CN111866111A

CN111866111A - Edge deployment method for sensing user demand growth trend

Info

Publication number: CN111866111A
Application number: CN202010668788.9A
Authority: CN
Inventors: 秦肖雲; 王博; 曹洁; 崔霄; 王昌海; 张志锋
Original assignee: Zhengzhou University of Light Industry
Current assignee: Zhengzhou University of Light Industry
Priority date: 2020-07-13
Filing date: 2020-07-13
Publication date: 2020-10-30
Anticipated expiration: 2040-07-13
Also published as: CN111866111B

Abstract

The invention relates to the technical field of information technology, in particular to an edge deployment method for perceiving the growth trend of user demand, which provides an edge deployment scheme meeting the growth demand of users for service providers; the method is characterized by comprising the following steps: step 1, dividing historical operation time of an edge computing center into m time periods; step 2, calculating the edge resource amount required by the user in each time period divided in the step 1; and 3, calculating the resource quantity required by the user in the central historical operation time period according to the m edges obtained in the step 2 by using a time sequence analysis method, and predicting the edge resource quantity required by the user in a future time period.

Description

Edge deployment method for sensing user demand growth trend

Technical Field

The invention relates to the technical field of information technology, in particular to an edge deployment method for sensing the increase trend of user demand.

Background

Many smart devices (e.g., smartphones) are unable to efficiently perform user tasks due to their limited resource capacity and battery power. Some service providers then employ edge computing techniques to reduce the computing latency of tasks and reduce the computing power consumption of devices by offloading some of the user tasks of smart devices (especially mobile devices) to nearby edge computing centers. However, due to the limitations of the site space and the cooling conditions, only limited server resources can be deployed by one edge computing center, and how to efficiently use edge resources is one of the problems that needs to be solved, where designing an efficient edge deployment scheme is one of the effective methods for solving the problem.

In the edge computing environment, the edge deployment scheme reasonably deploys corresponding edge resources for each computing center, so that the idle edge resource amount during operation is optimized, and the user requirements during edge operation are met. Some works have designed effective edge deployment schemes by using historical service information of operation, however, all of the works only adopt the existing information and do not consider the future change trend of user requirements.

With the development of information technology and the increasing demand of people for nice life, smart devices, such as smart phones, wearable devices and smart homes, are becoming more popular. And with the rapid development of artificial intelligence algorithms (such as deep learning) and communication technologies (such as 5G, WiFi6/7) in recent years, the internet services facing intelligent devices are rapidly increased in diversity and complexity, which results in that the existing edge server deployment method is likely to design a scheme which cannot meet the requirements of users during operation.

Disclosure of Invention

In order to solve the technical problem, the invention provides an edge deployment method for providing an edge deployment scheme meeting the user growth requirement for a service provider, and the edge deployment method is used for sensing the user requirement growth trend.

The invention discloses an edge deployment method for sensing the increase trend of user demand, which comprises the following steps:

step 1, dividing historical operation time of an edge computing center into m time periods;

step 2, calculating the edge resource amount required by the user in each time period divided in the step 1;

and 3, calculating the resource quantity required by the user in the central historical operation time period according to the m edges obtained in the step 2 by using a time sequence analysis method, and predicting the edge resource quantity required by the user in a future time period.

The invention discloses an edge deployment method for sensing the increase trend of user demand, which further comprises the following steps:

recording the data acquisition time of the edge computing center during historical operation as { tau_j1.. T }, wherein τ is when 1 ≦ j1 < j2 ≦ T_j1＜τ_j2(ii) a The collected information comprises task information requested by a user to the edge computing center, and the set of tasks requested by the user is recorded as { task_i1., n }, and s_iAnd f_iTask respectively_iThe start time and the end time of (2), note r_iIs task_iThe amount of resources consumed during execution; the kth time period comprises the time of acquiring the execution information of the user task { tau_j|j＝τ_(k-1)·Δ,...,τ_k·ΔIn which τ is_jRepresenting the jth information acquisition time;

is a rounded-down symbol; t represents the number of information collection times during the historical operation of the edge computing center.

in step 2, the calculation formula for calculating the amount of the edge resources required by the user in each time period divided in step 1 is as follows:

wherein R is_kRepresenting the amount of edge resources required by the user in the kth time period; r is_iIs task_iThe amount of resources consumed during execution; task_iRepresenting the ith task requested by a user during the operation of the edge computing center; z is a radical of_i,jRepresenting task_iWhether to run at jth data acquisition time: when s is_i≤τ_j≤f_iWhen z is_i,j1 is ═ 1; in other cases z_i,j＝0，s_iAnd f_iTask respectively_iThe start time and the end time of (c); tau is_jAnd showing the collection time of the j-th user task execution information.

step 3 is used for predicting the edge resource amount required by the user in a future time period, and the calculation formula is as follows:

wherein R is_m+1Representing the amount of edge resources required by the user for a period of time in the future; r_kRepresenting the edge resource amount required by the user in the kth time period, and calculating the edge resource amount in the step 2; and the self-correlation offset factor set for the user meets the condition that m-1 is more than or equal to 1.

