Suddhasil De

Modeling an Enhanced Tuple Space Based Mobile Middleware in UNITY

Tuple Space based Mobile Middleware (TSMM) is developed to tackle emerging dynamics in underlying infrastructure. Its coordination medium, viz. tuple space model, has multiple inherent decoupled behaviors to coordinate interactions between different agents of supported applications. This paper focuses on one such decoupling quality, synchronization decoupling, by which agents are decoupled from their invoked primitives. This decoupling ability is accomplished by including reactivity in tuple space model. Reactive tuple space model also enables development of robust and flexible applications for TSMM. However, in existing TSMM, this decoupling behavior has not been fully attained, as reactions are themselves coupled with invoked primitives, leading to additional network and processing overheads. This paper proposes an improved reactive tuple space model for TSMM with enhanced decoupling behavior of reactions to achieve robustness for supported applications. Proposed tuple space model is also formally modeled and verified by notation and proof logic of UNITY model.

On performance improvement issues in unordered Tuple Space based Mobile Middleware

Tuple Space based Mobile Middleware (TSMM), a new genre of mobile middleware, is developed to tackle emerging dynamics in underlying infrastructure. It uses tuple space model to coordinate interactions between different active components (agents) of supported applications. This paper focusses on a primary design issue of tuple space model, viz. tuple-antituple structure, which specifies arities and nature of arrangements of constituent fields of tuples and antituples. This factor not only affects application design, but also impacts simplicity, flexibility, scalability and performance of TSMM. Broadly, two types of arrangements are possible: ordered (where arity and arrangement of fields are predefined), and unordered (where none of them are predefined). Ordered structure lacks flexibility and restricts the design of TSMM and its applications. Unordered structure removes these drawbacks, but degrades TSMMs performance and scalability, as additional creation and lookup overheads are introduced here. Among the existing TSMM, LIMONE incorporates unordered tuple-antituple structure. In this paper, we modify LIMONEs tuple-antituple structure for improving its performance and scalability. Both original and modified models are analyzed and experimented to show the improvements.

Tuple space enhancements for mobile middleware

Tuple space-based mobile middleware TSMM is affected by two elementary factors of the tuple space model, viz. the tuple/antituple structure and the tuple space structure. The tuple/antituple structure influences application design as well as simplicity and flexibility of TSMM, while the tuple space structure regulates its performance and scalability. In existing TSMM, the unordered tuple/antituple structure improves simplicity and flexibility, while degrades performance and scalability due to the additional overhead in tuple generation and lookup. Again, the sequential tuple space structure in existing TSMM produces considerable performance overhead for the tuple-reading and -consuming primitives. This paper proposes two enhancements in the tuple space model for TSMM: 1 enhancement in the unordered tuple/antituple structure for reducing the additional overhead as well as preserving the simplicity and flexibility of TSMM; and, 2 introduction of the enhanced tuple space structure for further reducing the overhead of the tuple-reading and -consuming primitives. Analysis and experiments confirm the superiority of TSMM with proposed enhancements over existing ones.

Uncoupling in services of mobile cloud computing using tuple space model: design and formal specifications

Mobile Cloud Computing widens the service of existing facility of cloud computing paradigm to user applications executing in mobile/portable devices, called mobile applications. It enables existing cloud services to serve these applications, thereby helps to overcome the inherent limitations of mobile

A new tuple space structure for tuple space based mobile middleware platforms

/

On the Notion of Decoupling in Mobile Cloud Computing

portable devices. However, the use of cloud services in existing architecture require tight coupling between mobile applications and cloud services, which is highly undesirable in dynamic and unreliable mobile cloud computing. A new architecture of mobile cloud computing, where interactions between mobile applications and cloud services will be decoupled, is proposed in this paper. Tuple space model is used for uncoupling these interactions in proposed architecture, improving flexibility and efficiency of mobile cloud computing to a great extent. This paper also suggests an approach of formalizing proposed architecture to depict the uncoupling of service access and delivery to mobile applications. Formalization is carried out using Mobile UNITY.

Formalization of a Fully-Decoupled Reactive Tuple Space Model for Mobile Middleware

Tuple space model is a simple and efficient data-driven coordination model, suitable for different mobile middleware platforms. A significant performance factor of this model is tuple space structure. Existing tuple space models have sequential tuple space structure, which works efficiently with tuple-producing primitives. But, tuple-reading and -consuming primitives that drive overall performance of tuple space model, generate substantial overhead with this sequential structure. In this paper, we propose an efficient tuple space structure, which is not only simple, but also improves performance and scalability of tuple space model. We have compared proposed structure with existing ones to show the improvements. Tuple space model with proposed tuple space structure not only preserves the ease of developing mobile applications, but also facilitates performance gain during their execution.

Architectures of mobile middleware: A taxonomic perspective

Mobile cloud computing extends the existing facility of cloud computing paradigm to different mobile applications executing in mobile/portable devices of end users. It caters these applications with the support of existing cloud services, thereby helping them to overcome the inherent limitations of mobile/portable devices. However, existing cloud services in mobile cloud architecture get tightly coupled with these applications, which is highly undesirable in dynamic and unreliable mobile cloud computing. A new mobile cloud architecture, where interactions between mobile applications and existing cloud services are decoupled in nature, is proposed in this paper. Tuple space model is used for decoupling these interactions in proposed architecture, which improves flexibility and efficiency of mobile cloud computing. This paper also suggests an approach of formalizing proposed architecture in order to appropriately analyze its flexibility in decoupling mobile applications from their leased cloud services. Formalization is carried out using Mobile UNITY.

Modeling decoupled mobile cloud computing using Mobile UNITY

This paper suggests an approach for formalizing Tuple Space based Mobile Middleware (TSMM) that contains a fully-decoupled reactive tuple space model as coordination medium. Formalization of TSMM is carried out using Mobile UNITY.

Formalization of discovery and communication mechanisms of Tuple Space based Mobile Middleware for underlying unreliable infrastructure

Complex decentralized applications, developed for emerging mobile computing paradigm, need support of mobile middleware platforms to enable coordination of active components of these applications. Recent works demonstrated that tuple space model can act as coordination medium for mobile middleware platforms, which are termed as tuple space based mobile middleware. Importance of these middleware are escalating day-by-day with changing network scenario, thereby demanding a better understanding of their current and future research directions. This motivating factor drives us to present a taxonomy of architectures for tuple space based mobile middleware. Proposed taxonomy is the new of its kind and is based on distribution pattern of tuple spaces within middleware and nature of identification of these tuple spaces. Multiple existing tuple space based mobile middleware are also considered and classified according to proposed taxonomy.