By Topic

Towards a Generic Design Space Exploration Framework

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Saxena, T. ; Dept. of Electr. Eng. & Comput. Sci., Vanderbilt Univ., Nashville, TN, USA ; Karsai, G.

The set of all possible design alternatives for a system is referred to as a design-space, and design-space exploration (DSE) is the systematic exploration of the elements in a design-space. Various DSE techniques have been used for hardware/software co-design, configuration of software product lines and real-time software synthesis. Although at an abstract level DSE steps performed in these domains are similar, most of the current research is focused on domain specific frameworks which are tightly coupled with tools that evaluate point designs and use domain specific knowledge. There is a need for a generic tool that can be easily configured to model exploration problems from different domains as well on different levels of abstraction. In this paper we present Generic Design Space Exploration (GDSE) framework for domain independent DSE. This framework supports generic modeling of DSE problems from different domains using a language that allows a design-space to be encoded using domain-specific concepts and a simple constraint language that we designed. Rather than forcing the user to encode their design problem in a low-level constraint language, we advocate a higher level approach. The main contribution of this framework are: 1) it is able to support modeling of DSE in different domains 2) it supports a user-friendly constraint language that is expressive enough to specify constraints 3) the automated generation of low level constraint language code from the model alleviates the user from encoding the entire problem by hand 4) solver independence allows the user to experiment with different encodings.

Published in:

Computer and Information Technology (CIT), 2010 IEEE 10th International Conference on

Date of Conference:

June 29 2010-July 1 2010