Kumputer & Masyarakat

A survey of mobile cloud computing: architecture, applications, and approaches

AUTHOR :
Hoang T. Dinh
Chonho Lee
Dusit Niyato
Ping Wang
URL : https://onlinelibrary.wiley.com/doi/full/10.1002/wcm.1203

1. INTRODUCTION

This paper gives a survey of MCC, which helps general readers have an overview of the MCC including the definition, architecture, and applications. The issues, existing solutions, and approaches are presented. In addition, the future research directions of MCC are discussed. Copyright © 2011 John Wiley & Sons, Ltd.

This paper presents a comprehensive survey on MCC. Section 2 provides a brief overview of MCC including definition, architecture, and its advantages. Section 3 discusses the use of MCC in various applications. Then, Section 4 presents several issues that arise in MCC and approaches to address the issues. Next, the future research directions are outlined in Section 5. Finally, we summarize and conclude the survey in Section 6. The list of acronyms appeared in this paper is given in Table 1.

Table 1. Acronyms

4G

Fourth Generation

AAA

Authentication, Authorization, Accounting

APDV

Application Protocol Data Unit

API

Application Programing Interface

ARM

Advanced RISC Machine

AV

Antivirus

B2B

Business to Business

B2C

Business to Customer

BTS

Base Transceiver Station

CC

Cloud Computing

CSP

Cloud Service Provider

EC2

Elastic Compute Cloud

GPS

Global Positioning System

HA

Home Agent

IaaS

Infrastructure as a Service

IA

Integrated Authenticated

ID

Identifier

IMERA

French acronym for Mobile Interaction in Augmented Reality Environment

ISP

Internet Service Provider

IRNA

Intelligent Radio Network Access

JME

Java ME, a Java platform

LBS

Location Base Service

LTE

Long Term Evolution

LTS

Location Trusted Server

MAUI

Memory Arithmetic Unit and Interface

MC

Mobile Computing

MCC

Mobile Cloud Computing

MDP

Markov Decision Process

MSC

Mobile Service Cloud

P2P

Peer‐to‐Peer

PaaS

Platform as a Service

QoS

Quality of Service

RACE

Resource‐Aware Collaborative Execution

REST

Repretational State Transfer

RFS

Random File System

RTP

Real‐time Transport Protocol

S3

Simple Storage Service

SaaS

Software as a Service

TCC

Truster Crypto Coprocessor

URI

Uniform Resource Identifier

2 OVERVIEW OF MOBILE CLOUD COMPUTING

 This section provides an overview of MCC including definition, architecture, and advantages of MCC. 

2.1 What is mobile cloud computing?

The MCC forum defines MCC as follows 4: 

‘Mobile cloud computing at its simplest, refers to an infrastructure where both the data storage and data processing happen outside of the mobile device. Mobile cloud applications move the computing power and data storage away from mobile phones and into the cloud, bringing applications and MC to not just smartphone users but a much broader range of mobile subscribers’.

Aepona 5 describes MCC as a new paradigm for mobile applications whereby the data processing and storage are moved from the mobile device to powerful and centralized computing platforms located in clouds. These centralized applications are then accessed over the wireless connection based on a thin native client or web browser on the mobile devices. 

Alternatively, MCC can be defined as a combination of mobile web and CC 6, 7, which is the most popular tool for mobile users to access applications and services on the Internet. 

Briefly, MCC provides mobile users with the data processing and storage services in clouds. The mobile devices do not need a powerful configuration (e.g., CPU speed and memory capacity) because all the complicated computing modules can be processed in the clouds.

