The Embedded Systems Design Challenge

[attachment=17732]

1 Motivation
Computer Science is going through a maturing period. There is a perception
that many of the original, defining problems of Computer Science either have
been solved, or require an unforeseeable breakthrough (such as the P versus NP
question). It is a reflection of this view that many of the currently advocated
challenges for Computer Science research push existing technology to the limits
(e.g., the semantic web [4]; the verifying compiler [15]; sensor networks [6]), to
new application areas (such as biology [12]), or to a combination of both (e.g.,
nanotechnologies; quantum computing). Not surprisingly, many of the brightest
students no longer aim to become computer scientists, but choose to enter
directly into the life sciences or nanoengineering [8].


2 Current Scientific Foundations for Systems Design,
and their Limitations
2.1 The Embedded Systems Design Problem
What is an embedded system? An embedded system is an engineering artifact
involving computation that is subject to physical constraints. The physical
constraints arise through two kinds of interactions of computational processes
with the physical world: (1) reaction to a physical environment, and (2) execution
on a physical platform. Accordingly, the two types of physical constraints
are reaction constraints and execution constraints. Common reaction constraints
specify deadlines, throughput, and jitter; they originate from the behavioral requirements
of the system. Common execution constraints put bounds on available
processor speeds, power, and hardware failure rates;



3 Current Engineering Practices for Embedded Systems
Design, and their Limitations
3.1 Model-based Design
Language-based and synthesis-based origins. Historically, many methodologies
for embedded systems design trace their origins to one of two sources:
there are language-based methods that lie in the software tradition, and synthesisbased
methods that come out of the hardware tradition. A language-based approach
is centered on a particular programming language with a particular target
run-time system. Examples include Ada and, more recently, RT-Java [5]. For
these languages, there are compilation technologies that lead to event-driven
implementations on standardized platforms (fixed-priority scheduling with preemption).
The synthesis-based approaches, on the other hand, have evolved
from hardware design methodologies. They start from a system description in
a tractable (often structural) fragment of a hardware description language such
as VHDL and Verilog and, ideally automatically, derive an implementation that
obeys a given set of constraints.