Mobile terminal calculation unloading method based on task distribution

Abstract

The invention discloses a mobile terminal calculation unloading method based on task distribution, and belongs to the field of mobile communication. The calculation unloading method comprises the following steps: a) firstly, carrying out task division on a task to be executed to form a series of subtasks; b) then, according to the execution time of the mobile terminal, sorting the subtasks obtained in the a), wherein calculation time in direct proportion to priority; c) comprehensively considering the execution efficiency ratio k of each cloud server and the mobile terminal, the execution time t1 of each subtask on the mobile terminal, the residence time t2 of the cloud servers and network bandwidth B to find the cloud server to which each subtask can be unloaded; and d) finally, according to a descending order of the propriety, successively the subtasks to the optimal cloud server in the cloud servers to which the subtasks can be unloaded. The mobile terminal calculation unloading method divides the task to be executed into one series of subtasks, and loads the subtasks to a plurality of cloud servers with higher efficiency to be executed, so that the calculation time of the mobile terminal can be shortened, and energy consumption is lowered.

Description

technical field

[0001] The mobile terminal computing unloading method based on task allocation of the present invention belongs to the field of mobile communication. Background technique

[0002] Currently, the usage rate of portable mobile terminals such as smartphones and tablet computers is far greater than that of PCs. However, because mobile terminals are limited by battery life, storage space, and information processing capabilities, considerable energy consumption and computing costs will be generated when performing computationally intensive tasks.

[0003] Computing offloading is a way of placing mobile terminal computing tasks in cloud servers for execution, which can reduce computing time and energy consumption by reducing the computing tasks of mobile terminals.

[0004] How to achieve the matching between the tasks to be processed and the cloud server is directly related to the efficiency of computing offloading. However, the computing method of the current com...

Examples

specific Embodiment 1

[0021] The mobile terminal computing offloading method based on task assignment in this embodiment includes the following steps:

[0022] Step a, the mobile terminal divides the tasks to be executed to form a series of subtasks;

[0023] Step b, sort the subtasks obtained in step a according to the length of execution time in the mobile terminal; the longer the calculation time, the higher the priority;

[0024] Step c, comprehensively consider the execution efficiency ratio k between the cloud server and the mobile terminal, and the execution time t of the subtask on the mobile terminal 1 , the residence time of the cloud server t 2 , network bandwidth B, find the cloud server that can offload each subtask;

[0025] Step d, according to the order of priority from high to low, offload the subtasks to the optimal cloud server among the cloud servers from which the subtask can be offloaded.

specific Embodiment 2

[0026] The mobile terminal computing offloading method based on task allocation in this embodiment, on the basis of the specific embodiment 1, further limits the subtasks to be the smallest tasks that cannot be further divided.

[0027] Such a technical solution can ensure that the tasks to be executed are divided into the most detailed, that is, the most sub-tasks are obtained, then the most parallel tasks are executed in the cloud server, and the time to complete all sub-tasks is also the shortest, further reducing the calculation of the mobile terminal time and reduce energy consumption.

specific Embodiment 3

[0028] The mobile terminal computing offloading method based on task allocation in this embodiment, on the basis of the specific embodiment 1, further defines the cloud server that can be offloaded in step c, k>1, t 2 >kt 1 cloud server.

[0029] in:

[0030] k>1, which can ensure that the execution efficiency of the cloud server is higher than that of the mobile terminal, and ensure that running subtasks in the cloud server can reduce the computing time;

[0031] t 2 >kt 1 , which can ensure that the subtasks are calculated in a single server without task migration between servers, thereby saving the time spent on task migration and also reducing the calculation time.

technical field

[0001] The mobile terminal computing unloading method based on task allocation of the present invention belongs to the field of mobile communication. Background technique

[0002] Currently, the usage rate of portable mobile terminals such as smartphones and tablet computers is far greater than that of PCs. However, because mobile terminals are limited by battery life, storage space, and information processing capabilities, considerable energy consumption and computing costs will be generated when performing computationally intensive tasks.

[0003] Computing offloading is a way of placing mobile terminal computing tasks in cloud servers for execution, which can reduce computing time and energy consumption by reducing the computing tasks of mobile terminals.

[0004] How to achieve the matching between the tasks to be processed and the cloud server is directly related to the efficiency of computing offloading. However, the computing method of the current com...

Examples

specific Embodiment 1

[0021] The mobile terminal computing offloading method based on task assignment in this embodiment includes the following steps:

[0022] Step a, the mobile terminal divides the tasks to be executed to form a series of subtasks;

[0023] Step b, sort the subtasks obtained in step a according to the length of execution time in the mobile terminal; the longer the calculation time, the higher the priority;

[0024] Step c, comprehensively consider the execution efficiency ratio k between the cloud server and the mobile terminal, and the execution time t of the subtask on the mobile terminal 1 , the residence time of the cloud server t 2 , network bandwidth B, find the cloud server that can offload each subtask;

[0025] Step d, according to the order of priority from high to low, offload the subtasks to the optimal cloud server among the cloud servers from which the subtask can be offloaded.

specific Embodiment 2

[0026] The mobile terminal computing offloading method based on task allocation in this embodiment, on the basis of the specific embodiment 1, further limits the subtasks to be the smallest tasks that cannot be further divided.

[0027] Such a technical solution can ensure that the tasks to be executed are divided into the most detailed, that is, the most sub-tasks are obtained, then the most parallel tasks are executed in the cloud server, and the time to complete all sub-tasks is also the shortest, further reducing the calculation of the mobile terminal time and reduce energy consumption.

specific Embodiment 3

[0028] The mobile terminal computing offloading method based on task allocation in this embodiment, on the basis of the specific embodiment 1, further defines the cloud server that can be offloaded in step c, k>1, t 2 >kt 1 cloud server.

[0029] in:

[0030] k>1, which can ensure that the execution efficiency of the cloud server is higher than that of the mobile terminal, and ensure that running subtasks in the cloud server can reduce the computing time;

[0031] t 2 >kt 1 , which can ensure that the subtasks are calculated in a single server without task migration between servers, thereby saving the time spent on task migration and also reducing the calculation time.