Multi-terminal nets do change conventional wire length distribution models

Authors:
Dirk Stoobandt
Affiliations:
Ghent Univ.,Gent, Beligum
Venue:
Proceedings of the 2001 international workshop on System-level interconnect prediction
Year:
2001

Citing 0
Cited 7

A priori system-level interconnect prediction: Rent's rule and wire length distribution models

Proceedings of the 2001 international workshop on System-level interconnect prediction
On rent's rule for rectangular regions

Proceedings of the 2001 international workshop on System-level interconnect prediction
Prediction of interconnect net-degree distribution based on Rent's rule

Proceedings of the 2004 international workshop on System level interconnect prediction
Empirical models for net-length probability distribution and applications

IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Post-placement interconnect entropy: how many configuration bits does a programmable logic device need?

Proceedings of the 2006 international workshop on System-level interconnect prediction
Microminiature packaging and integrated circuitry: the work of E. F. Rent, with an application to on-chip interconnection requirements

IBM Journal of Research and Development - POWER5 and packaging
Post-placement interconnect entropy

IEEE Transactions on Very Large Scale Integration (VLSI) Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Conventional models for estimating wire lengths in computer chips use Rent's rule to estimate the number of terminals between sets of gates. The number of interconnections then follows by taking into account that most nets are point-to-point connections. In this paper, we introduce a model for multi-terminal nets and we show that such nets have a fundamentally different influence on the wire length estimations than point-to-point nets. The multi-terminal net model is then used to estimate the wire length distribution in two cases: (i)m the distribution of source-sink pairs for applications of delay estimation and (ii) the distribution of Steiner tree lengths for applications related to routing resource estimation. The effects of including multi-terminal nets in the estimations are highlighted. Experiments show that the new estimated wire length distributions are close to the measured ones.