John J. Granacki
USC/Information Sciences Institute
4676 Admiralty Way Suite 1000
Marina Del Rey, CA 90292, U.S.A.
Voice: (310) 822-1511 x770
Fax: (310) 823-6714
Pager: (888) 424-2651
email: granacki@isi.edu
Education
Ph.D., Electrical Engineering, University of Southern California, 1986
Engineer, Electrical Engineering, University of Southern California, 1982
M.S., Physics, Drexel University, 1970
B.A., Physics, Rutgers University, 1968
Research Interests
John's current professional interests include: data intensive systems;
embedded systems; wireless computing systems; distributed computing
systems; tools for system-level specification and models of system
behavior; tools for designing and analyzing adaptable/reconfigurable
computing systems.
Recent Research
Problem Statement and Objective
As technology allows higher and higher packaging densities at the
chip, module and board levels, the task of designing and implementing
computer systems is becoming increasing complex.
Furthermore, the rapidly evolving fabrication technologies requires
system designers to continually monitor changes in manufacturing
details and to incorporate these details into CAD tools and designs.
Our objective is to develop tools that automatically incorporate the
details of the fabrication technologies and allow the designer to
focus on the system-design problem. These tools allow designer to
make high-level trade-offs and analyze multiple design iterations
quickly facilitating the rapid prototyping of complex designs.
Method of Solution
We are developing improved models of the design process based on
techniques borrowed from data-flow analysis, Petri Nets and
programming languages. To test these models, we have developed and
implemented a Library Browser, a Design Flow Manager and an Automatic
Script Generation Tool that allow novice designers to produce
realizable physical implementations of systems.
We are also working on enhanced estimators for the physical design of
systems that can be used in high-level synthesis and can dramatically
shorten the time it takes to perform a design iteration.
Expected Significance
Our tools have already been used by several teams of academic
researchers to successfully implement their experimental computer
systems. Using these tools reduced the cost and time required to
prototype these systems and allowed researchers to focus on system
instead of implementation issues.
Selected Publications
1. Ivan Hom and John Granacki, "Estimation of the Number of Routing
Layers and Total Wirelength in a PCB through Wiring Distribution
Analysis," in Proceedings of the European Design Automation
Conference, pp. 310-315, September, 1996.
2. Sally A. Hayati, Alice C.Parker and John J. Granacki,
Representation of Control and Timing Behavior with Applications to
Interface Synthesis, In Proceedings of the IEEE International
Conference on Computer Design, October 1988.
3. John J. Granacki, Alice C. Parker and Yigal Arens, Understanding
System Specifications Written in Natural Language, In Proceedings of
the Tenth International Joint Conference on Artificial Intelligence,
August 1987.