Extreme DA new-generation timing analysis tools


	Variability The GoldTime suite models physical variations in several different ways. Systematic, or deterministic, variations can be attributed to a single governing principle and are usually dependent on die position and surrounding components. Examples are metal thickness, temperature, IR-drop variation, and proximity dependent gate length variations. Modeling systematic variations entails giving GoldTime the physical variation values or a method of calculating them. Random variations, such as dopant fluctuations, are modeled statistically. GoldTime can model random variations as die-to-die, spatially correlated, or mismatch. GoldTime can handle arbitrary correlations among input parameters. Variation data can be provided to GoldTime in terms of transistor and interconnect geometry parameters and in terms of E-test data. You can learn more about variation-aware design by reading the following technical articles: "An Exploratory Study on Statistical Timing Analysis and Parametric Yield Optimization" presented by Extreme DA and UMC at the International Symposium On Quality Electronic Design '07. (PDF) "Parametric Analysis to Determine Accurate Interconnect Extraction Corners for Design Performance" presented by Extreme DA at International Symposium On Quality Electronic Design '08. (PDF) For a discussion of the data and characterization requirements for variation-aware analysis see the following EEtimes article: "Cell model creation for statistical timing analysis" "Who needs statistical timing and how to use it," an interview with Chandu Visweswariah, research staff member at IBM's T.J.Watson Research Center in Yorktown Heights, by Richard Goering of SCDSource.com.