2.2 Architectures of mobile cloud computing

From the concept of MCC, the general architecture of MCC can be shown in Figure 1. In Figure 1, mobile devices are connected to the mobile networks via base stations (e.g., base transceiver station, access point, or satellite) that establish and control the connections (air links) and functional interfaces between the networks and mobile devices. Mobile users' requests and information (e.g., ID and location) are transmitted to the central processors that are connected to servers providing mobile network services. Here, mobile network operators can provide services to mobile users as authentication, authorization, and accounting based on the home agent and subscribers' data stored in databases. After that, the subscribers' requests are delivered to a cloud through the Internet. In the cloud, cloud controllers process the requests to provide mobile users with the corresponding cloud services. These services are developed with the concepts of utility computing, virtualization, and service‐oriented architecture (e.g., web, application, and database servers). 

 

This architecture is commonly used to demonstrate the effectiveness of the CC model in terms of meeting the user's requirements 12. 

Generally, a CC is a large‐scale distributed network system implemented based on a number of servers in data centers. The cloud services are generally classified based on a layer concept (Figure 2). In the upper layers of this paradigm, Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS) are stacked. 

• Data centers layer. This layer provides the hardware facility and infrastructure for clouds. In data center layer, a number of servers are linked with high‐speed networks to provide services for customers. Typically, data centers are built in less populated places, with a high power supply stability and a low risk of disaster. 
• IaaS. Infrastructure as a Service is built on top of the data center layer. IaaS enables the provision of storage, hardware, servers, and networking components. The client typically pays on a per‐use basis. Thus, clients can save cost as the payment is only based on how much resource they really use. Infrastructure can be expanded or shrunk dynamically as needed. The examples of IaaS are Amazon Elastic Cloud Computing and Simple Storage Service (S3).
•  PaaS. Platform as a Service offers an advanced integrated environment for building, testing, and deploying custom applications. The examples of PaaS are Google App Engine, Microsoft Azure, and Amazon Map Reduce/Simple Storage Service. 
• SaaS. Software as a Service supports a software distribution with specific requirements. In this layer, the users can access an application and information remotely via the Internet and pay only for that they use. Salesforce is one of the pioneers in providing this service model. Microsoft's Live Mesh also allows sharing files and folders across multiple devices simultaneously.

Although the CC architecture can be divided into four layers as shown in Figure 2, it does not mean that the top layer must be built on the layer directly below it. For example, the SaaS application can be deployed directly on IaaS, instead of PaaS.

2.3 Advantages of mobile cloud computing

1. Extending battery lifetime.Computation offloading technique is proposed with the objective to migrate the large computations and complex processing from resource‐limited devices (i.e., mobile devices) to resourceful machines (i.e., servers in clouds). This avoids taking a long application execution time on mobile devices which results in large amount of power consumption.

Rudenko et al. 18 and Smailagic and Ettus 19 evaluate the effectiveness of offloading techniques through several experiments. The results demonstrate that the remote application execution can save energy significantly.  

2. Improving data storage capacity and processing power. 

3. Improving reliability. For example, the cloud can be used to protect copyrighted digital contents (e.g., video, clip, and music) from being abused and unauthorized distribution. Also, the cloud can remotely provide to mobile users with security services such as virus scanning, malicious code detection, and authentication 31. Also, such cloud‐based security services can make efficient use of the collected record from different users to improve the effectiveness of the services. 

 

In addition, MCC also inherits some advantages of clouds for mobile services as follows: 

• Dynamic provisioning. Dynamic on‐demand provisioning of resources on a fine‐grained, self‐service basis is a flexible way for service providers and mobile users to run their applications without advanced reservation of resources.
• Scalability. The deployment of mobile applications can be performed and scaled to meet the unpredictable user demands due to flexible resource provisioning. Service providers can easily add and expand an application and service without or with little constraint on the resource usage.
• Multitenancy. Service providers (e.g., network operator and data center owner) can share the resources and costs to support a variety of applications and large number of users.
• Ease of integration. Multiple services from different service providers can be integrated easily through the cloud and Internet to meet the user demand.

