The number of the complete records which should be collected in a specified time or frequency period;

ratio of effective recording, FI₅＝A₅/B₅，A₅number of valid records collected within a specified time or frequency period, B₅The number of records actually acquired in a specified time or frequency period;

the data timeliness include at least:

Data update timeliness ratio of FI₆＝A₆/B₆，A₆The number of access equipment data values acquired in time in a specified time or frequency period, B₆The number of the access equipment data values which are required to be acquired in a specified time or frequency period;

The data consistency includes at least:

Data update consistency ratio of FI₇＝A₇/B₇，A₇average period of access device data values acquired in real time within a defined time or frequency period, B₇is the collection period;

data value validity ratio of FI₈＝A₈/B₈，A₈The number of effective data values of the access equipment, B, collected in time in a specified time or frequency period₈The number of access device data values to be acquired in a predetermined time or frequency period.

8. The computing power migration system of claim 7, wherein the evaluation result is DQ, thenWherein FI_i＝A_i/B_i，W_iIs a weight parameter andn∈[8,+∞]And i is a positive integer.

9. the edge gateway data quality assessment based computing power migration system of claim 8, wherein there is at least one edge gateway.

Technical Field

the present disclosure relates to the field of edge computing, and in particular, to a computing power migration method and system based on edge gateway data quality evaluation.

Background

In edge computing, data quality is the basis of data application, and high-quality data is possible to mine high-value data application. The data quality evaluation is one of the main functions of the edge computing platform and is used for evaluating the data of the edge computing edge gateway and the access equipment acquired by the edge computing edge gateway, and the main evaluation indexes comprise data access conditions, data accuracy, completeness, timeliness and consistency. The evaluation index of the data is generally based on the quality characteristics of the data finally stored in the platform, and according to the data quality evaluation result, each part in the edge calculation is improved and continuously optimized (for example, communication quality is improved, gateway computing capacity is improved, and the like). The basis of data quality evaluation generally relates to integrity information such as data time stamp, data check and the like uploaded through the edge gateway.

Stability of communications between the edge gateway and the edge computing platform, availability and reliability of the edge gateway itself, all have an impact on the quality characteristics of the uploaded data. The data quality evaluation result based on a single platform can only be aimed at an autonomous area formed by an edge gateway and related access equipment, and the evaluation value of the data quality inside the area is not very high. As shown in fig. 1, the communication instability of the edge gateway (EG007) will affect the integrity and timeliness of the data uploaded to the edge computing platform, and the result is also worse by a single evaluation based on the quality of the data uploaded by the edge gateway, whereas the quality of the data processed on the edge gateway is better. If the local computing capability of the edge computing platform (for example, the computing decision-making capability based on the data in the autonomous region) can be set down in the edge gateway based on the above evaluation, the efficiency of data application can be greatly improved.

disclosure of Invention

The disclosure provides a computing power migration method and a computing power migration system based on edge gateway data quality evaluation, and the technical purpose of improving data application efficiency is achieved.

the technical purpose of the present disclosure is achieved by the following technical solutions:

A computing power migration method based on edge gateway data quality evaluation comprises the following steps:

the method comprises the steps that an edge gateway collects data of access equipment in an autonomous region, evaluates the quality of the data according to evaluation indexes to obtain an evaluation result DQ (gw), and configures local computing power service;

The data and the DQ (gw) are sent to an edge computing platform, the edge computing platform evaluates the quality of the data according to the evaluation index to obtain an evaluation result DQ (pf), and the edge computing platform judges whether DQ (gw)/DQ (pf) reaches the standard or not;

If the DQ (gw)/DQ (pf) is greater than a first threshold, the edge computing platform sends a start instruction to the edge gateway, the edge gateway starts the local computing capability service, and the edge computing platform stops the related local computing capability;

if the DQ (gw)/DQ (pf) is less than a second threshold, the edge computing platform starts up the associated local computing power or

The edge computing platform sends a stopping instruction to the edge gateway, the edge gateway stops the local computing capacity service, and meanwhile, the edge computing platform starts the related local computing capacity;

wherein DQ (pf) is more than 0 and less than or equal to DQ (gw) and less than or equal to 1.

Further, the first threshold is 1.2, and the second threshold is 1.05.

