Search

With high integration density due to scaling of IC fabrication processes, faster design cycles, are the need of today along with development of energy efficient architectures fulfilling the speed requirements. The focus here is the to build up features of stochastic computing for data centric or data steaming applications. High level synthesis (HLS) allows the designer to incorporate transformations on the applications before the register transfer level (RTL) generation to incorporate design space explorations. Herein the design space explorations start with retiming. Also the novelty is that filter bench marks considered are implemented by incorporating stochastic computing functions. The same are used to develop the filter benchmarks and then retiming transformation is applied in HLS. The design space trade off results with respect to retiming transformations on conventional computation and stochastic computing architectures are presented, for area and power and it is found that stochastic computing results in area saving and power saving scales relatively for higher order applications namely the filter benchmarks showing that stochastic structures to be energy efficient than conventional computing. The future scope of the work is to consider accuracy analysis based on end user demand of accuracy versus energy efficiency.

Agricultural geography is one of the subportion of human and our economic geography which examines the primary, secondary, tertiary and quaternary yield types activities that are carried out in agriculture. In this paper, we are examining the spatial distribution and concentration of crops and their yields along with their crop periods using Big data Analytics to find out the cropping patterns and combinations that varies in space and time. Our primary objective is to findout the crop associations and patterns under climatic influence for each geographical segmentation, to give better yeilds. The Big data analytic based association won’t to last as many of the farmers and scientists are rightly challenging over the agricultural sustainability. However, there is a strong possibility of the farmers to adopt a new combination in the coming decades as the Big data analytic based crop pattern decision facilitates the farmers, always try to optimize their agricultural re-turns and adopt new innovations which gives better yeilds since we are performing factual data centric analytics.

We have been developing BriCA (Brain-inspired Computing Architecture), the generic software platform that can combine an arbitrary number of machine learning modules to construct higher structures such as cognitive architectures inspired by the brain. We discuss requirements analysis and design principles of this cognitive computing platform, report its implementation, and describe plans for further development.

Nowadays, knowledge is the key to success in all software engineering processes. This valuable knowledge, obtained through analysis, design, development, and maintenance processes of the system, should be saved and reused in designing and developing current and similar systems. Using pre-existing knowledge is a practical approach which reduces design complexity, improves software architecture design and manages software quality. In this paper, we describe an approach to create architecture design knowledge using a hierarchical structure of architectural styles based on quality attributes. Knowledge is most importance asset of our approach that reuses similar domains, correlates architectures and reduces complexity by using and composing rules. Finally, we executed our approach on a standard case study.

Supercomputer architectures are being upgraded using different level of parallelism to improve computing performance. This makes it difficult for scientists to develop high performance code in a short time. From the viewpoint of productivity and software life cycle, a concise yet effective infrastructure is required to achieve parallel processing. In this paper, we propose a usable building block framework to build parallel applications on large-scale Cartesian data structures. The proposed framework is designed such that each process in a simulation cycle can easily access the generated data files with usable functions. This framework enables us to describe parallel applications with fewer lines of source code, and hence, it contributes to the productivity of the software. Further, this framework was considered for improving performance, and it was confirmed that the developed flow simulator based on this framework demonstrated considerably excellent weak scaling performance on the K computer.

Improved techniques to develop future System of Systems solution architectures are a systems engineering imperative. The systems engineer must be able to produce quantitative Measures of Performance that parse good performing architectures from bad architectures in a very complex and dynamic domain often and early during concept development. This paper proposes innovative System of Systems Engineering techniques to manage risks and reduce costs in complex systems. We suggest a combination of graph theory, Big Data and Uncertainty Quantification are foundational tools that can model architectures and more importantly the movement of data across architecture interfaces. This provides a mathematical foundation for quantitative architecture, model based systems engineering and simulation environments enabling continuous prototype testing in a model environment. The result is a means to quantitatively validate viable complex architectures early in the lifecycle. This paper describes a graph-based approach to model complex architectures. Complexity will be defined mathematically as a function of the number of entities, inter-entity relationships and active event sequences in the graphical architecture. Uncertainty quantification compares the graph model to the physical System of Systems based on quantities of interest. Big Data facilitates analytics across continuous data threads in the architecture.

The evolution of System of Systems (SoS) architectures is a difficult process due to the complex interactions between entities that exist in an SoS. Adequate tools are thus needed in capturing these interdependencies to support informed decisions on architectural evolutions. In this paper, an approach is introduced to facilitate decision making on architecture evolution strategies. The main purpose of evolving architecture is to satisfy advanced capability and performance objectives through adding new systems, replacing existing systems, changing links, etc. The paper employs Colored Petri Nets (CPN), a powerful discrete event dynamic simulation tool, to model, simulate and evaluate the existing and evolving architectures. However, it is unrealistic to improve performance without considering the cost of architecture evolutions. This paper regards complexity as an indicator of the architecture evolving cost. Compared to other complexity metrics, dynamic complexity is included in the complexity metric in this paper. Eventually, an appropriate evolution choice could be achieved by examining the tradeoff space between complexity and performance. This approach is illustrated with a conceptual SoS problem.

This paper describes an empirical study of decision-making when changing the architecture in embedded systems. A refactoring of the system architecture often gives effects on both system properties and functions in the company organization, and there is a lack of efficient analysis methods for decision support in the system architecture process. This study investigates the information needed to make a decision about a system refactoring. Scenario-based interviews have been conducted with managers and system architects from companies developing embedded systems. The results show that the companies investigate similar issues regardless of their industry sector. The most wanted information prior to a decision is also presented.

Objective of the proposed architecture is to enable representing an electronic document as a multi-layered structure of executable digital objects, which is extensible and without a need to support any particular formats or user interfaces. IODA layers are intended to reflect document content organization levels rather then system abstraction or functional levels, as in software architecture models.