Calculation unloading method in multi-user multi-edge server scene

Abstract

The invention discloses a calculation unloading method in a multi-user multi-edge server scene, and belongs to the field of mobile cloud computing. The method comprises the following specific steps that: 1, a controller models a multi-user multi-edge server calculation unloading problem by utilizing a calculation unloading modeling method under a multi-user multi-edge server scene according to collected all user information and a target of a system, so that a double-layer optimization model is obtained; 2, the controller converts an original double-layer optimization problem into an equivalentsingle-layer optimization problem according to the characteristics of the initial problem model obtained in the previous step; and 3, the controller solves the single-layer optimization problem modelobtained in the previous step by using a branch limit method. According to the method provided by the invention, the problems of calculation time delay and calculation energy of the user are considered at the same time, the diversity of user requirements is met, and the economic benefit is greatly improved; according to the method provided by the invention, local and cloud computing costs of theuser are simultaneously considered, and the computing cost of the overall user is effectively reduced.

Description

technical field [0001] The invention relates to a calculation unloading method in a multi-user multi-edge server scenario, belonging to the field of mobile cloud computing. This method minimizes user computational cost and minimizes operator loss. Background technique [0002] With the development of edge computing, operators have deployed a large number of edge servers around users to provide users with computing offloading services. In the scenario of multiple users and multiple edge servers, different users can dynamically choose which edge servers to use for computing according to their own device status and server computing prices. At the same time, in order to earn more profits, edge servers will also adjust prices and services accordingly to attract more customers. Therefore, how to optimize user decision-making and edge server pricing to minimize user computing costs and maximize edge server benefits is particularly important. [0003] The main problem of computat...

AI Technical Summary

Problems solved by technology

It developed an optimal offloading algorithm based on the Markov decision process (MDP) model for mobile devices in intermittently connected cloud systems to minimize user computing costs (Offloading in Mobile Cloudlet Systems with Intermittent Connectivity.IEEE Transactions on Mobile Computing, 2015.14(12): 2516-2529.) The second category mainly studies how to minimize the loss of operators, and studies the pricing strategy of CSP that has given computing resources but does not combine it with user computing costs (Price Competition in a Duopoly IaaSCloud Market.201416Asia-Pacific.IEEE,2014:1-4.)

[0004] In summary, the current related computing offloading problems have not conducted joint research on user decision-making and edge server pricing in the multi-user and multi-edge server scenario.

Images