The Ontology-based Software Co-Evolution of Cloud and Mobile Computing - Literature review Example

Running Head: The Ontology-Based Software The Ontology-Based Software Co-Evolution of Cloud and Mobile Computing: A Literature Review University Name Subject Name Instructor Name Date of Submission The Ontology-Based Software Co-Evolution of Cloud and Mobile Computing: A Literature Review Software Co-Evolution Since the information and communication technology systems are becoming more and more complex, new technology vendors and developers are appearing in the industry and academia. Experts who design computer based information systems are utilising a number of tricks, techniques, methods and programming constructs with issue and case specific precision. Now when different systems and software applications are required to inter-network and communicate, problems arise due to the cross-platform differences, technical clashes and theoretical ambiguities. (Sheng, 2012; de Oliveira et al, 2010; Kozuch et al, 2004) The issues encountered while synchronising different information systems and IT platforms (as explained above) can be solved by applying the concept of co-evolution. A similar complex situation is being come across while reflecting on the technology migration and orchestration issues between mobile computing and cloud computing. According to Alam and Padenga (2010), the term co-evolution originated in the realm of biological sciences to explain the developmental processes that take place during speciation, mutation and extinction. Analogically, software co-evolution in the realm of computer science signifies the importance of synchronising the different software species across the different IT platforms that have diverse purposes and taxonomies. Co-evolution in the realm of understanding the developmental processes of different software applications and utilities involves robust support and correlation mechanisms between the different source codes, programming environments and conceptual frameworks. For example, co-evolution of two software applications can be analysed by studying and cross-correlating the different portions of their source codes and respective programming constructs. Software, at different levels of abstraction, and the programming language in which it is developed, must be compared with the historic trends seen in the other software suites in that category (Kozuch et al, 2004). For example, a programmer must compare the evolution of MS Office Outlook with that of IBM Lotus Notes to get an idea of how writing email messages can be simplified across the different system platforms and technologies, particularly when cross-platform compatibility issues are encountered between the Microsoft and IBM products. Otherwise, according to Alam and Padenga (2010, p. 70), “Whenever a new version of the (given) programming, modelling, or specification language is provided, it is quite possible that programs that worked perfectly in a previous version of the language fail to function in the new version.” Understanding Cloud Computing Cloud computing is an extensively service oriented technology which can be explained with the help of business intelligence paradigm. From a technical viewpoint, Gens (2008) has described as “an emerging IT development, deployment and delivery model, enabling real-time delivery of products, services and solutions over the Internet (i.e. enabling cloud services)”, as defined by the International Data Consortium. Researchers from InfoWorld, Inc. surveyed through various stakeholders in the realm of cloud computing such as the users, enterprises, vendors and experts to develop a part by part framework of the technology. According to Knorr and Gruman (2009), the research revealed that cloud computing can be roughly divided into seven major elements from a utility based technical perspective. These components are helpful in developing cloud computing taxonomy that can help to standardise research and development parameters in this field. The seven components are SaaS, utility computing, web services support, service oriented platform, managed services, service commerce platforms and Internet integration (Knorr and Gruman, 2009). Cloud Computing: Major Elements and Taxonomy 1. SaaS At this level, cloud computing is operational in the form of a multi-tenant architecture. In conventional sense, it is an application layer interface where the cloud service provider distributes some sort of software application at lower license fees (Lanois, 2011; Knorr and Gruman, 2009). 