A Meta-Heuristic-Based Approach for Qos-Aware Service Composition

Chenyang Li,Huiling Chen,Jun Li

doi:10.1109/access.2020.2987078

Chenyang Li, Huiling Chen + Show 1 more

Open Access

https://doi.org/10.1109/access.2020.2987078

Copy DOI

Abstract

Recently, with the rapid increase in the number of web services, QoS-aware Web Service Composition(QWSC) has become a popular topic in both industry and academia. Meta-heuristic algorithm, as an effective way to solve classical optimization problems, has been successfully applied to QWSC nowdays. However, such approach has intrinsic drawbacks and usually lack of good performance in large-scale scenarios. For example, some meta-heuristic algorithms are suitable for continuous search space, while the search space of QWSC is discrete. For solving those problems which were commonly faced when applying meta-heuristic algorithm on QWSC, in this research, we firstly introduce a preprocessing approach for constructing fuzzy continuous neighborhood relations of concrete services, which makes the local search strategy of meta-heuristic algorithms be as effective in discrete space as in continuous space, thus improving the optimization performance. Second, we combine Harris Hawks Optimization (HHO) algorithm and logical chaotic sequence to propose an improved meta-heuristic algorithm named CHHO for solving QWSC. The ergodic and chaotic characteristics of chaotic sequences are used to enhance the ability of the CHHO to jump out of the local optimum for further optimization. Experimental results show that the CHHO has better optimization performance by comparing with the existing mainstream algorithms when solving QWSC problems. Additionally, the preprocessing approach not only greatly improves the optimization performance of the CHHO but also can be freely utilized in other meta-heuristics based approaches.

Highlights

Service Oriented Computing (SOC) is a computing method that uses services as the basic unit to rapidly build distributed software systems and enterprise applications through service composition technology [1]
The meta-heuristic algorithms compared in this paper include: Chaos Harris Hawk Optimization (CHHO), Harris Hawks Optimization (HHO) [21], Eagle Strategy with Whale Optimization Algorithm (ESWOA) [35], mABC [36]
We firstly proposed a preprocessing approach for concrete service datasets called FOCC, which makes the local search strategy of meta-heuristic algorithms be as effective in discrete space as in continuous space, thereby making some meta-heuristic algorithms designed for continuous problems suitable for solving QWSC(discrete problem)

Summary

INTRODUCTION

Service Oriented Computing (SOC) is a computing method that uses services as the basic unit to rapidly build distributed software systems and enterprise applications through service composition technology [1]. This characteristic makes QWSC problem difficult to solve, especially with the explosive growth of the number of web services in recent years Various approaches such as deterministic algorithms, heuristic. C. Li et al.: Meta-Heuristic-Based Approach for Qos-Aware Service Composition algorithms, meta-heuristic algorithms, and hybrid methods have been proposed to solve this problem, there are still many problems in QWSC that need to be further studied and solved to meet the increasing requirements of users in a realistic and complex network environment [9], [10]. To address the issues discussed above, we propose an improved meta-heuristic algorithm based on HHO and a preprocessing approach for constructing fuzzy continuous neighborhood relations of concrete services.

RELATED WORKS

LOGISTIC CHAOTIC SINGLE-DIMENSIONAL PERTURBATION

30: Return Xbest

EVALUATION

CONCLUSION AND FUTURE WORK