Further, the evaluation index at least includes data access condition, data completeness, data timeliness and data consistency, where the data access condition at least includes:

real-time on-line rate of access device, FI₁＝A₁/B₁，A₁For the number of data values of the access device in the acquisition cycle, B₁The number of data values of the equipment to be accessed in the acquisition period;

Historical online rate of access device, FI₂＝A₂/B₂，A₂Number of data values of access devices in a defined time or frequency period, B₂The number of data values of the equipment to be accessed in a specified time or frequency period;

the completeness of the data at least comprises:

Completeness of data batch FI₃＝A₃/B₃，A₃collecting the number of data items in a batch record for a single access device, B₃Acquiring the number of data items to be acquired in the batch record for a single access device;

Completeness of data record FI₄＝A₄/B₄，A₄The number of complete records collected in a given time or frequency period, B₄the number of the complete records which should be collected in a specified time or frequency period;

Ratio of effective recording, FI₅＝A₅/B₅，A₅Number of valid records collected within a specified time or frequency period, B₅The number of records actually acquired in a specified time or frequency period;

The data timeliness include at least:

data update timeliness ratio of FI₆＝A₆/B₆，A₆The number of access equipment data values acquired in time in a specified time or frequency period, B₆The number of the access equipment data values which are required to be acquired in a specified time or frequency period;

the data consistency includes at least:

Data update consistency ratio of FI₇＝A₇/B₇，A₇Average period of access device data values acquired in real time within a defined time or frequency period, B₇Is the collection period;

Data value validity ratio of FI₈＝A₈/B₈，A₈the number of effective data values of the access equipment, B, collected in time in a specified time or frequency period₈The number of access device data values to be acquired in a predetermined time or frequency period.

Further, if the evaluation result is DQ, thenwherein FI_i＝A_i/B_i，W_iIs a weight parameter andn∈[8,+∞]and i is a positive integer.

A computing power migration system based on edge gateway data quality evaluation comprises:

At least one access device;

the edge gateway is used for acquiring data of access equipment in the autonomous region, evaluating the data quality according to evaluation indexes to obtain an evaluation result DQ (gw), uploading the data and the DQ (gw) to an edge computing platform, and configuring local computing capability service;

the edge computing platform evaluates the data quality according to the evaluation index to obtain an evaluation result DQ (pf);

The edge computing platform comprises a judging unit, and the judging unit judges whether DQ (gw)/DQ (pf) reaches the standard:

If the DQ (gw)/DQ (pf) is greater than a first threshold, the edge computing platform sends a start instruction to the edge gateway, the edge gateway starts the local computing capability service, and the edge computing platform stops the related local computing capability;

if the DQ (gw)/DQ (pf) is less than a second threshold, the edge computing platform starts up the associated local computing power; or

The edge computing platform sends a stopping instruction to the edge gateway, the edge gateway stops the local computing capacity service, and meanwhile, the edge computing platform starts the related local computing capacity;

Wherein DQ (pf) is more than 0 and less than or equal to DQ (gw) and less than or equal to 1.

further, the first threshold is 1.2, and the second threshold is 1.05.

Further, the evaluation index at least includes data access condition, data completeness, data timeliness and data consistency, where the data access condition at least includes:

Real-time on-line rate of access device, FI₁＝A₁/B₁，A₁For the number of data values of the access device in the acquisition cycle, B₁The number of data values of the equipment to be accessed in the acquisition period;

Historical online rate of access device, FI₂＝A₂/B₂，A₂Number of data values of access devices in a defined time or frequency period, B₂The number of data values of the equipment to be accessed in a specified time or frequency period;

the completeness of the data at least comprises:

Completeness of data batch FI₃＝A₃/B₃，A₃Collecting the number of data items in a batch record for a single access device, B₃acquiring the number of data items to be acquired in the batch record for a single access device;

completeness of data record FI₄＝A₄/B₄，A₄The number of complete records collected in a given time or frequency period, B₄The number of the complete records which should be collected in a specified time or frequency period;

ratio of effective recording, FI₅＝A₅/B₅，A₅Number of valid records collected within a specified time or frequency period, B₅the number of records actually acquired in a specified time or frequency period;

The data timeliness include at least:

Data update timeliness ratio of FI₆＝A₆/B₆，A₆The number of access equipment data values acquired in time in a specified time or frequency period, B₆The number of the access equipment data values which are required to be acquired in a specified time or frequency period;

the data consistency includes at least:

data update consistency ratio of FI₇＝A₇/B₇，A₇average period of access device data values acquired in real time within a defined time or frequency period, B₇Is the collection period;

Data value validity ratio of FI₈＝A₈/B₈，A₈The number of effective data values of the access equipment, B, collected in time in a specified time or frequency period₈the number of access device data values to be acquired in a predetermined time or frequency period.

further, if the evaluation result is DQ, thenwherein FI_i＝A_i/B_i，W_iis a weight parameter andn∈[8,+∞]And i is a positive integer.

further, the number of the edge gateways is at least one.