3.  ISSUES AND APPROACHES OF MOBILE CLOUD COMPUTING

 3.1  Issues in mobile communication side

1. Low bandwidth. Bandwidth is one of the big issues in MCC because the radio resource for wireless networks is much scarce as compared with the traditional wired networks.
Jin and Kwok 63 proposes a solution to share the limited bandwidth among mobile users who are located in the same area (e.g., a workplace, a station, and a stadium) and involved in the same content (e.g., a video file). However, the proposed solution is only applied in the case when the users in a certain area are interested in the same contents. Also, it does not consider a distribution policy (e.g., who receives how much and which part of contents) which leads to a lack of fairness about each user's contribution to a coalition.  

2. Availability. Service availability becomes a more important issue in MCC than that in the CC with wired networks. Mobile users may not be able to connect to the cloud to obtain a service due to traffic congestion, network failures, and the out‐of‐signal. 

3. Heterogeneity. Mobile cloud computing will be used in the highly heterogeneous networks in terms of wireless network interfaces. Different mobile nodes access to the cloud through different radio access technologies such as WCDMA, GPRS, WiMAX, CDMA2000, and WLAN. As a result, an issue of how to handle the wireless connectivity while satisfying MCC's requirements arises (e.g., always‐on connectivity, on‐demand scalability of wireless connectivity, and the energy efficiency of mobile devices).  

 

Context management architecture introduced in 67.  

3.2 Issues in computing side

1. Computing offloading. Offloading is one of the main features of MCC to improve the battery lifetime for the mobile devices and to increase the performance of applications. However, there are many related issues including efficient and dynamic offloading under environment changes.  

a. Offloading in the static environment. Experiments in 18 show that offloading is not always the effective way to save energy. For a code compilation, offloading might consume more energy than that of local processing when the size of codes is small. For example, when the size of altered codes after compilation is 500 KB, offloading consumes about 5% of a device's battery for its communication, whereas the local processing consumes about 10% of the battery for its computation. In this case, the offloading can save the battery up to 50%. However, when the size of altered codes is 250 KB, the efficiency reduces to 30%. When the size of altered codes is small, the offloading consumes more battery than that of local processing.  

b. Offloading in the dynamic environment. This subsection introduces few approaches to deal with offloading in a dynamic network environment (e.g., changing connection status and bandwidth). The environment changes can cause additional problems. For example, the transmitted data may not reach the destination, or the data executed on the server will be lost when it has to be returned to the sender.

Ou et al. 79 analyzes the performance of offloading systems operating in wireless environments. In this work, the authors take into account three circumstances of executing an application, thereby estimating the efficiency of offloading. They are the cases when the application is performed locally (without offloading), performed in ideal offloading systems (without failures), and performed with the presence of offloading and failure recoveries. In the last case, when a failure occurs, the application will be re‐offloaded. This approach only re‐offloads the failed subtasks, thereby improving the execution time. However, this solution has some limitations. That is, the mobile environment is considered as a wireless ad hoc local area network (i.e., broadband connectivity is not supported). Also, during offloading execution, a disconnection of a mobile device is treated as a failure. 

1. Security. Protecting user privacy and data/application secrecy from adversary is a key to establish and maintain consumers' trust in the mobile platform, especially in MCC. In the following, the security‐related issues in MCC are introduced in two categories: the security for mobile users and the security for data. Also, some solutions to address these issues are reviewed. 

a. Security for mobile users. Mobile devices such as cellular phone, personal digital assistant (PDA), and smartphone are exposed to numerous security threats like malicious codes (e.g., virus, worm, and Trojan horses) and their vulnerability. In addition, with mobile phones integrated global positioning system (GPS) device, they can cause privacy issues for subscribers. Two main issues are as follows:

• Security for mobile applications. Installing and running security softwares such as Kaspersky, McAfee, and AVG antivirus programs on mobile devices are the simplest ways to detect security threats (e.g., virus, worms, and malicious codes) on the devices. However, mobile devices are constrained in their processing and power; protecting them from the threats is more difficult than that for resourceful device (e.g., PC). For example, it is impossible to keep running the virus detection software on mobile devices. Oberheide et al. 31 presents an approach to move the threat detection capabilities to clouds. This paradigm is an extension of the existing cloud AV platform that provides an in‐cloud service for malware detection.The platform consists of host agent and network service components 82 83. Host agent is a lightweight process that runs on mobile devices, and its function is to inspect the file activity on a system (i.e., it is similar to the function of antivirus software). If an identified file is not available in a cache of previous analyzed files, this file will be sent to the in‐cloud network service for verification. The second major component of CloudAV is a network service that is responsible for file verification. The network service will determine whether a file is malicious or not. The most advantage of this solution is that moving the detection capabilities to a network service enables the use of multiple antivirus engines in parallel by hosting them in virtualized containers. However, to apply CloudAV platform for the mobile environment, a mobile agent should be improved and customized to fit in the mobile devices. Oberheide et al.
• Privacy. With the advantages of GPS positioning devices, the number of mobile users using the location based services (LBS) increases. However, the LBS faces a privacy issue when mobile users provide private information such as their current location. This problem becomes even worse if an adversary knows the user's important information. Location trusted server (LTS) 86 is presented to address this issue. As shown in Figure 5, after receiving the mobile users' requests, LTS gathers their location information in a certain area and cloaks the information called ‘cloaked region’ based on a ‘k‐anonymity’ concept 87 to conceal the user's information. The ‘cloaked region’ is sent to LBS, so LBS knows only general information about the users but cannot identify them. Wang and Wang 88 point out the problem that if LTS reveals the users' information, or if LTS colludes with LBS, the users' information will be in danger. The authors propose to generate the ‘cloaked region’ on mobile devices based on Casper cloaking algorithm 89. Meanwhile, gathering the information of other users around the sender will be done on the cloud to reduce cost and improve speed and scalability. When launching the program on the sender's mobile device, the program will require the cloud to provide information about surrounding users. After that, the mobile client will generate the ‘cloaked region’ by itself and send ‘cloaked region’ to the LBS. In this way, both LTS and LBS cannot know the sender's information.

 b. Securing data on clouds. Although both mobile users and application developers benefit from storing a large amount of data/applications on a cloud, they should be careful of dealing with the data/applications in terms of their integrity, authentication, and digital rights. The data‐related issues in MCC are as follows:

·         Integrity. Several solutions are proposed to address this issue (e.g., 90 91). However, such solutions do not take the energy consumption of mobile users into account. Itani et al. 92 considers the energy consumption issue. This scheme consists of three main components: a mobile client, a cloud storage service, and a trusted third party. The scheme performs three phases: the initialization, update, and verification. In the first phase, files (Fx) that need to be sent to the cloud will be assigned with a message authentication code urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0001. These MACFx will be stored locally, while the files will be sent and stored on the cloud. In the update phase, a case when a user wants to insert the data into file (Fx) is considered. The cloud then sends file (Fx) to this user. At the same time, the cloud also sends a requirement to the trusted crypto coprocessor (TCC) to generate urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0002. TCC then sends urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0003 to the client to verify Fx by comparing it with urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0004. If everything is properly authenticated, the user can insert/delete data. Finally, the mobile client can request the integrity verification of a file, collection of files, or the whole file system stored in the cloud. This phase starts when the user sends a requirement to verify integrity of files to TCC. TCC then retrieves files that need to be checked from the cloud and generates urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0005 to send to the client. The client only compares the received urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0006 and urn:x-wiley:15308669:media:wcm1203:wcm1203-math-0007 that are stored on its device to verify the integrity of such files. This approach not only verifies the integrity of data but also saves energy for the device and bandwidth for the communication network. The reason is that checking and verification are processed on TCC and the client just runs a simple code for comparison. The result shows that this solution can save 90% processing requirements, thus saving significant energy for mobile device.

