CARROT – A Tool for Fast and Accurate Soft Error Rate Estimation

We present a soft error rate (SER) analysis methodology within a simulation and design environment that covers a broad spectrum of design problems and parameters. Our approach includes modeling of the particle hit at the transistor level, fast Monte-Carlo type simulation to obtain the latching proba...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bountas, Dimitrios, Stamoulis, Georgios I.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present a soft error rate (SER) analysis methodology within a simulation and design environment that covers a broad spectrum of design problems and parameters. Our approach includes modeling of the particle hit at the transistor level, fast Monte-Carlo type simulation to obtain the latching probability of a particle hit on all nodes of the circuit, embedded timing analysis to obtain the latching window, and fine-grained accounting of the electrical masking effects to account for both the effects of scaling and of pulse duration versus the period of the system clock to get an estimate of the maximum SER of the circuit. This approach has been implemented in CARROT and placed under a broad design environment to assess design tradeoffs with SER as a parameter.
ISSN:0302-9743
1611-3349
DOI:10.1007/11796435_34