2. Utility computing At the presentation and session layers of the OSI model, cloud computing is designed to fulfil the functional requirements of distributed computing such as data representation, encryption, inter-host communication, etc. (Shang et al, 2010; Knorr and Gruman, 2009; Gens, 2008) 3. Web services support Following the OSI model, cloud based web services are an element of the overall cloud computing framework that serves the functions of the network layer (Lanois, 2011). In the technology category, utility computing is made available along with the networking facilities (Knorr and Gruman, 2009). 4. Service oriented platforms Experts like Knorr and Gruman (2009) have defined the element of service oriented platform as “another SaaS variation”, and this form of cloud computing can deliver development and/or programming environments as a service. 5. Managed services A managed service is principally an application layer element exposed to the information and communication technology uses rather than to the end-users. Examples include virus scanning and application monitoring services. (Knorr and Gruman, 2009). Experts like de Oliveira have attributed the managed services under taxonomic category of “architecture” in close affinity with the framework of “standards” (de Oliveira et al, 2010, p. 52). 6. Service commerce platforms “A hybrid of SaaS and MSP, this cloud computing service offers a service hub that users interact with” (Knorr and Gruman, 2009). They are mainly widespread in trade environments like the expense management. 7. Internet integration Internet integration is a fundamental functionality of cloud computing which establishes it firmly at the network and transport layers of the OSI model. The basic concept of Internet integration is based on the quest of synchronising the service providers and the service users more coherently using even the unprotected public networks. (Knorr and Gruman, 2009; Zhou et al, 2007) Taxonomy of Cloud Computing de Oliveira et al (2010) have attempted to elucidate the taxonomy of cloud computing. Unlike the analysis of Infoworld (Knorr and Gruman, 2009), de Oliveira et al (2010) have taken up an approach that is based on information and communication technology rather than business systems. One of the key differences between de Oliveira et al (2009) and Knorr and Gruman (2010) is that the former have defined users as client side entities while the latter have defined the users simply as customers. So the main taxonomical assortments under cloud computing are architecture, standards, orientation, business model, technology, access, pricing and privacy. Another interesting approach is the taxonomy adopted by OpenCrowd. It is a more detailed description of the taxonomic categories that can be attributed to the different intricacies of cloud computing. OpenCrowd’s goal is to provide information on cloud services, to create a dialogue between cloud providers and services, consumers and developers, and to promote understanding and adoption of cloud computing solutions (see Figure – 1). (Bauer and Adams, 2012; OpenCrowd, 2010) Figure – 1: Cloud Taxonomy as developed by OpenCrowd (Source: OpenCrowd, 2010; http://www.eci.com/blog/images/11-29-11_views_cloud-tax-lrg.png) Understanding Mobile Computing Mobile computing can be regarded as a well established realm of information and communication technology. Mobile computing has given rise to scalable but larger networks with the help of diverse handheld and transferable computing devices available to the IT customers. Experts such as Satyanarayanan (1995) defined adaptableness to be the most decisive futuristic trait of mobile computing. Mobile computing networks can be constituted with the help vastly diversified range of software applications and hardware platforms. Further in this context, IBM (2013) states: “Mobile Computing is both old and new … Today with the penetration of a new generation of mobile devices such as smartphones, iPads, the emphasis is increasingly shifting to mobile and multi-device user experience, mobile social computing, next generation mobile solutions, mobile cloud services and infrastructure.” Before directly reviewing co-evolution dynamics of the cloud and mobile computing technologies, it is an imperative to find out where mobile computing should be placed in the cloud taxonomy so that correlation and mapping activities can be embarked on from an ontological perspective. Under the methodology adopted by de Oliveira et al (2010, p. 52), mobile computing must be placed in the technical assortment of “access”, which “depends on” the taxonomic category of pricing. Access and pricing have been placed at the same level, while mobile computing is at one of the terminal points of cloud computing concept diagram (see Figure – 2). Figure – 2: Taxonomical position of the mobile technologies and devices under cloud computing. Adopted from de Oliveira et al, 2010, p. 52 Review of the Cloud and the Mobile Ontology Paradigms Before facilitating ontology based co-evolution between cloud and mobile computing technologies, it is an imperative to define and unify the cloud computing ontology itself (see Figure – 3). By interpreting and unifying the different cloud computing systems as constructible services that vary from one vendor to another, a researcher can define a more powerful interaction model among the diverse cloud entities, on both the semantic and functional levels (Youseff, et al, 2008). However, achieving this conceptual commonality has different technical issues. Youseff et al, (2008, p. 7) explain the situation as the following: “Although the cloud providers have deployed their DaaS-systems near the computational infrastructure in order to decrease the latency in transferring the data from its storage to its processing site, many users still fear data leakage.” Figure – 3: Cloud computing ontology proposed by Youseff et al (2008, p. 4) Furthermore, de Oliveira et al (2010) stress on the monitoring issues that are encountered while scrutinising an elaborate cloud computing system. This happens mainly due to the lack of preset universal standards. Moreover, problems have been encountered in mapping the pieces of information, data and variables from one part onto another part of two different cloud platforms (Arizona State University, 2011). In order to overcome the issues in developing a unified ontology for cloud computing, research works of Smr and Novek (2006) can be reviewed. Smr and Novek (2006) advocate direct acquisition of the ontological terms, concepts and constructs and build scientific portals with context aware search tools. This extensive assemblage and organization of complex information can be extended to the sphere of cloud computing too. That will considerably help the researchers to redefine cloud computing too from a more systemic view leading to a unified ontology (Youseff et al, 2008, p. 7). Now, in the course of cloud computing ontology development, experts like de Oliveira et al (2010), Smr and Novek (2006), etc. are emphasising on the need to study other ontological paradigms too that already exist or are emerging in other disciplines of IT. In assessing a co-evolution of cloud computing and mobile technologies, the researchers must be capable of organising the ontological characteristics of both the subject areas. In explaining pervasive computing, Satyanarayanan et al (2009) state that after nearly two decades of continual effort by numerous research experts, the core concepts, methods and mechanisms entailed in mobile computing technology have matured considerably. But it is still a “fast-growing area” (Satyanarayanan et al, 2009, p. 14), where a comprehensive set of cross-platform standards and compatible ontology are still not available. The SPICE Consortium seeks to solve this problem and they took initiative in this direction as early as 2008. SPICE Consortium’s prospective mobile ontology (see Figure – 4) has been structurally divided into sub-ontological categories that cover a diverse range of domains. The novel web ontology language (OWL) is being used to construct the high level mobile ontology. In the OWL executable files that would encapsulate mobile ontology information, SPICE Consortium (2008) claims that the main concepts are being written and executed. “The main concepts to be defined are part of the Mobile Ontology Core and the other sub-ontologies inherit from it. The Mobile Ontology makes also use of existing ontologies to represent concepts like time” (SPICE Consortium, 2008, paragraph 2). Figure – 4: A description of the core ontology constituents (Source: SPICE Consortium, 2008) Technology Migration and Scope for Co-Evolution Technology migration from cloud computing to mobile computing and vice versa can be practically achieved by adopting integration and/or synchronisation based technical strategies for the purpose to culminate at parallel computing or knowledge based “concurrent systems” (Mills and Gomaa, 2002, p. 228). The principal objective here is to find out the simplest and fastest techniques for the same, provided they are also sufficiently secure and inexpensive (Caceres et al, 2005; Clark et al, 2005). Ontology Based System Integration Zhou et al (2007, p. 143) have designed an integration model for the purpose of software migration which the authors call OPTIMA (acronym for “ontology-based platform-specific software migration approach”). Their research is based on the technique of ontology-based platform specific software migration, which can be utilised for cross-platform integration too through performance enhanced “virtual environments” (Koh et al, 2007, p. 200). Zhou et al (2007) further hold that the method of interface specification can simplify the research problem. To facilitate prospective migratory facilities, Zhou et al (2007, pp. 145-146) have devised eight principles for ontology designing. These principles are explained as the following: Principle 1 Instance biased definitions must be used to execute one to one, many to one, and/or one to many mapping. Principle 2 Designing must be done as per specifications of application. Principle 3 Ontological concepts must be classified according as codes, code behaviours and code attributes. Principle 4 Ontological concepts should be organised as per behavioural specifications. Principle 5 Inter-conceptual relationships should be restricted by cardinality. Principle 6 A programme specific but reasonable naming regulation is necessary for the concepts. Principle 7 Aspect oriented restructuring should be implemented. Principle 8 Ontology design must be characterised by extensibility through multiple layers of the software source code. (Zhou et al, 2007, pp. 145-146) Zhou et al’s (2007) approach appear to be materialistic since most of the experts in the field of ontology research, analysis and development advocate a systemic but structure approach to be deployed the issues in the way of system migration and cross-platform integration. Examples include Devedzic (2002), Corcho et al (2003), etc. Synchronisation of Mobile technologies and Cloud Computing: Concept of Mobile Clouds Synchronisation between cloud computing and mobile technologies is a precondition to the evolution of mutually compatible ontology between the two. Caceres et al (2005) point out that personal computer technology is being used to create portable devices with enhanced computing. In other words, this is like adding portability to the personal computers (Surie et al, 2007). Contextually, Annamalai et al (2006) has advocated a comparative approach to evaluate the operability of portable desktop development and deployment. In order to improve the performance and technical adaptability of the portable desktop, Wolbach et al (2008) advocate an infrastructure based transient convertibility and modification paradigm, which will focus on the technical intricacies of both software and hardware migration. Creation of such a strong theory based transition methodology further necessitates optimisation of the migrant virtual computers, because virtual computing can be embedded in mobile devices as well provided that the device is equipped with powerful processors (Sapuntzakis et al, 2002). At this point, it should also be noted that researchers like Zhou et al (2007) and Clark et al (2005) have conducted trans-platform software migration rather successfully. In fact, Clark et al (2005) tested live migration methodology in a virtual environment with much success. Concept of Cloudlet: A Transitory Migration Paradigm One of the latest techniques (quite tricky although!) of facilitating software migration from mobile to cloud platforms and vice versa is the creation of cloudlets. Cloudlets can be regarded as compact and scalable cloud clusters that can launch virtual machine platforms culminating at parallel computing and facilitate even the most complicated software migrations (Lagar-Cavilla et al, 2009; Tolia et al, 2003). For example, Satyanarayanan et al (2009) have advocated utilisation of virtual machine based cloudlets (see Figure - 5) to sophisticate the existing mobile technologies. Satyanarayanan’s research team (2009, p. 18) states: “It is useful to contrast dynamic VM synthesis with the alternative approach of assembling a large file from hash-addressed chunks. Variants of the alternative approach have been used in (different) systems.” The most compatible systems in this regard are LBFS and ISR. LBFS is a highly adaptable low bandwidth network level file system helpful in distributive computing (Muthitachareon et al, 2004). ISR stands for Internet Suspend/Resume. Internet suspension and resumption, when done in quick succession as per the durations and requirements of Internet packet transmissions can help to establish small scale but efficient cloud based networks (Satyanarayanan et al, 2007; Kozuch et al, 2004). Figure – 5: Concept of cloudlet as explained by Satyanarayanan et al, 2009, p. 18) List of References Alam, A. and Padenga, T. 2010, Applied Software Reengineering. Noida: Pearson Annamalai, M. et al 2006, Implementing portable desktops: A new operation and comparison. In: Technical Report MSR-TR-2006-151. Redmond: Microsoft. Arizona State University 2011, Use of Ontology in Cloud Computing. Available at: http://ontologyincloud.wikispaces.asu.edu/Use+of+Ontology+in+Cloud+Computing. Last accessed on 24th January, 2013 Bauer, E. and Adams, R. 2012, Reliability and Availability of Cloud Computing. Chichester: Wiley Caceres, R. et al 2005, Reincarnating