2019Unpublished venueRequires access

Exploring area and total wirelength using a cell merging technique

Kevin A. Caceres Albinagorta, Calebe Conceição, Carlos Silva-Cárdenas, Ricardo Reis

Open publisher page 7 citations

Abstract

The industry of Integrated Circuits (ICs) has been making increasingly complex chips with up to billions of transistors in a single die. As we cannot do the design flow by hand, the leading adopted solution to deal with this challenge has been to use a pre-designed library of standard cells and using EDA tools to automate the process. Nevertheless, the resulting netlist is not as efficient in terms of the number of transistors as a handmade design, possibly reflecting in the overall area, power, and delay of the circuit. To generate cells on-demand is a way to improve this inherent limitation, as previous works demonstrate. In this paper, we investigate a netlist optimization methodology based on gate merging and its impacts regarding area and wire-length when applied to the Nagate's Open Cell Library for 45nm. We obtained a reduction in area and total wire-length of 3.5% and 4.2% on average, respectively.

About this research paper

What this paper is about

The industry of Integrated Circuits (ICs) has been making increasingly complex chips with up to billions of transistors in a single die. As we cannot do the design flow by hand, the leading adopted solution to deal with this challenge has been to use a pre-designed library of standard cells and using EDA tools to automate the process. Nevertheless, the resulting netlist is not as efficient in terms of the number of transistors as a handmade design, possibly reflecting in the overall area, power, and delay of the circuit. To generate cells on-demand is a way to improve this inherent limitation, as previous works demonstrate. In this paper, we investigate a netlist optimization methodology based on gate merging and its impacts regarding area and wire-length when applied to the Nagate's Open Cell Library for 45nm. We obtained a reduction in area and total wire-length of 3.5% and 4.2% on average, respectively.

Why it matters

OpenAlex reports 7 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The industry of Integrated Circuits (ICs) has been making increasingly complex chips with up to billions of transistors in a single die. As we cannot do the design flow by hand, the leading adopted solution to deal with this challenge has been to use a pre-designed library of standard cells and using EDA tools to automate the process. Nevertheless, the resulting netlist is not as efficient in terms of the number of transistors as a handmade design, possibly reflecting in the overall area, power, and delay of the circuit. To generate cells on-demand is a way to improve this inherent limitation, as previous works demonstrate. In this paper, we investigate a netlist optimization methodology based on gate merging and its impacts regarding area and wire-length when applied to the Nagate's Open Cell Library for 45nm. We obtained a reduction in area and total wire-length of 3.5% and 4.2% on average, respectively.

Key concepts: Netlist, Standard cell, Design flow, Computer science, Integrated circuit layout, Transistor, Reduction (mathematics), Process (computing)

Related papers

Back to paper searchBrowse research topicsOriginal source
Exploring area and total wirelength using a cell merging technique — Research Paper | ScholarLens