Secure Code Offloading in Mobile Cloud Computing: A Review
Anoop S Dr.J Amar Pratap Singh
Asst.Professor,College of Engg,Perumon and Research Guide and Director
Research Scholar,Noorul Islam Centre for Higher Education,Thuckalay. Noorul Islam Centre for Higher Education,Thuckalay.
Abstract—Mobile technology have made smartphones a future venture in computing and service access devices. Users are running computational applications on Smart Mobile Devices (SMDs). However in spite of all these advancements they are still having many limitations. Mobile Cloud Computing (MCC) is an innovative solution for maintaining this inefficiency by collaborating services and resources of computational clouds to devices on demand basis. It integrates mobile computing and Cloud Computing to extend capabilities of mobile devices using offloading techniques.Security and privacy of offloaded data is a key factor. The paper reviews the current research trends in mobile cloud computing offloading and its significance. This paper also emphasis on the security aspects and design issues related to in developing and executing computational mobile applications within the domain.
Index Terms:-Cloud Computing, Mobile Cloud Computing, Offloading Methods, Security
Introduction
Cloud computing is a form of distributed computing model which implements the utility computing vision [1][13] services that is provided on demand basis. The objective of the cloud computing model is to make an increased capability for the client devices by accessing to resources on leased infrastructure and software applications instead buying them using high cost. Cloud computing had given an introduction to a new era in information and services there by making a collaboration among various handheld devices. Cloud enabled to create an online social network infrastructure in which people share data and thoughts to build research communities. As mobiles are becoming inevitable part in our human life, all people are doing their computation task on mobiles rather depending on traditional devices. With this an over task of computation is happening on mobile devices. They are converting from the role of telephones to smart phones. Due to this high usage paradigm, there is a chance of reducing performance and efficiency. By embedding them to cloud and migrating the high task applications to run on cloud makes it easier to reduce the same. So there comes the role of cloud offloading and its importance to understand the technologies which completely offload data on to cloud and provide a way of shifting the high application part on to the cloud to reduce the overhead on the mobile device. Cloud computing provides an on-Demand Service mechanism by which its customers can do and retrieve their services automatically with the help of internet. The users can access heterogeneous data from various sits there by increasing their improved accessibility. It provides rapid elasticity and measured usage to the end users.
Mobile Cloud Computing[2] is used where all the resources are provided and retrieved from the cloud it takes the information from handheld devices and do its processing within the devices, stores the result which is then given back them, saving the resources of mobile devices and its battery. It describes an architectural framework of how computation intensive task must be transferred from mobile devices to cloud and how cloud computing capabilities can be utilized in mobile devices.
Currently, there are several works and research in mobile cloud which are intended to enhance functionalities of resource starving devices by providing mobile clients get connected to cloud platform, software, and computing services. For example, Amazon web services have a Simple Storage Service to protect their user profile and data.
[5]. In addition, there are several frameworks that allow to process data intensive tasks remotely on cloud servers. In ASM computation offloading framework [6] Computation offloading reduced the energy consumption cost of mobile devices by 33%.
Computation offloading is the process of transferring high energy intensive computation application components to a remote cloud from where it can be processed. A large number of computation offloading frameworks have been proposed with a variety of approaches for reducing computation overhead on mobile devices. These tasks are identified and partitioned based on different granularity levels and the applications are offloaded to remote servers for execution there by extending capabilities. But still, the computation offloading mechanisms still faces different levels of challenges.
Offloading Process is a step by step mechanism in deciding whether and which applications to b send to the remote server. It requires a classification process regarding which process to be run remotely and which applications to be retained in the mobile devices. That is the first step is to identify the offloadable and non-offloadable components in the mobile devices. The decision parameters may change according to the applications. In the next step the identification of the remote server and the identification of codes which are to be send must be selected. On the last step the offloading decision must be taken and must identify whether offloading to be done in runtime.
Security Issues and Aspects in MCC
Security of data transmission is a key concern in cloud based application processing. Security and privacy are two important concepts that need to be maintained during the offloading process. These concepts can be evaluated from different aspects: (a) Mobile device, (b) cloud data centers, and (c) at the time of data transmission over the network. Besides all the technologies, there is a great increase in the number of attacks on mobile devices which become the main aim for attackers. Threats happen in cloud data center occurs when a data is transferred between different nodes over the network. Thus, excellent levels of security are expected by both the mobile clients and the cloud providers. Binary transfer of mobile codes[13],[14],[15],[16] are subjected to vulnerability Despite the available solutions, strong measures and a secure environment are required overcoming the security breaches in mobile cloud computing.
1) Data confidentiality. Confidentiality defines the protection of data from an unauthorized disclosure. The host that an agent created can only obtain the data that are computed from other hosts.
2) Nonrepudiation. Nonrepudiation provides protection against denial by one of the entities involved in a communication. Therefore, no host that an agent has moved on to can deny the data results that is computed in them and the agent’s passing through.
3) Integrity. Integrity is an assurance that the data that are received are exactly as calculated and sent by the host that the agent has moved on to. Therefore, an intermediate host can modify, insert, and delete previous hosts’ data.
4) Authentication. Authentication provides the assurance that the source of the received data is as claimed. Therefore, any host where the agent migrates to should authenticate the data that are computed on its platform.
5) Malicious host identification. The originator can identify the malicious host by verifying the chain of encapsulated offers.
6) Forward privacy. The originator can only extract the visited host’s integrity.
7) Truncation resilience. The chain of encapsulated offers cannot be broken in between due to the existence of two or more malicious hosts.
In this paper review the different approaches and challenges in offloading sensitive data in cloud and its security aspects.
Literature survey
K. Kumar et al.[3] on his paper that mobile cloud offloading will be beneficial for users those who deals with mobile devices. The high intensive computing applications can be executed in clouds rather than in mobiles so that computation must be improved. The energy of mobile devices can be saved by offloading and the applications can run on different machines on cloud platform..
Satyanarayanan et al.[4] proposed a novel architecture which replaces the remote cloud with nearby cloudlets to reduce the latency created due to transmission. In this paper he designed an infrastructure which provides security and privacy towards cloud offloading mechanisms. A cloudlet can be defined as a hosting environment for offloaded tasks that is deployed to remote machines as different as individual servers systems. Cloudlets are virtual-machine (VM) based on support scalability, mobility, and elasticity. They are located in single-hop nearness to mobile devices.
MAUI [7] is a framework that illustrates that the main aim of smartphones is to reduce energy consumption of high task intensive applications with the help of offloading process. MAUI defines a dynamic offloading framework by understanding the continuous profiling process. The framework is transparent and it encapsulates the complexity in running a remote execution application from the end user and gives awareness to user that the whole application is processing in the mobile device.
Chun et al. present in [6] the Clone Cloud framework with the purpose of improving e the battery life and performance on the mobile device by transferring high battery draining components to cloud servers. It uses a program profiling method for the analysis of the offloadable components and other sensing mechanisms are run locally. It uses static analysis to identify the constraints and build a cost model for execution. A clone of smart phone software is stored in the cloud server and at the run time its makes the offloading mechanism by the migration of threads. A virtual machine instance migration is happening by the object passing methods. There is a partition configuration file which keeps track of current execution conditions such as bandwidth and other cloud resources. Thus the clone cloud provides a dynamic offloading strategy.
Xia et al. present in [8] a computation offloading framework called Phone2Cloud. The purpose was to enhance energy efficiency of mobile devices and improve the application’s performance. A fully quantitative analysis on energy saving mechanisms was conducted based on application and scenario experiments. Phone2Cloud provided a semi-automatic offloading framework. It was designed in such a way that it has to be manually modified for running applications on the cloud and to receive the results. The user’s delay-tolerance threshold was the factor to identify the offloading decision in a static analysis manner. The average power consumption and the execution time was the factor being considered for making an offloading decision by a decision engine.
Zhao et al. in a paper [9] discussed about the mirror server framework used by Telecommunication Service Provider (TSP). A TSP is a type of communication service provider which is used to enable voice communication services .Mirror server extends computation offloading, security, and storage to the mobile devices. Mirror server uses a server with VM templates on heterogeneous platforms. No partition of offloaded data is required but in the first stage a virtual machine instance must be created which act as a mirror server taking care of managing and deploying a computing infrastructure in the network. It provides an optimized mechanism of offloading.
Al-Hamadi et al. [10] proposes a system with the Kerberos protocol which is based on a symmetric key. The data are which are to be transferred are encrypted by generating a symmetric key from the hash function that is present in the server. The ticket-granting service issues the key, which can be only accessed by the client and the server. This protocol provides confidentiality, integrity, authentication and authorization. The main drawback is that since the keys are shared, large data transfer will become risky.
Kemp et al. present in [17] the Cuckoo framework for computation offloading for smartphones. This framework uses a java stub model to do offloading mobile device applications onto a cloud server. Cuckoo’s aims to enhance device performance and reduce battery usage. The framework was integrated with Eclipse development platform tool with the open source Android framework.
Challenges and Broader perspective
Offloading is not suitable for every mobile application, but when an application uses complex or time consuming algorithms such as recursion, by offloading those parts into the cloud, time and energy consumption are reduced so that the local execution time is reduced. The important aspect is identification of code that needs to be offloaded. Proper classification algorithms must be needed to evaluate the constraints and it must improve the overall efficiency of the system. A decision maker is necessary for the proper implementation of offloading.Decsicion maker keeps track of the constraints, analyze the environment and make a decision. A proper planning based on feasibility study, requirements analysis and migration strategy is to be done. Then an execution scenario is required which enables data extraction, modification and transformation of offloaded data. After that an evaluation based on these must be done for testing and validation purposes.
