Offloading as a Service Middleware for Mobile Cloud Apps

Offloading as a Service Middleware for Mobile Cloud Apps

Hamid A. Jadad, Abderezak Touzene, Khaled Day
Copyright: © 2020 |Pages: 20
DOI: 10.4018/IJCAC.2020040103
(Individual Articles)
No Current Special Offers


Recently, much research has focused on the improvement of mobile app performance and their power optimization, by offloading computation from mobile devices to public cloud computing platforms. However, the scalability of these offloading services on a large scale is still a challenge. This article describes a solution to this scalability problem by proposing a middleware that provides offloading as a service (OAS) to large-scale implementation of mobile users and apps. The proposed middleware OAS uses adaptive VM allocation and deallocation algorithms based on a CPU rate prediction model. Furthermore, it dynamically schedules the requests using a load-balancing algorithm to ensure meeting QoS requirements at a lower cost. The authors have tested the proposed algorithm by conducting multiple simulations and compared our results with state-of-the-art algorithms based on various performance metrics under multiple load conditions. The results show that OAS achieves better response time with a minimum number of VMs and reduces 50% of the cost compared to existing approaches.
Article Preview

1. Introduction

During the last decade, we have witnessed unprecedented growth in the mobile phone industry due to the popularity of smart devices owing to their affordability and advancement in related technologies. One reason for this popularity of smartphones is the rapid demand for mobile applications (GSMA corporate, 2017). For example, as of May 2019, the Google Play store contained around 2.1 million mobile apps, whereas the Apple Store offered 1.8 million apps (Statista, 2019). These apps can be classified into various categories such as entertainment, business, health care, and education. In the recent past, most of these apps required manageable computation, memory, and power. However, many advanced apps like multimedia processing, video gaming, speech recognition, and natural language processing which require a high level of computing power, memory, and energy. This scenario is quickly making it impractical to run such demanding apps locally on the devices. Hence, the cloud computing paradigm has come up with a solution.

Recently, cloud computing has changed the way of delivering computing services. Cloud providers can offer Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS) to the end-user at low cost. Cloud characteristics such as flexibility, reliability, and cost-effectiveness are the major benefits of moving business to the cloud. Many recent cloud computing issues are discussed in (Botta, de Donato, Persico, & Pescapé, 2016), (Skourletopoulos et al., 2017a), and (Singh, Jeong, & Park, 2016). The work in (Botta et al., 2016) discusses the challenges to integrate Internet of Things (IoT) applications and cloud computing. The authors in (Al Ridhawi, Aloqaily, Kantarci, Jararweh, & Mouftah, 2018) introduce a service provision scheme to provide continuous availability of diversified cloud services targeting vehicular cloud users. The study in (Skourletopoulos et al., 2017a) shows current big data techniques and models that exploit cloud technologies. Singh et al. present a comprehensive (Singh et al., 2016) present a comprehensive review of the security challenges in cloud computing.

In order to minimize latency and enhance the performance of the offered services, public cloud providers have expanded their global infrastructure. They group their cloud data centers within a given geographic area called a region. Each region may contain multiple cloud data centers from different cloud providers. For example, IBM deploys 60 cloud data centers within 6 regions (IBM, n.d.). Each data center service has a different price because of different business expenses such as energy, carbon penalties, real estate taxes and operating costs in each data center location. For example, an instance of m4.xlarge on-demand Windows EC2 in the AWS U.S. East region costs $0.404 per hour. The same instance in the Asia Pacific (Singapore) region costs $0.455 (Computing, 2017). This cost difference becomes more significant for a large number of instances over longer periods. Therefore, the cost estimation of utilization of these large-scale systems is not easy to ascertain.

Mobile cloud computing is a new paradigm that appeared from merging cloud computing and mobile computing (Skourletopoulos et al., 2017b). It allows mobile users to utilize cloud services on demand. It is envisioned that this paradigm will help to overcome the limitations of the mobile device's hardware. In (Lewis & Lago, 2015), the authors have proposed a taxonomy of mobile cloud computing based on the key issues and how they have been tackled in research. One of the key issues is job offloading which consists of migration of jobs (data or code) from resource-constrained mobile devices to the resource-rich cloud.

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2024)
Volume 13: 1 Issue (2023)
Volume 12: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing