Towards Dynamic Task Scheduling and Reconfiguration using an Aspect Oriented Approach applied on Real-time concerns of Industrial Systems Alécio Pedro Delazari Binotto*, Edison Pignaton de Freitas* , *** Carlos Eduardo Pereira**, Tony Larsson*** * Informatics Institute, Federal University of Rio Grande do Sul, Porto Alegre, Brazil (e-mail: abinotto@inf.ufrgs.br, epfreitas@inf.ufrgs.br) ** Electrical Engineering Department, Federal University of Rio Grande do Sul, Porto Alegre, Brazil (e-mail: cpereira@ece.ufrgs.br) *** School of Information Science, Computer and Electrical Engineering, Halmstad University, Halmstad, Sweden (e-mail: edison.pignaton@hh.se, tony.larsson@hh.se) Abstract: High performance computational platforms are required by industries that make use of automatic methods to manage modern machines, which are mostly controlled by high-performance specific hardware with processing capabilities. It usually works together with CPUs, forming a powerful execution platform. On an industrial production line, distinct tasks can be assigned to be processed by different machines depending on certain conditions and production parameters. However, these conditions can change at run-time influenced mainly by machine failure and maintenance, priorities changes, and possible new better task distribution. Therefore, self-adaptive computing is a potential paradigm as it can provide flexibility to explore the machine resources and improve performance on different execution scenarios of the production line. One approach is to explore scheduling and run-time task migration among machines’ hardware towards a balancing of tasks, aiming performance and production gain. This way, the monitoring of time requirements and its crosscutting behaviour play an important role for task (re)allocation decisions. This paper introduces the use of software aspect-oriented paradigms to perform machines’ monitoring and a self-rescheduling strategy of tasks to address non- functional timing constraints. As case study, tasks for a production line of aluminium ingots are designed. Keywords: distributed systems, task scheduling/reconfiguration, aspect orientation, software engineering. 1. INTRODUCTION Due to the presence of timing constraints, modern applications usually require high performance platforms to deal with distinct algorithms and massive calculations. The development of low-cost powerful and application specific hardware offer several alternatives for configuration of execution platforms and its application implementation, aiming better performance, programmability and data control. The resulting heterogeneity can be viewed as an asymmetric multi-core cluster and it is intensified with the new generation of multi-core CPUs, being a challenge to perform simple programming and efficient resource utilization. In this sense, low-cost hybrid hardware architectures are becoming attractive to compose adaptable execution platforms; and software applications must benefit from that powerfulness. This leads to the creation of new strategies to distribute applications’ workload (tasks, algorithms, or even full applications that must run at the same time) to execute in the asymmetric Processing Units (PUs) in order to better meet application requirements, such as performance and timeliness, without loosing flexibility. Dynamic and reconfigurable load-balancing computing (by means of task allocation) is a potential paradigm for those scenarios, providing flexibility, improving efficiency, and offering simplicity to program an (balanced) application on heterogeneous and multi-core architectures. Figure 1 shows such a theoretical scenario of a desktop platform composed of several devices. Fig. 1. System overview. Based on that, an important step is to create a real-time workload scheduling framework to be used by applications composed of different algorithms (sensor data processing, massive mathematical calculations, artificial intelligence, graphics, cryptography, etc.), executing on a platform under time constraints to achieve a minimal Quality of Service. In addition, it has to be predicted that during execution time, Proceedings of the 13th IFAC Symposium on Information Control Problems in Manufacturing Moscow, Russia, June 3-5, 2009 978-3-902661-43-2/09/$20.00 © 2009 IFAC 1423 10.3182/20090603-3-RU-2001.0465