First ACM and IEEE International Conference on Formal Methods and Models for Co-Design (MEMOCODE?03)
PANEL: Should the space of implementation possibilities be determined by the abilities of high-level synthesis and validation?
Mont Saint-Michel, France
June 24-June 26
ISBN: 0-7695-1923-7
Abstraction vs. implementation raises a lot of questions in the minds of designers. Researchers in the EDA field claim that abstraction helps validation, conceptualization, and must be instrumental to coping with the rising complexity of design. Design engineers often argue that abstraction means distance from real implementation details, and hence optimality is compromised. One may argue that some compromise on optimality may be necessary to cope with the mythical productivity gap, but experienced designers may not agree to do that, especially in the field of performance driven design, such as general purpose microprocessors. In this panel, we ask industry and academic experts about their views on the compromise one should be ready to make in order for higher abstraction driven design methodology to materialize.
Citation:
Rajesh Gupta, Sandeep Shukla, "PANEL: Should the space of implementation possibilities be determined by the abilities of high-level synthesis and validation?," memocode, pp.277, First ACM and IEEE International Conference on Formal Methods and Models for Co-Design (MEMOCODE?03), 2003