Statistical Timing Analysis with Extended Pseudo-Canonical Timing Model

State of the art statistical timing analysis (STA) tools often yield less accurate results when timing variables become correlated due to global source of variations and path reconvergence. To the best of our knowledge, no good solution is available dealing both types of correlations simultaneously....

Full description

Saved in:
Bibliographic Details
Main Authors: Zhang, Lizheng, Chen, Weijen, Hu, Yuhen, Chen, Charlie Chung-Ping
Format: Conference Proceeding
Language:English
Subjects:
Benchmark testing
Circuit testing
Computational efficiency
Computing methodologies > Modeling and simulation > Model development and analysis > Modeling methodologies
Computing methodologies > Symbolic and algebraic manipulation > Symbolic and algebraic algorithms
Delay
Estimation error
General and reference > Cross-computing tools and techniques > Performance
Hardware > Emerging technologies
Hardware > Very large scale integration design
History
Mathematics of computing > Discrete mathematics > Graph theory > Paths and connectivity problems
Mathematics of computing > Probability and statistics > Distribution functions
Random variables
Runtime
Theory of computation > Design and analysis of algorithms
Timing
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:State of the art statistical timing analysis (STA) tools often yield less accurate results when timing variables become correlated due to global source of variations and path reconvergence. To the best of our knowledge, no good solution is available dealing both types of correlations simultaneously. In this paper, we present a novel extended pseudo-canonical timing model to retain and evaluate both type of correlation during statistical timing analysis with minimum computation cost. Also, an intelligent pruning method is introduced to enable trade-off runtime with accuracy. Tested with ISCAS benchmark suites, our method shows both high accuracy and high performance. For example, on the circuit c6288, our distribution estimation error shows 15× accuracy improvement compared with previous approaches.
ISSN:1530-1591
1558-1101
DOI:10.1109/DATE.2005.280