Abstract
The expansion of human knowledge in all areas is largely the outcome of the activity of academic institutions and the result of their mission to contribute to the cultural, intellectual and economic development of the society, involving education, research and university extension activities. For many years, the academic community has been organizing itself in all different ways to respond to current and future needs, ensuring research integrity and recognition, and building on successive generations of peers to validate and support the launching and development of novel research streams. We owe the current state of research and development of our society to generations of scholars and scientists that have brought all of us here.(...)

Abstract

Abstract
Self-Regulation and retail problem solving using Multi-Agent Systems are considered two promising areas but relatively little explored. The regulation in these environments must be able to handle with the implicit dynamism and variation in a complex area such as retail systems with a minimum human interference. Agents must be able to change their behaviour based on rules previously formatted, using an autonomic process that does the maintenance of a knowledge base which is well defined and consistent, to be possible to meet all the objectives and to provide support on the need of change. © 2013, Springer Science+Business Media Dordrecht.

Abstract
Current mobile applications treat the end-user device as a 'thin client,' with all of the heavy computations being offloaded to an infrastructure cloud. However, the computational capabilities of mobile devices are constantly improving, and it is worthwhile considering whether an edge-cloud that consists purely of mobile devices (operating effectively as 'thick clients') can perform as well as, or even better than, an infrastructure cloud. In this paper, we study the trade-offs between offloading computation to an infrastructure cloud versus retaining the computation within a mobile edge-cloud. To this end, we develop and run two classes of applications on both types of clouds, and we analyze the performance of the two clouds in terms of the time taken to run the application, along with the total amount of battery power consumed in both cases. Our results indicate that there are indeed classes of applications where an edge-cloud can outperform an infrastructure cloud in terms of both latency and battery power. © 2013 IEEE.

Abstract
This paper presents an extension to a hardware/software system architecture in which repetitive instruction traces, called Megablocks, are accelerated by a Reconfigurable Processing Unit (RPU). This scheme is supported by a custom toolchain able to automatically generate a RPU tailored for the execution of one or more Megablocks detected offline. Switching between hardware and software execution is done transparently, without modifications to source code or executable binaries. Our approach has been evaluated using an architecture with a MicroBlaze General Purpose Processor (GPP) softcore. By using a memory sharing mechanism, the RPU can access the GPP's data memory, allowing the acceleration of Megablocks with load/store operations. For a set of 21 embedded benchmarks, an average speedup of 1.43x is achieved, and a potential speedup of 2.09x is predicted for an implementation using a low overhead interface for communication between GPP and RPU.

Abstract