A security aspect is needed by a cloud service provider to ensure that malicious attackers are not attacking the cloud infrastructure or host malicious software. Security in MCC is important in the following perspectives: security for mobile devices, security for data transmission over the wireless medium and security in the cloud datacenter nodes Every user has to have the security functionalities which must forms part of the security baseline that offers basic security guarantees at any mode of operation. However there will be others who would needs additional security services from the service provider to satisfy their requirements as well as to protect themselves from other malicious attackers. For that purpose a good security model which meets the security requirements is required to maintain the integrity and privacy of the users. A good service level agreement policy between both is needed to be maintained for better security concerns. Since offloaded data is located at the cloud server ,if the server is breached then sensitive data will be available to the attackers. A trust based system[11] using symmetric key mechanism plays a key role in maintaining privacy. A proper use of encryption algorithms must be applied to offloaded data to ensure security.Even these algorithms also can be compromised, the relevance of using steganography methods provide a way to overcome security attacks when offloading sensitive data. Steganography is to hide data before sending them to servers so that unauthorized access of data can be prevented.Steganography hides data so that the server is unaware of the existence of information. Image processing is computation-intensive and a good candidate for offloading.With the help of a cover image and data image ,a new stego image can be created and can be evaluated based on further processing.
Conclusion
In mobile clouds,the computation intensive tasks ruin the energy efficiency and battery life.The code offloading technique provides a way to reduce the local execution time by pushing the computation to the remote cloud server. It helps to lower the CPU load on mobile devices there by saving lots of energy and improves battery time.Offloading components and its security are key challenges that must be analyzed and taken care of.The method of encryption and data hiding improves the security and privacy.Thus this work focuses on the challenges and research aspects of mobile cloud offloading and its security aspects.
References
[1]http://en.wikipedia.org/wiki/Computer_vision
[2]Hoang T. Dinh, Chonho Lee, Dusit Niyato, and Ping Wang” A Survey of Mobile Cloud Computing: Architecture, Applications, and Approaches” Wiley,pp. 1-38.
[3]K. Kumar and Y.-H. Lu, “Cloud computing for mobile users:Can offloading computation save energy?”Computer, vol. 43, pp. 51–56,2010.
[4]M. Satyanarayanan, P. Bahl, R. Caceres, and N. Davies, “The case for vm-based cloudlets in mobile computing,” IEEE Pervasive Computing ,vol. 8, pp. 14–23, 2009.
[5]S.Mathe,Overview of Amazon web Service,Amazon WhitePapers,2014.
[6] B.-G. Chun, S. Ihm, P. Maniatis, M. Naik, A. Patti, CloneCloud: elastic execution between mobile device and cloud, in: Proceedings of the Sixth Conference on Computer Systems, ACM, 2011, pp. 301–314.
[7]E. Cuervo, A. Balasubramanian, D.-K. Cho, A. Wolman, S.Saroiu, R. Chandra, P. Bahl, MAUI: making smartphones last longer with code offload, in: Proceedings of the 8th International Conference on Mobile Systems, Applications, and Services,ACM, 2010, pp. 49–62.
[8] F. Xia, F. Ding, J. Li, X. Kong, L.T. Yang, J. Ma, Phone2cloud:exploiting computation offloading for energy saving on smartphones in mobile cloud computing, Inform. Syst. Front.16 (1) (2014) 95–111.
[9]B. Zhao, Z. Xu, C. Chi, S. Zhu, G. Cao, Mirroring smartphones for good: a feasibility study, in: Mobile and Ubiquitous Systems:Computing, Networking, and Services, Springer, 2010, pp. 26–38.
[10] H. M. N Al-Hamadi, C. Y. Yeun, M. J. Zemerly, and M. Al-Qutayri,“Distributed lightweight kerberos protocol for mobile agent systems,” in Proc. IEEE GCC GCC, 2011, vol. 19, pp. 233–236, No. 2.
[11]G. Geetha and C. Jayakumar, “Trust enhanced data security in free roaming mobile agents using symmetric key cryptography,” Int. J. Netw. Sec.Appl., vol. 3, no. 5, pp. 217–228, 2011
[12] M. Armbrust, A. Fox, R. Griffith, A.D. Joseph, R. Katz, A.Konwinski, G. Lee, D. Patterson, A. Rabkin, I. Stoica, and M.Zaharia, “A View of Cloud Computing,” Comm. the ACM, vol. 53,no. 4, pp. 50-58, 2010.
[13] M. Armbrust, A. Fox, R. Griffith, A.D. Joseph, R. Katz, A.Konwinski, G. Lee, D. Patterson, A. Rabkin, I. Stoica, and M.Zaharia, “A View of Cloud Computing,” Comm. the ACM, vol. 53,no. 4, pp. 50-58, 2010.
[14] Y. Begum and M. Mohamed, “A dht-based process migration policy for mobile clusters,” in Information Technology: New Generations (ITNG),2010 Seventh International Conference on. IEEE, 2010, pp. 934–938.
[15] C. Lai and R. Ko, “Dishes: A distributed shell system designed for ubiquitous computing environment,” International Journal of Computer Networks & Communications, vol. 2, no. 1, pp. 66–83, 2010.
[16] I. Giurgiu, O. Riva, D. Juric, I. Krivulev, and G. Alonso, “Calling the cloud: Enabling mobile phones as interfaces to cloud applications,” Middleware 2009, pp. 83–102, 2009.
[17] R. Kemp, N. Palmer, T. Kielmann, H. Bal, Cuckoo: a computation offloading framework for smartphones, in:Mobile Computing, Applications, and Services, Springer, 2010,pp. 59–79.