Table 1 Comparison of related work.

From: Efficient workflow scheduling using an improved multi-objective memetic algorithm in cloud-edge-end collaborative framework

Work

Scenario

Solution method

Task dependencies

Scheduling constraints

Objective

16

Edge-end

LBCMAB

✗

✗

minimize the latency and offloading cost(S)

17

Edge-end

IGOWOA

✗

✗

minimize user energy consumption, task response delay, and the number of deployed cloudlets(M)

18

Edge-end

PSO

✗

✓

minimize time delay, energy consumption and task execution cost(M)

19

Edge-end

MCO-A

✗

✗

minimize the network latency and energy consumption(S)

20

Cloud-edge-end

EDP-TO

✗

✓

minimize the latency

21

Edge-end

GA

✓

✗

minimize latency and offloading failure probability(S)

22

Edge-end

MAMTS

✓

✗

minimize the average completion time of multiple applications(S)

23

Edge-end

MDP

✓

✓

minimizes the total completion time(S)

29

Cloud-edge-end

ACO

✗

✗

minimize system latency, energy consumption and load balancing rate(M)

33

Edge-end

SCO

✓

✗

minimize energy consumption and time consumption(S)

36

Edge-end

GA

✓

✗

minimize energy consumption(S)

37

Cloud-edge-end

HORSA

✓

✗

minimize makespan, cost and maximize resource utilization (M)

39

Edge-end

MOEA/D

✗

✓

minimize energy consumption and makespan (M)

Ours

Cloud-edge-end

IMOMA

✓

✓

minimize energy consumption and makespan(M)

  1. LBCMAB (Learning-Based Co-Offloading Approach Based on MAB); S: Single-objective optimization; M: Multi-objective optimization