Compared with the prior art, the invention has the beneficial effects that: the invention uses the task information of the edge computing center during the historical operation to compute the edge resource amount needed by the user in each time period, and uses the time sequence prediction method to obtain the change trend of the edge resource amount needed by the user, and predicts the edge resource amount needed by the user in a future time period, thereby providing an edge deployment scheme which accords with the change trend of the user requirement for the service provider.

Drawings

FIG. 1 is a logic flow diagram of the present invention.

Detailed Description

The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.

Examples

Recording the data acquisition time of the edge computing center during historical operation as { tau_j1.. T }, wherein τ is when 1 ≦ j1 < j2 ≦ T_j1＜τ_j2. The collected information comprises task information requested by a user to the edge computing center, and the set of tasks requested by the user is recorded as { task_i1., n }, and s_iAnd f_iTask respectively_iThe start time and the end time of (2), note r_iIs task_iThe amount of resources consumed during execution.

As shown in fig. 1, a specific flow of an edge deployment method for sensing a user demand growth trend is as follows:

step 1: dividing the historical operation time of the edge computing center into m time periods, wherein the kth time period comprises the user task execution information acquisition time { tau_j|j＝τ_(k-1)·Δ,...,τ_k·ΔIn which τ is_jRepresenting the jth information acquisition time;

Step 2: calculating the edge resource amount required by the user in each time period divided in the step 1, wherein the calculation formula is as follows:

Wherein R is_kRepresenting the amount of edge resources required by the user in the kth time period; r is_iIs task_iThe amount of resources consumed during execution; task_iRepresenting the ith task requested by a user during the operation of the edge computing center; z is a radical of_i,jRepresenting task_iWhether to run at jth data acquisition time: when s is_i≤τ_j≤f_iWhen z is_i,j1, otherwise z_i,j＝0；s_iAnd f_iTask respectively_iThe start time and the end time of (c); tau is_jAnd showing the collection time of the j-th user task execution information.

And step 3: and (3) calculating the resource quantity required by the user in the historical operation time period of the center according to the m edges obtained in the step (2) by using a time series analysis method, predicting the edge resource quantity required by the user in a future time period, wherein the calculation formula is as follows:

The invention uses the task information of the edge computing center during the historical operation to compute the edge resource amount needed by the user in each time period, and uses the time sequence prediction method to obtain the change trend of the edge resource amount needed by the user, and predicts the edge resource amount needed by the user in a future time period, thereby providing an edge deployment scheme which accords with the change trend of the user requirement for the service provider.

The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims

1. An edge deployment method for sensing the increase trend of user demand is characterized by comprising the following steps:

2. The edge deployment method for sensing the increasing trend of user demand as claimed in claim 1, wherein the data collection time during the historical operation of the edge computing center is recorded as { τ }_j1.. T }, wherein τ is when 1 ≦ j1 < j2 ≦ T_j1＜τ_j2(ii) a The collected information comprises task information requested by a user to the edge computing center, and the set of tasks requested by the user is recorded as { task _i1., n }, and s_iAnd f_iTask respectively_iThe start time and the end time of (2), note r_iIs task_iThe amount of resources consumed during execution; the kth time period comprises the time of acquiring the execution information of the user task { tau_j|j＝τ_(k-1)·Δ,...,τ_k·ΔIn which τ is_jRepresenting the jth information acquisition time;

3. The edge deployment method for sensing the increasing trend of the user demand as claimed in claim 2, wherein the calculation formula for calculating the amount of the edge resources required by the user in each time segment divided in step 1 in step 2 is as follows:

wherein R is_kRepresenting the amount of edge resources required by the user in the kth time period; r is_iIs task_iThe amount of resources consumed during execution;task_irepresenting the ith task requested by a user during the operation of the edge computing center; z is a radical of_i,jRepresenting task_iWhether to run at jth data acquisition time: when s is_i≤τ_j≤f_iWhen z is_i,j1, otherwise z_i,j＝0；s_iAnd f_iTask respectively_iThe start time and the end time of (c); tau is_jAnd showing the collection time of the j-th user task execution information.

4. The edge deployment method for sensing the increasing trend of the user demand as claimed in claim 3, wherein the step 3 is used for predicting the amount of the edge resources required by the user in a future time period, and the calculation formula is as follows: