Skip to main content

Advances, Systems and Applications

Table 1 Some existing surveys on computing offloading in edge computing

From: Task partitioning and offloading in IoT cloud-edge collaborative computing framework: a survey

Survey paper

Summary

Heidari et al. [10]

The authors outlined task offloading for different computing environments, with offloading strategies and decision processes.

Zhang et al. [11]

The authors focused on architecture, computation migration, edge caching, and service orchestration in task offloading.

Saeik et al. [12]

The authors emphasized on solving the task offloading problem with the mathematical, artificial intelligence and control theory optimization approaches.

Feng et al. [13]

The authors compared the approaches including mathematical solver, heuristic algorithms, Lyapunov optimization, game theory, and MDP/RL for solving the task offloading problem.

Mach et al. [14]

The authors focused mainly on computation offloading decisions, by considering allocation of computing resources and mobility management in MEC.

Jiang et al. [15]

The authors classified the existing work on task offloading into two categories, namely gaming and cooperation between edge and the cloud, and heuristic algorithms.

Wang et al.  [16]

The authors classified the existing work on task offloading into five categories, based on offloading destination, load balance of edge servers, device mobility, application partitioning, and partition granularity, respectively.

Lin et al. [17]

The authors classified the computation offloading from offloading flow and offloading scenario. They also classified the computation offloading schemes into five categories, which are (non)convex optimization, MDP, game theory, Lyapunov optimization, and machine learning.

Shakarami et al. [18]

The authors only reviewed the computation offloading approaches based on game-theoretic for edge computing.

Wang et al. [19]

The authors reviewed the task offloading schemes considering response time, device energy, service provider cost, load balance between edge and cloud, and device mobility.

This paper

We consider task partitioning together with computing offloading, and propose a general framework including an application model and a decision engine, and more comprehensive taxonomy metrics for classifying task partitioning and offloading approaches.