In conclusion, the beneficial effects of the present disclosure are: and configuring local computing capability service for the edge gateway, acquiring data of the access equipment by the edge gateway and uploading the data to an edge computing platform, and performing quality evaluation on the data received by the edge gateway and the edge computing platform according to evaluation indexes to obtain an evaluation result DQ (gw) of the edge gateway and an evaluation result DQ (pf) of the edge computing platform. Then, whether DQ (gw)/DQ (pf) reach the standard or not is judged through a judging unit of the edge computing platform, when DQ (gw)/DQ (pf) are larger than a first threshold value, the edge computing platform sends a starting instruction to an edge gateway, the edge gateway starts local computing capability service, and meanwhile, the edge computing platform stops relevant local computing capability; when DQ (gw)/DQ (pf) is smaller than a second threshold, the edge computing platform sends a stop instruction to the edge gateway, the edge gateway stops its local computing capability service, and the edge computing platform starts the related local computing capability, so as to finally improve the data application efficiency.

drawings

FIG. 1 is a schematic diagram of an edge computation topology;

FIG. 2 is a schematic diagram of data evaluation indexes inside an autonomous region;

FIG. 3 is an algorithmic flow chart of the method of the present disclosure.

Detailed Description

the present disclosure is described in further detail below with reference to the attached drawing figures.

in the description of the present disclosure, it is to be understood that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated, but merely as being used to distinguish between different elements.

The system disclosed by the invention comprises access equipment, an edge gateway and an edge computing platform, wherein referring to fig. 1, the edge gateway has certain computing capacity, and the edge computing platform comprises a judging unit.

Fig. 3 is an algorithm flow chart of the method of the present disclosure, and first, a data quality evaluation method, a local computing power dropping condition, and a local computing power dropping trigger mechanism are configured for an edge computing platform, and a data quality evaluation method is configured for an edge gateway and a local computing power service is registered.

Then the edge gateway collects the data of the access equipment in the autonomous region, and evaluates the data quality according to the evaluation index to obtain an evaluation result DQ (gw); and the edge gateway sends the acquired data and DQ (gw) to an edge computing platform, and the edge computing platform evaluates the quality of the data according to the evaluation indexes to obtain an evaluation result DQ (pf). The evaluation index is shown in fig. 2 and table 1.

TABLE 1

Judging by a judging unit of the edge computing platform, if DQ (gw)/DQ (pf) is greater than a first threshold (namely judging a local computing power transfer condition), sending a starting instruction to an edge gateway by the edge computing platform, starting local computing power service by the edge gateway, and stopping related local computing power by the edge computing platform; if DQ (gw)/DQ (pf) is less than the second threshold, the edge computing platform starts its own associated local computing power.

After the edge computing platform transfers the local computing capacity to the edge gateway, the edge gateway continues to acquire data and upload the data to the edge computing platform, if the situation that the DQ (gw)/DQ (pf) is smaller than the second threshold value occurs (i.e. judging the local computing capacity recovery condition), the edge computing platform should send a stop instruction to the edge gateway, the edge gateway stops the local computing capacity service, and meanwhile, the edge computing platform starts the relevant local computing capacity of the edge computing platform. DQ (pf) is more than 0 and less than or equal to DQ (gw) is less than or equal to 1. Fig. 3 shows a case where an edge computing platform corresponds to one edge gateway, and when there are multiple edge gateways, the algorithm flows between each edge gateway and the edge computing platform are the same.

Here, the evaluation resultsWherein FI_i＝A_i/B_i，W_iis a weight parameter andn∈[8,+∞]i is a positive integer, FI_iFor the quantitative function of the evaluation index, the corresponding relationship is shown in fig. 2. DQ (gw) is the quality evaluation result of the data in the edge gateway autonomous region, DQ (pf) is the quality evaluation result of the data received by the edge computing platform, all obtained from the above formula, and their FIs_iAnd W_iDifferent. In addition, the number of terms of the evaluation index is not limited to eight as described in the present disclosure, so n ∈ [8, + ∞]non-inventive changes that can be made under the teachings of the present disclosure are within the scope of the present disclosure.

The first threshold value is 1.2, and the second threshold value is 1.05. When the threshold is 1.05,1.2, the distribution of the computing power has no great influence on the system, but if the first threshold and the second threshold are the same, the switching of the computing power may be repeated and unstable.

The foregoing is an exemplary embodiment of the present disclosure, and the scope of the present disclosure is defined by the claims and their equivalents.