PCs with portable soul-pads. In: MobiSys ’05: The Third International Conference on Mobile Systems, Applications, and Services. New York: IEEE Chen, F. et al 2009. Service identification via ontology mapping. In: 33rd IEEE International Computer Software and Application Conference (COMPSAC09). Seattle and Washington, USA July 2009. New York: IEEE. Clark, C. et al 2005, A live migration of virtual machines. In: Proc. of the 2nd USENIX Symposium on Networked Systems Design and Implementation. Boston, USA, May 2005. New York: IEEE Corcho, O. et al 2003, "Methodogies, Tools and Language for Building Ontologies. Where is Their Meeting Point?", Data and Knowledge Engineering, 46, pp. 41-64. de Oliveira, D. et al 2010, Towards a taxonomy for cloud computing from an e-science perspective. In: N. Antonopoulos and L. Gillam (Eds.), 2010. Cloud Computing: Principles, Systems, and Applications. New York: Springer, pp. 47-62. Devedzic, V. 2002, Understanding ontological engineering, Comm. of the ACM, 45, pp. 136-144 Gens, F. 2008, Defining “Cloud Services” and “Cloud Computing.” IDCeXchange, International Data Corporation. Available at: http://blogs.idc.com. Last accessed on 24th January, 2013 IBM 2013, Mobile Computing. Available at: http://researcher.watson.ibm.com/researcher/view_pic.php?id=145. Last accessed on 24th January, 2013 Knorr, E. and Gruman, G. 2009, What Cloud Computing Really Means? InfoWorld, Inc. Available at: http://www.infoworld.com/d/cloud-computing/what-cloud-computing-really-means-031. Last accessed on 24th January, 2013 Koh, Y. et al 2007, An analysis of performance interference effects in virtual environments. In: ISPASS. New York: IEEE, pp. 200-209 Kozuch, M. and Satyanarayanan, M. 2002, Internet Suspend/Resume. In: Proceedings of the 4rth IEEE Workshop on Mobile Computing Systems and Applications. New York, USA, June 2002. New York: IEEE Lagar-Cavilla, H.A. et al 2009, SnowFlock: Rapid virtual machine cloningfor cloud computing. In: Proc. of EuroSys, Nuremberg, Germany, March 2009. New York: IEEE Lanois, P. 2011, Caught in the clouds: The web 2.0, cloud computing and privacy? Northwestern Journal of Technology and Intellectual Property, 9(2), pp. 29-49. Mills K.L. and Gomaa, H. 2002, Knowledge-based automation of a design method for concurrent systems. IEEE Transaction on Software Engineering, 28, pp. 228-255 Muthitachareon,, A. et al 2001, A low-bandwidth network file system. In: Proc. of the 18th ACM Symposium on Operating Systems Principles. Banff, Canada, October 2001. New Jersey: ACM OpenCrowd 2010. Cloud Taxonomy. Available at: http://www.eci.com/blog/images/11-29-11_views_cloud-tax-lrg.png. Last accessed on 24th January, 2013 Sapuntzakis, C. et al 2002, Optimizing the migration of virtual computers. In: Proc. Of the 5th Symposium on Operating Systems Design and Implementation. Boston, USA, December 2002. New York: IEEE Satyanarayanan, M., 1995. Fundamental challenges in mobile computing. School of Computer Science, Carnegie Mellon University. Available: http://www.cs.cmu.edu/~coda/docdir/podc95.pdf. Last accessed on 24th January, 2013 Satyanarayanan, M. et al 2009, The case of VM-based cloudlets in mobile computing, Pervasive Computing, 8, pp. 14-23 Satyanarayanan, M. et al 2007, Pervasive Personal Computing in an Internet Suspend/Resume System. IEEE Internet Computing, 11 (2) Shang, L. et al 2010, YML-PC: A reference architecture based on workflow for building scientific working clouds, In: N. Antonopoulos and L. Gillam (Eds.), 2010. Cloud Computing: Principles, Systems, and Applications. New York: Springer, pp. 145-162. Smr, P. and Novek, V. 2006, Ontology acquisition for automatic building of scientific portals. “Ontology acquisition for automatic building of scientific portals,” in SOFSEM 2006: Theory and Practice of Computer Science: 32nd Conference on Current Trends in Theory and Practice of Computer Science. Heidelberg and London: Springer Verlag, pp. 493–500. SPICE Consortium 2008, Mobile Ontology Files and Documentation. Available at: http://ontology.ist-spice.org/spice_ontologies_files.htm. Last accessed on 24th January, 2013 Surie, A., et al 2007, Rapid Trust Establishment for Pervasive Personal Computing, IEEE Pervasive Computing 6 (4) Tolia, N. et al 2006, Quantifying Interactive Experience on Thin Clients, IEEE Computer, 39 (3) Wolbach, A. et al 2008, In: Proc. of the MobiVirt 2008 Workshop on Virtualization in Mobile Computing. Breckenridge, USA, June 2008. New York: IEEE Youseff, L. et al 2008, Toward a unified ontology of cloud computing. In: GCE’08. Austin, USA, November 2008. New York: IEEE, pp. 1-10 Zhou, H. et al 2007, OPTIMA: An Ontology-based PlaTform-specIfic software Migration Approach. In: QSIC’O7. Portland, USA, October 2007. New York: IEEE, pp. 143-152 Read More