Motivations Being the fastest growing segment of the microelectronics market,
FPGAs have become mainstream already years ago in all kinds of embedded
systems. More recently FPGAs and other Reconfigurable Computing (RC)
platforms are rapidly moving into practically every application area,
such as automotive, aerospace, defense, medical, chemistry, molecular
biology, physics, astrophysics, high performance computing,
supercomputing, and many other areas. Productivity Crisis.
Rapidly growing complexity and pervasiveness
of RC-based
multi-paradigm
devices leads
to a productivity crisis of major proportions. On the other hand RC is
an efficient approach to cope with
the accelerating VLSI design crisis. While the economic importance of
RC and its FPGAs is widely acknowledged, but the strategic dimension of
RC has not been
appreciated until recently, academia has failed
to
pay sufficient attention to the
education of a community of high-quality system designers and
configware
programmers using such platforms. This has motivated a
recent but ever growing interest in the question of educating
specialists in this domain and this has also been recognized as a
particularly difficult problem. The von Neumann Syndrome.
The manycore programming crisis requires a re-definition of CS
fundamentals. Typical CS graduates are not qualified for the current
and future IT job market. In CS education we cannot avoid the
transistion from the dictatorship of the von Neumann paradigm, over to a
twin
paradigm approach interlocking the instruction-stream-based
programmer mind set with the data-stream-based mind set of
reconfigurable computing. Need for unifying the
foundations. We need to counter
the current trend, where specialization is the target of education
systems. We need to go toward interdisciplinary CS-related
curricula for unifying the foundations of the discipline since it has
become evident that
fundamental problems are shared across several different application
domains. It is
the goal of this first workshop to bring together researchers,
educators, and
industrial representatives to share design, research, and
education experiences in Reconfigurable Computing and a wide variety of
its applications. RC-based
design insolves not only
hardware-software co-design. What is really needed is a much more
interdisciplinary approach of
hardware-configware-software co-design, not only as a design practice,
but also as part of CS, DE, and EE curricula. New educational approaches
needed. Although configware
engineering
is a discipline of its own, fundamentally different from software
engineering, and, a configware industry is already existing
and growing, it is too often ignored by our curricula. Modern
FPGAs as COTS (commodities off the shelf) have all 3
paradigms on board of the same VLSI chip: hardwired accelerators,
microprocessors (and memory banks), and FPGAs, and we need software and
configware to program the same chip. To cope with the clash of
cultures we need interdisciplinary
curricula
merging all these different backgrounds in a systematic
way.
We need innovative lectures and lab courses supporting the integration
of reconfigurable computing into progressive curricula. The workshop
intends to provide a forum for presenting experiences and new
educational
approaches and for discussing the pros and cons.
The 3rd
International
Workshop on
Reconfigurable
Computing Education
RC education 2008
____________________________________________________________________
Fragmentation. Each of
these application domains
has only a limited view of
computing and takes it more as a mere technique than as a science on
its own. Consequences are, that it makes it very difficult to
bridge the cultural and practical gaps. Given this fragmentation, it
can be rather hard to investigate, since there are so many different
actors and departments involved. Including and programming
reconfigurable platforms in the design of
embedded systems
as well as embedded real-time systems and all other application areas
requires more skills at least from computer sciences.
Currently it requires to
involve experts from different backgrounds, with dissenting points of
view, not only for test and verification of such designs, if at all
possible, being very expensive and delaying significantly the
introduction of products.