Abstract-Reconfigurable FPGA based Systems-on-Chip (SoC) architectures are increasingly becoming the preferred solution for implementing modern embedded systems, due to their flexible nature. However due to the tremendous amount of hardware resources available in these systems, new design methodologies and tools are required to reduce their design complexity. In this paper we present an exploratory analysis for specification of these systems, while utilizing the UML MARTE (Modeling and Analysis of Real-time and Embedded Systems) profile. Our contributions permit us to model fine grain reconfigurable FPGA based SoC architectures while extending the profile to integrate new features such as Partial Dynamic Reconfiguration supported by these modern systems. Finally we present the current limitations of the MARTE profile and ask some open questions regarding how these high level models can be effectively used as input for commercial FPGA simulation and synthesis tools. Solutions to these questions can help in creating a design flow from high level models to synthesis, placement and execution of these reconfigurable SoCs.
Dynamic Partial Reconfiguration (DPR) has been introduced in recent years as a method to increase the flexibility of FPGA designs. However, using DPR for building complex systems remains a daunting task. Recently, approaches based on Model-Driven Engineering (MDE) and UML MARTE standard have emerged which aim to simplify the design of complex SoCs, and in some cases, DPR systems. Nevertheless, many of these approaches lacked a standard intermediate representation to pass from high-levels of descriptions to executable models. However, with the recent standardization of the IP-XACT specification, there is an increasing interest to use it in MDE methodologies to ease system integration and to enable design flow automation. In this paper we propose an MARTE/MDE approach which exploits the capabilities of IP-XACT to model and automatically generate DPR SoC designs. We present the MARTE modeling concepts and how these models are mapped to IP-XACT objects; the emphasis is given to the generation of IP cores that can be used in the Xilinx EDK (Embedded Design Kit) environment, since we aim to develop a complete flow around their Dynamic Partial Reconfiguration design flow. Finally, we present a case study integrating the presented concepts, showing the benefits in design efforts compared with a purely VHDL approach and using solely EDK. Experimental results show a reduction of
Controlling dynamic and partial reconfigurations becomes one of the most important key issues in modern embedded systems design. In fact, in such systems, the reconfiguration controller can significantly affect the system performances. Indeed, the controller has to handle efficiently three major tasks during runtime: observation (monitoring), taking reconfiguration decisions and notify decisions to the rest of the system in order to realize it. We present in this paper a novel high level approach permitting to model, using MARTE UML profile, modular and flexible distributed controllers for dynamic reconfiguration management. This approach permits components/ models reuse and allows systematic code generation. It consequently makes reconfigurable systems design less tedious and reduces time to market.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.