• Authentication. Chow 93 presents an authentication method using CC to secure the data access suitable for mobile environments. This scheme combines TrustCube 94 and implicit authentication 95 96 to authenticate the mobile clients. TrustCube is a policy‐based cloud authentication platform using the open standards, and it supports the integration of various authentication methods. The authors build an implicit authentication system using mobile data (e.g., calling logs, SMS messages, website accesses, and location) for existing mobile environment. The system requires input constraints that make it difficult for mobile users to use complex passwords. As a result, this often leads to the use of simple and short passwords or personal identification numbers (PINs). Figure 6 shows the system architecture and how the system secures mobile users' access. When a web server receives a request from a mobile client, the web server redirects the request to the integrated authenticated (IA) service along with the details of the request. The IA service retrieves the policy for the access request, extracts the information that needs to be collected, and sends an inquiry to the IA server through the trusted network connect protocol. The IA server receives the inquiry, generates a report, and sends it back to the IA service. After that, the IA service applies the authentication rule in the policy and determines the authentication result (whether or not the mobile client is authenticated successfully for the access request) and sends the authentication result back to the web server. Based on the authentication result, the web server either provides the service or denies the request. 
• Digital rights management. The unstructured digital contents (e.g., video, image, audio, and e‐book) have often been pirated and illegally distributed. Protecting these contents from illegal access is of crucial importance to the content providers in MCC like traditional CC and peer‐to‐peer networks. Zou et al. 30 proposes Phosphor, a cloud‐based mobile digital rights management (DRM) scheme with a subscriber identity module (SIM) card in mobile phone to improve the flexibility and reduce the vulnerability of its security at a low cost. The authors design a license state word (LSW) located in a SIM card and the LSW protocol based on the application protocol data unit (APDU) command. In addition, the cloud‐based DRM with an efficient unstructured data management service can meet the performance requirements with high elasticity. Thus, when a mobile user receives the encrypted data (e.g., video stream) from the content server via real‐time transport protocol, he/she uses the decryption key from a SIM card via APDU command to decode. If the decoding is successful, the mobile user can watch this video on his/her phone. The drawback of this solution is that it is still based on the SIM card of the mobile phone; so, it cannot be applied for other kinds of access; that is, a laptop using WiFi to access these contents.

3. Enhancing the efficiency of data access. With an increasing number of cloud services, the demand of accessing data resources (e.g., image, files, and documents) on the cloud increases. As a result, a method to deal with (i.e., store, manage, and access) data resources on clouds becomes a significant challenge. However, handling the data resources on clouds is not an easy problem because of the low bandwidth, mobility, and the limitation of resource capacity of mobile devices.

For commercial cloud storage providers (e.g., Amazon S3), every I/O operations (e.g., put, copy, cut, and list) are taken by the cloud provider. The I/O operations are executed at a file‐level in general, so this increases the cost of network communication and service for mobile users. Nam et al. 97 proposes an algorithm in which I/O operations are executed at a block‐level. The algorithm uses log‐structured I/O transaction 98 to minimize the number of the block‐level I/O operations. The main idea here is to allow the cloud storage log‐structure perform write operation with the optimal number of data blocks that adaptively changes with I/O and cloud storage pricing policy. The authors demonstrate that, through experimentation, the proposed solution reduces the total I/O costs considerably up to 54%.

4. Context‐aware mobile cloud services. It is important for the service provider to fulfill mobile users' satisfaction by monitoring their preferences and providing appropriate services to each of the users. A lot of research work try to utilize the local contexts (e.g., data types, network status, device environments, and user preferences) to improve the quality of service (QoS). 

Samimi et al. builds a model, called Mobile Service Clouds (MSCs), which is extended from service clouds paradigm. In this model, when a customer uses a service on the cloud, the user's request firstly goes to a service gateway. The gateway will choose an appropriate primary proxy to meet the requirements (e.g., the shortest path and minimum round‐trip time) and then sends the result to the user. In the case of disconnection, MSCs will establish transient proxies 106 for mobile devices to monitor the service path, and support dynamic reconfiguration (with minimum interruption). The advantages of this model are that the model addresses the disconnection issue and can maintain the QoS at an acceptable level.  

