In this paper, we propose a configuration-aware data-partitioning approach for reconfigurable computing, We show how the reconfiguration overhead impacts the data-partitioning process. Moreover, we explore the system-...
详细信息
In this paper, we propose a configuration-aware data-partitioning approach for reconfigurable computing, We show how the reconfiguration overhead impacts the data-partitioning process. Moreover, we explore the system-level power-performance tradeoffs available when implementing streaming embedded applications on fine-grained reconfigurable architectures. For a certain group of streaming applications, we show that an efficient hardware/software partitioning algorithm is required when targeting low power. However, if the application objective is performance, then we propose the use of dynamically reconfigurable architectures. We propose a design methodology that adapts the architecture and algorithms to the application requirements. The methodology hag been proven to work on a real research platform based on Xilinx devices. Finally, we have applied our methodology, and algorithms to the case study of image sharpening, which is required nowadays in digital cameras and mobile phones.
There is nowadays a wide recognition that low coupling is a main tenet in the measurement of high-quality modular software. In fact, coupling is one of the few internal software attributes that has been both theoretic...
详细信息
ISBN:
(纸本)1595935843
There is nowadays a wide recognition that low coupling is a main tenet in the measurement of high-quality modular software. In fact, coupling is one of the few internal software attributes that has been both theoretically and empirically shown to have a large impact on a variety of external software qualities, such as reusability, maintainability, evolvability and testability. Aspect-oriented programming (AOP) is an emerging technique that advocates enhanced modularization of certain widely-scoped system properties, the so-called crosscutting concerns. However, there is a poor common understanding of coupling in the context of AOP. Most of the existing metrics and assessment frameworks concentrate on the coupling evaluation of the AspectJ programming language. In addition, they took their own particular view of what coupling means in this particular language. This paper presents the definition of a generic coupling framework that takes into account both AspectJ and CaesarJ, two representatives of the most well-known families of available AOP languages. The current version of the proposed framework allows for the definition of different coupling metrics, which in turn permits the analysis and comparison of Java, AspectJ and CaesarJ implementations. We also illustrate how the framework can be applied to the characterization of existing coupling metrics. Copyright 2006 acm.
It is a great pleasure to welcome you to the 21st annual acmsymposium on appliedcomputing, Special Track on Distributed Systems and Grid computing (DSGC) in Dijon, France. The objective of the DSGC track is to provi...
ISBN:
(纸本)9781595931085
It is a great pleasure to welcome you to the 21st annual acmsymposium on appliedcomputing, Special Track on Distributed Systems and Grid computing (DSGC) in Dijon, France. The objective of the DSGC track is to provide a forum for scientists, engineers and practitioners in academia, industry and research institutes to share technical ideas, experiences and results and to present their latest findings in any aspects of Parallel, Distributed, and Grid computing. The topics of the track emphasize the design, architecture, and software of distributed systems and grid computing environments with their scientific and engineering applications.
暂无评论