A Mechanism for Communicating in Dynamically Reconfigurable Embedded Systems

نویسندگان

  • Mehrdad Hassani
  • David B. Stewart
چکیده

We present a time-bounded state-based communication mechanism for dynamically reconfigurable embedded systems. The mechanism is a single-processor, low-overhead version of the Chimera state-variable mechanism, that was developed for state-based communication in multi-processor environments. The new design is suitable for execution on low-performance embedded processors, uses less memory, and supports dynamic binding, one-to-one, one-to-many, and broadcast capabilities in a time-deterministic manner. The Chimera Project[4] demonstrated the use of state-based communication to create dynamically reconfigurable real-time software objects for a multiprocessor environment. Its state variable (SVAR) mechanism provides the user with operating system services that support transparent time-bounded inter-processor communication for high assurance control systems. The mechanism is fully deterministic, and has guaranteed worst-case waiting and transfer times for shared data. In conjunction with Chimera's port-based object model of real-time software components, it improves predictability and performance by minimizing inter-object dependencies as compared to message-based systems. The mechanism also supports dynamic binding, one-to-one, one-to-many, and broadcast capabilities. The SVAR model of communication is also suitable for creating dynamically reconfigurable software for embedded systems[6]. The characteristics of embedded systems introduce different challenges into the design of an SVAR mechanism , as compared to Chimera's version. In particular, embedded systems usually use lower performance processors with less flexibility, have significant memory and CPU band-width limitations, and timing constraints are often more rigid than those allowed in a multiprocessor environment. The binding of communicating objects must also be fast and time-bounded to support dynamic reconfigurability. Although the Chimera mechanism is very effective for meeting its goals in a multiprocessor environment, it has several shortcomings when applied to embedded systems. Its usage of memory is inefficient, as the local table is a duplicate of the global table, even though a process accesses only a small subset of it. It has larger overhead for locking the table, due to its multiprocessor support. The analysis of the mechanism is specific to a VMEbus hardware environment. Chi-mera's mechanism is also specific to its port-based object model, and only supports two pre-programmed operations, one for reading, one for writing. By redesigning the mechanism, we have made improvements that address all of the above shortcoming, making the SVAR communication practical for embedded systems. The mechanism has also been decoupled from the internals of the Chimera Real-Time Operating System[5], allowing it to be ported to any environment. State-based communication is preferred in embedded systems to provide higher assurability. In a message-based system …

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A review of high-level synthesis for dynamically reconfigurable FPGAs

Dynamically Reconfigurable Field Programmable Gate Arrays (DR FPGAs) change many of the basic assumptions of what hardware is. DR FPGA-based dynamically reconfigurable computing has become a powerful methodology for achieving high performance while minimizing the resource required in the implementation of many applications. The key to harnessing the power of DR FPGAs for most applications is to...

متن کامل

An O(1) Distributed Exception Handling Mechanism for Dynamcially Reconfigurable Real-time Software

Title of Thesis: AN O(1) DISTRISBUTED EXCEPTION HANDLING MECHANISM FOR DYNAMICALLY RECONFIGURABLE REAL-TIME SOFTWARE Degree Candidate: Jun Lang Degree and Year: Master of Science, Electrical Engineering, 1997 Thesis directed by: David B. Stewart, Assistant Professor Department of Electrical Engineering With programs growing in size and complexity, the quality and cost of developing and maintain...

متن کامل

Model-based platform-specific co-design methodology for dynamically partially reconfigurable systems with hardware virtualization and preemption

To facilitate the development of the dynamically partially reconfigurable system (DPRS), we propose a model-based platform-specific co-design (MPC) methodology for DPRS with hardware virtualization and preemption. For DPRS analysis and validation, a model-based verification and estimation framework is proposed to make model-driven architecture (MDA) more realistic and applicable to the DPRS des...

متن کامل

Dynamically Co-synthesis of H/w & S/w and Optimization in Reconfigurable Embedded System

Field Programming Gate Array (FPGA) play an important role in reconfigurable computing. Reconfigurable computing has been used to build reconfigurable embedded system. Reconfigurable embedded System is dynamically changing in the hardware circuit at runtime with the reconfigurable characteristic of Programmable Logic Devices like FPGA, to give the system advantages in both hardware and software...

متن کامل

Configuration Management for Networked Reconfigurable Embedded Devices

Distribution of product updates to embedded devices can increase product lifetimes for consumers and boost revenues for vendors. Dynamic provisioning of application solutions to embedded devices are complex due to their heterogeneous nature. This paper proposes a configuration management architectural framework which incorporates a mobile agent based push methodology for networked reconfigurabl...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1997