The Coign automatic distributed partitioning system: an application is transformed into a distributed application by inserting the Coign runtime, profiling the instrumented application, and analyzing the profiles to cut the network‐based graph.

Table 3. Common mobile computing environmental changes.

Changes	Priority level	Description
Client side power level	1	Power can be divided into sufficient and insufficient power levels, which will depend on the particular situation.
Connection status	2	The connection status can be faded, disconnected from the mobile network, or reconnected to the mobile network.
Bandwidth	3	The bandwidth varies from time to time and depends on several factors, such as the network traffic condition, and so on.

Figure 5
Overall Architecture of Spatial Cloaking.

 
Figure 6
TrustCube architecture.

Figure 7
Architecture of E‐Recall system.

Figure 8
Random file system architecture.



Figure 9
The VOLARE middleware modules.

4. OPEN ISSUES AND FUTURE RESEARCH DIRECTIONS

Several research works contribute to the development of MCC by tackling issues as presented in the previous section. However, there are still some issues which need to be addressed. This section discusses several open issues and possible research directions in the development of MCC.

4.1 Low bandwidth

Although many researchers propose the optimal and efficient way of bandwidth allocation, the bandwidth limitation is still a big concern because the number of mobile and cloud users is dramatically increasing. We consider that fourth generation (4G) network and Femtocell are emerging as promising technologies that overcome the limitation and bring a revolution in improving bandwidth.

1. 4G network. Fourth generation network is a technology that significantly increases bandwidth capacity for subscribers. 4G network is capable of providing up to 100 Mbit/s (for ‘LTE Advanced’ standard) and 128 Mbit/s (for ‘WirelessMAN‐Advanced’ standard) for mobile users, whereas the current 3G network supports a maximum of 14.4 Mbit/s. Furthermore, 4G network also promises other advantages such as widering mobile coverage area, smoothering quicker handoff, varied services, and so on 109. Nevertheless, 4G wireless networks still have several issues related to network architecture, access protocol, or QoS that are taken into account in 110.
2. Femtocell. Femtocell 111 is a small cellular base station, designed for use in a small area. Hay Systems Ltd (HSL) 112 develops a service to combine femtocells and CC to deliver a highly economical, scalable, and secure network for mobile operators. This allows the resources employed in delivering mobile services over the femtocell network to expand or contract as user demands for services increase or decrease, respectively. The result is a highly economical femtocell network with only sufficient resources being used at any given point, without impacting the ability to immediately scale to meet demands. In this paradigm, femtocells located in homes and offices of users connect via the Internet to the cloud to gain access to their operator's network. Mobile operators connect with the cloud enabling their subscribers to gain access to their network when using a femtocell connected to the cloud. However, 112 just shows that femtocell is practically useful when used with clouds. We need to investigate a uniform standard and performance impact of using femtocells in MCC.

4.2 Network access management

An efficient network access management not only improves link performance for mobile users but also optimizes bandwidth usage. Cognitive radio can be expected as a solution to achieve the wireless access management in mobile communication environment 113. Cognitive radio increases the efficiency of the spectrum utilization significantly, by allowing unlicensed users to access the spectrum allocated to the licensed users. When this technique is integrated into MCC, the spectrum can be utilized more efficiently. The spectrum scarcity can be solved and thus millions of dollars for network providers can be saved 114. However, cognitive radio is defined as wireless communication technology in which each node communicates via an optimal wireless system based on recognition of radio resource availability in heterogeneous wireless communication environment. Therefore, mobile users in MCC must be able to detect this radio resource availability (through spectrum sensing) while ensuring that the traditional services will not be interfered.

4.3 Quality of service

In MCC, mobile users need to access to servers located in a cloud when requesting services and resources in the cloud. However, the mobile users may face some problems such as congestion due to the limitation of wireless bandwidths, network disconnection, and the signal attenuation caused by mobile users' mobility. They cause delays when the users want to communicate with the cloud, so QoS is reduced significantly. Two new research directions are CloneCloud and Cloudlets that are expected to reduce the network delay.

1. CloneCloud. CloneCloud brings the power of CC to your smartphones 115. CloneCloud uses nearby computers or data centers to increase the speed of running smartphone applications. The idea is to clone the entire set of data and applications from the smartphone onto the cloud and to selectively execute some operations on the clones, re‐integrating the results back into the smartphone. One can have multiple clones for the same smartphone, and clones pretend to be more powerful smartphones, and so on. CloneCloud is limited in some respects by its inability to migrate native state and to export unique native resources remotely. A related limitation is that CloneCloud does not virtualize access to native resources that are not virtualized already and are not available on the clone.
2. Cloudlets. A cloudlet is a trusted, resource‐rich computer or cluster of computers which is well‐connected to the Internet and available for use by nearby mobile devices. Thus, when mobile devices do not want to offload to the cloud (maybe due to delay and cost), they can find and use a nearby cloudlet. In this way, mobile users may meet the demand for real‐time interactive response by low‐latency, one‐hop, high‐bandwidth wireless access to the cloudlet. If no cloudlet is available nearby, the mobile device may refer to the default mode that will send requirements to a distant cloud, or in the worst case, solely its own resources. Satyanarayanan et al. 116 builds an architecture through exploiting a virtual machine technology to rapidly instantiate customized service software on a nearby cloudlet and then uses that service over a wireless local area network. This technology can help mobile users overcome the limitations of CC due to wide area network latency and low bandwidth. However, there are some considerations that need to be addressed before this idea can be applied widely in practical system. For example, how to distribute processing, storage, and networking capacity for each cloudlet? How to manage policies for cloudlet providers to maximize user experience while minimizing cost? Also, trust and security for cloudlet are other issues in implementing this idea because adversaries can create a fake cloudlet to steal the user's information.

4.4 Pricing

Using services in MCC involves both mobile service provider (MSP) and cloud service provider (CSP). However, MSPs and CSPs have different services management, customers management, methods of payment, and prices. Therefore, this will lead to many issues; that is, how to set price, how the price will be divided among different entities, and how the customers pay. For example, when a mobile user runs mobile gaming application on the cloud, this involves the game service provider (providing a game license), mobile service provider (accessing the data through base station), and CSP (running game engine on a data center). The price paid by the game player has to be divided among these three entities such that all of them are satisfied with the division. It is clear that the business model including pricing and revenue sharing has to be carefully developed for MCC.

4.5 Standard interface

Interoperability becomes an important issue when mobile users need to interact and communicate with the cloud. The current interface between mobile users and cloud are mostly based on the web interfaces. However, using web interfaces may not be the best option. First, web interface is not specifically designed for mobile devices. Therefore, web interface may have more overhead. Also, compatibility among devices for web interface could be an issue. In this case, the standard protocol, signaling, and interface for interacting between mobile users and cloud would be required to ensure seamless services. In the future, HTML5 is expected as a promising technique to address this issue. HTML5 WebSockets offer an interface. However, an extensive performance evaluation and feasibility study have to be performed to ensure that it will work in MCC efficiently.

5.6 Service convergence

The development and competition of CSPs can lead to the fact that in the near future, these services will be differentiated according to the types, cost, availability and quality. Moreover, in some cases, a single cloud is not enough to meet the mobile user's demands. Therefore, the new scheme is needed in which the mobile users can utilize multiple clouds in a unified fashion. In this case, the scheme should be able to automatically discover and compose services for user. One of the potential solutions of this issue is the sky computing, which will be the next step of cloud computing. Sky computing is a computing model where resources from multiple cloud providers are leveraged to create a large scale distributed infrastructure 117. Similarly, the mobile sky computing will enable the providers to support a cross‐cloud communication and enable users to implement mobile services and applications. However, to offer a service to mobile user in a unified way, the service integration (i.e., convergence) would need to be explored.